Commit
•
4481ad3
1
Parent(s):
cd1ec8a
initial commit
Browse files- .gitattributes +1 -0
- README.md +105 -3
- config.json +248 -0
- configuration_ovis.py +359 -0
- generation_config.json +11 -0
- model-00001-of-00005.safetensors +3 -0
- model-00002-of-00005.safetensors +3 -0
- model-00003-of-00005.safetensors +3 -0
- model-00004-of-00005.safetensors +3 -0
- model-00005-of-00005.safetensors +3 -0
- model.safetensors.index.json +924 -0
- modeling_ovis.py +691 -0
- preprocessor_config.json +24 -0
- special_tokens_map.json +34 -0
- tokenizer.json +3 -0
- tokenizer.model +3 -0
- tokenizer_config.json +1757 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,3 +1,105 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- AIDC-AI/Ovis-dataset
|
| 5 |
+
library_name: transformers
|
| 6 |
+
tags:
|
| 7 |
+
- MLLM
|
| 8 |
+
pipeline_tag: image-text-to-text
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
## Introduction
|
| 12 |
+
Ovis is a novel Multimodal Large Language Model (MLLM) architecture, designed to structurally align visual and textual embeddings. For a comprehensive introduction, please refer to [Ovis paper](https://arxiv.org/abs/2405.20797) and [Ovis GitHub](https://github.com/AIDC-AI/Ovis).
|
| 13 |
+
|
| 14 |
+
<div align="center">
|
| 15 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/658a8a837959448ef5500ce5/TIlymOb86R6_Mez3bpmcB.png" width="100%" />
|
| 16 |
+
</div>
|
| 17 |
+
|
| 18 |
+
## Model
|
| 19 |
+
As always, Ovis1.5 remains fully open-source: we release the training datasets, training & inference codes, and model weights for **reproducible transparency** and community collaboration.
|
| 20 |
+
|
| 21 |
+
| Ovis MLLMs | ViT | LLM | Training Datasets | Code | Model Weights |
|
| 22 |
+
|:-------------------------|:-----------:|:------------------:|:-------------------------------------------------------------------:|:-------------------------------------------:|:----------------------------------------------------------------:|
|
| 23 |
+
| Ovis1.5-Llama3-8B | Siglip-400M | Llama3-8B-Instruct | [Huggingface](https://huggingface.co/datasets/AIDC-AI/Ovis-dataset) | [Github](https://github.com/AIDC-AI/Ovis) | [Huggingface](https://huggingface.co/AIDC-AI/Ovis1.5-Llama3-8B) |
|
| 24 |
+
| Ovis1.5-Gemma2-9B | Siglip-400M | Gemma2-9B-It | [Huggingface](https://huggingface.co/datasets/AIDC-AI/Ovis-dataset) | [Github](https://github.com/AIDC-AI/Ovis) | [Huggingface](https://huggingface.co/AIDC-AI/Ovis1.5-Gemma2-9B) |
|
| 25 |
+
|
| 26 |
+
## Performance
|
| 27 |
+
We evaluate Ovis across various multimodal benchmarks using [VLMEvalKit](https://github.com/open-compass/VLMEvalKit) and compare its performance to leading MLLMs with similar parameter scales.
|
| 28 |
+
|
| 29 |
+
| | MiniCPM-Llama3-V2.5 | GLM-4V-9B | Ovis1.5-Llama3-8B | Ovis1.5-Gemma2-9B |
|
| 30 |
+
|:------------------|--------------------:|----------:|------------------:|------------------:|
|
| 31 |
+
| Open Weights | ✅ | ✅ | ✅ | ✅ |
|
| 32 |
+
| Open Datasets | ❌ | ❌ | ✅ | ✅ |
|
| 33 |
+
| MMTBench-VAL | 57.6 | 48.8 | 60.7 | **62.7** |
|
| 34 |
+
| MMBench-EN-V1.1 | 74 | 68.7 | **78.2** | 78.0 |
|
| 35 |
+
| MMBench-CN-V1.1 | 70.1 | 67.1 | **75.2** | 75.1 |
|
| 36 |
+
| MMStar | 51.8 | 54.8 | 57.2 | **58.7** |
|
| 37 |
+
| MMMU-Val | 45.8 | 46.9 | 48.6 | **49.8** |
|
| 38 |
+
| MathVista-Mini | 54.3 | 51.1 | 62.4 | **65.7** |
|
| 39 |
+
| HallusionBenchAvg | 42.4 | 45 | 44.5 | **48.0** |
|
| 40 |
+
| AI2D | 78.4 | 71.2 | 82.5 | **84.7** |
|
| 41 |
+
| OCRBench | 725 | **776** | 743 | 756 |
|
| 42 |
+
| MMVet | 52.8 | **58** | 52.2 | 56.5 |
|
| 43 |
+
| RealWorldQA | 63.5 | 66 | 64.6 | **66.9** |
|
| 44 |
+
|
| 45 |
+
## Usage
|
| 46 |
+
Below is a code snippet to run Ovis with multimodal inputs. For additional usage instructions, including inference wrapper and Gradio UI, please refer to [Ovis GitHub](https://github.com/AIDC-AI/Ovis?tab=readme-ov-file#inference).
|
| 47 |
+
```bash
|
| 48 |
+
pip install torch==2.1.2 transformers==4.43.2 pillow==10.3.0
|
| 49 |
+
```
|
| 50 |
+
```python
|
| 51 |
+
import torch
|
| 52 |
+
from PIL import Image
|
| 53 |
+
from transformers import AutoModelForCausalLM
|
| 54 |
+
|
| 55 |
+
# load model
|
| 56 |
+
model = AutoModelForCausalLM.from_pretrained("AIDC-AI/Ovis1.5-Gemma2-9B",
|
| 57 |
+
torch_dtype=torch.bfloat16,
|
| 58 |
+
multimodal_max_length=8192,
|
| 59 |
+
trust_remote_code=True).cuda()
|
| 60 |
+
text_tokenizer = model.get_text_tokenizer()
|
| 61 |
+
visual_tokenizer = model.get_visual_tokenizer()
|
| 62 |
+
conversation_formatter = model.get_conversation_formatter()
|
| 63 |
+
|
| 64 |
+
# enter image path and prompt
|
| 65 |
+
image_path = input("Enter image path: ")
|
| 66 |
+
image = Image.open(image_path)
|
| 67 |
+
text = input("Enter prompt: ")
|
| 68 |
+
query = f'<image>\n{text}'
|
| 69 |
+
prompt, input_ids = conversation_formatter.format_query(query)
|
| 70 |
+
input_ids = torch.unsqueeze(input_ids, dim=0).to(device=model.device)
|
| 71 |
+
attention_mask = torch.ne(input_ids, text_tokenizer.pad_token_id).to(device=model.device)
|
| 72 |
+
pixel_values = [visual_tokenizer.preprocess_image(image).to(
|
| 73 |
+
dtype=visual_tokenizer.dtype, device=visual_tokenizer.device)]
|
| 74 |
+
|
| 75 |
+
# generate output
|
| 76 |
+
with torch.inference_mode():
|
| 77 |
+
gen_kwargs = dict(
|
| 78 |
+
max_new_tokens=1024,
|
| 79 |
+
do_sample=False,
|
| 80 |
+
top_p=None,
|
| 81 |
+
top_k=None,
|
| 82 |
+
temperature=None,
|
| 83 |
+
repetition_penalty=None,
|
| 84 |
+
eos_token_id=model.generation_config.eos_token_id,
|
| 85 |
+
pad_token_id=text_tokenizer.pad_token_id,
|
| 86 |
+
use_cache=True
|
| 87 |
+
)
|
| 88 |
+
output_ids = model.generate(input_ids, pixel_values=pixel_values, attention_mask=attention_mask, **gen_kwargs)[0]
|
| 89 |
+
output = text_tokenizer.decode(output_ids, skip_special_tokens=True)
|
| 90 |
+
print(f'Output: {output}')
|
| 91 |
+
```
|
| 92 |
+
|
| 93 |
+
## Citation
|
| 94 |
+
If you find Ovis useful, please cite the paper
|
| 95 |
+
```
|
| 96 |
+
@article{lu2024ovis,
|
| 97 |
+
title={Ovis: Structural Embedding Alignment for Multimodal Large Language Model},
|
| 98 |
+
author={Shiyin Lu and Yang Li and Qing-Guo Chen and Zhao Xu and Weihua Luo and Kaifu Zhang and Han-Jia Ye},
|
| 99 |
+
year={2024},
|
| 100 |
+
journal={arXiv:2405.20797}
|
| 101 |
+
}
|
| 102 |
+
```
|
| 103 |
+
|
| 104 |
+
## License
|
| 105 |
+
The project is licensed under the Apache 2.0 License and is restricted to uses that comply with the license agreements of Qwen, Llama3, Clip, and Siglip.
|
config.json
ADDED
|
@@ -0,0 +1,248 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Ovis"
|
| 4 |
+
],
|
| 5 |
+
"auto_map": {
|
| 6 |
+
"AutoConfig": "configuration_ovis.OvisConfig",
|
| 7 |
+
"AutoModelForCausalLM": "modeling_ovis.Ovis"
|
| 8 |
+
},
|
| 9 |
+
"conversation_formatter_class": "GemmaConversationFormatter",
|
| 10 |
+
"hidden_size": 3584,
|
| 11 |
+
"llm_config": {
|
| 12 |
+
"_name_or_path": "google/gemma-2-9b-it",
|
| 13 |
+
"add_cross_attention": false,
|
| 14 |
+
"architectures": [
|
| 15 |
+
"Gemma2ForCausalLM"
|
| 16 |
+
],
|
| 17 |
+
"attention_bias": false,
|
| 18 |
+
"attention_dropout": 0.0,
|
| 19 |
+
"attn_logit_softcapping": 50.0,
|
| 20 |
+
"bad_words_ids": null,
|
| 21 |
+
"begin_suppress_tokens": null,
|
| 22 |
+
"bos_token_id": 2,
|
| 23 |
+
"cache_implementation": "hybrid",
|
| 24 |
+
"chunk_size_feed_forward": 0,
|
| 25 |
+
"cross_attention_hidden_size": null,
|
| 26 |
+
"decoder_start_token_id": null,
|
| 27 |
+
"diversity_penalty": 0.0,
|
| 28 |
+
"do_sample": false,
|
| 29 |
+
"early_stopping": false,
|
| 30 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 31 |
+
"eos_token_id": 1,
|
| 32 |
+
"exponential_decay_length_penalty": null,
|
| 33 |
+
"final_logit_softcapping": 30.0,
|
| 34 |
+
"finetuning_task": null,
|
| 35 |
+
"forced_bos_token_id": null,
|
| 36 |
+
"forced_eos_token_id": null,
|
| 37 |
+
"head_dim": 256,
|
| 38 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 39 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 40 |
+
"hidden_size": 3584,
|
| 41 |
+
"id2label": {
|
| 42 |
+
"0": "LABEL_0",
|
| 43 |
+
"1": "LABEL_1"
|
| 44 |
+
},
|
| 45 |
+
"initializer_range": 0.02,
|
| 46 |
+
"intermediate_size": 14336,
|
| 47 |
+
"is_decoder": false,
|
| 48 |
+
"is_encoder_decoder": false,
|
| 49 |
+
"label2id": {
|
| 50 |
+
"LABEL_0": 0,
|
| 51 |
+
"LABEL_1": 1
|
| 52 |
+
},
|
| 53 |
+
"length_penalty": 1.0,
|
| 54 |
+
"max_length": 20,
|
| 55 |
+
"max_position_embeddings": 8192,
|
| 56 |
+
"min_length": 0,
|
| 57 |
+
"model_type": "gemma2",
|
| 58 |
+
"no_repeat_ngram_size": 0,
|
| 59 |
+
"num_attention_heads": 16,
|
| 60 |
+
"num_beam_groups": 1,
|
| 61 |
+
"num_beams": 1,
|
| 62 |
+
"num_hidden_layers": 42,
|
| 63 |
+
"num_key_value_heads": 8,
|
| 64 |
+
"num_return_sequences": 1,
|
| 65 |
+
"output_attentions": false,
|
| 66 |
+
"output_hidden_states": false,
|
| 67 |
+
"output_scores": false,
|
| 68 |
+
"pad_token_id": 0,
|
| 69 |
+
"prefix": null,
|
| 70 |
+
"problem_type": null,
|
| 71 |
+
"pruned_heads": {},
|
| 72 |
+
"query_pre_attn_scalar": 256,
|
| 73 |
+
"remove_invalid_values": false,
|
| 74 |
+
"repetition_penalty": 1.0,
|
| 75 |
+
"return_dict": true,
|
| 76 |
+
"return_dict_in_generate": false,
|
| 77 |
+
"rms_norm_eps": 1e-06,
|
| 78 |
+
"rope_theta": 10000.0,
|
| 79 |
+
"sep_token_id": null,
|
| 80 |
+
"sliding_window": 4096,
|
| 81 |
+
"sliding_window_size": 4096,
|
| 82 |
+
"suppress_tokens": null,
|
| 83 |
+
"task_specific_params": null,
|
| 84 |
+
"temperature": 1.0,
|
| 85 |
+
"tf_legacy_loss": false,
|
| 86 |
+
"tie_encoder_decoder": false,
|
| 87 |
+
"tie_word_embeddings": true,
|
| 88 |
+
"tokenizer_class": null,
|
| 89 |
+
"top_k": 50,
|
| 90 |
+
"top_p": 1.0,
|
| 91 |
+
"torch_dtype": "bfloat16",
|
| 92 |
+
"torchscript": false,
|
| 93 |
+
"typical_p": 1.0,
|
| 94 |
+
"use_bfloat16": false,
|
| 95 |
+
"use_cache": true,
|
| 96 |
+
"vocab_size": 256000
|
| 97 |
+
},
|
| 98 |
+
"model_type": "ovis",
|
| 99 |
+
"multimodal_max_length": 8192,
|
| 100 |
+
"torch_dtype": "bfloat16",
|
| 101 |
+
"transformers_version": "4.43.2",
|
| 102 |
+
"use_cache": true,
|
| 103 |
+
"visual_tokenizer_config": {
|
| 104 |
+
"_name_or_path": "",
|
| 105 |
+
"add_cross_attention": false,
|
| 106 |
+
"architectures": null,
|
| 107 |
+
"backbone_config": {
|
| 108 |
+
"_name_or_path": "google/siglip-so400m-patch14-384",
|
| 109 |
+
"add_cross_attention": false,
|
| 110 |
+
"architectures": null,
|
| 111 |
+
"attention_dropout": 0.0,
|
| 112 |
+
"bad_words_ids": null,
|
| 113 |
+
"begin_suppress_tokens": null,
|
| 114 |
+
"bos_token_id": null,
|
| 115 |
+
"chunk_size_feed_forward": 0,
|
| 116 |
+
"cross_attention_hidden_size": null,
|
| 117 |
+
"decoder_start_token_id": null,
|
| 118 |
+
"diversity_penalty": 0.0,
|
| 119 |
+
"do_sample": false,
|
| 120 |
+
"early_stopping": false,
|
| 121 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 122 |
+
"eos_token_id": null,
|
| 123 |
+
"exponential_decay_length_penalty": null,
|
| 124 |
+
"finetuning_task": null,
|
| 125 |
+
"forced_bos_token_id": null,
|
| 126 |
+
"forced_eos_token_id": null,
|
| 127 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 128 |
+
"hidden_size": 1152,
|
| 129 |
+
"id2label": {
|
| 130 |
+
"0": "LABEL_0",
|
| 131 |
+
"1": "LABEL_1"
|
| 132 |
+
},
|
| 133 |
+
"image_size": 384,
|
| 134 |
+
"intermediate_size": 4304,
|
| 135 |
+
"is_decoder": false,
|
| 136 |
+
"is_encoder_decoder": false,
|
| 137 |
+
"label2id": {
|
| 138 |
+
"LABEL_0": 0,
|
| 139 |
+
"LABEL_1": 1
|
| 140 |
+
},
|
| 141 |
+
"layer_norm_eps": 1e-06,
|
| 142 |
+
"length_penalty": 1.0,
|
| 143 |
+
"max_length": 20,
|
| 144 |
+
"min_length": 0,
|
| 145 |
+
"model_type": "siglip_vision_model",
|
| 146 |
+
"no_repeat_ngram_size": 0,
|
| 147 |
+
"num_attention_heads": 16,
|
| 148 |
+
"num_beam_groups": 1,
|
| 149 |
+
"num_beams": 1,
|
| 150 |
+
"num_channels": 3,
|
| 151 |
+
"num_hidden_layers": 27,
|
| 152 |
+
"num_return_sequences": 1,
|
| 153 |
+
"output_attentions": false,
|
| 154 |
+
"output_hidden_states": false,
|
| 155 |
+
"output_scores": false,
|
| 156 |
+
"pad_token_id": null,
|
| 157 |
+
"patch_size": 14,
|
| 158 |
+
"prefix": null,
|
| 159 |
+
"problem_type": null,
|
| 160 |
+
"pruned_heads": {},
|
| 161 |
+
"remove_invalid_values": false,
|
| 162 |
+
"repetition_penalty": 1.0,
|
| 163 |
+
"return_dict": true,
|
| 164 |
+
"return_dict_in_generate": false,
|
| 165 |
+
"sep_token_id": null,
|
| 166 |
+
"suppress_tokens": null,
|
| 167 |
+
"task_specific_params": null,
|
| 168 |
+
"temperature": 1.0,
|
| 169 |
+
"tf_legacy_loss": false,
|
| 170 |
+
"tie_encoder_decoder": false,
|
| 171 |
+
"tie_word_embeddings": true,
|
| 172 |
+
"tokenizer_class": null,
|
| 173 |
+
"top_k": 50,
|
| 174 |
+
"top_p": 1.0,
|
| 175 |
+
"torch_dtype": null,
|
| 176 |
+
"torchscript": false,
|
| 177 |
+
"typical_p": 1.0,
|
| 178 |
+
"use_bfloat16": false
|
| 179 |
+
},
|
| 180 |
+
"backbone_kwargs": {},
|
| 181 |
+
"bad_words_ids": null,
|
| 182 |
+
"begin_suppress_tokens": null,
|
| 183 |
+
"bos_token_id": null,
|
| 184 |
+
"chunk_size_feed_forward": 0,
|
| 185 |
+
"cross_attention_hidden_size": null,
|
| 186 |
+
"decoder_start_token_id": null,
|
| 187 |
+
"depths": null,
|
| 188 |
+
"diversity_penalty": 0.0,
|
| 189 |
+
"do_sample": false,
|
| 190 |
+
"drop_cls_token": false,
|
| 191 |
+
"early_stopping": false,
|
| 192 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 193 |
+
"eos_token_id": null,
|
| 194 |
+
"exponential_decay_length_penalty": null,
|
| 195 |
+
"finetuning_task": null,
|
| 196 |
+
"forced_bos_token_id": null,
|
| 197 |
+
"forced_eos_token_id": null,
|
| 198 |
+
"hd_booster": "s2wrapper",
|
| 199 |
+
"hidden_stride": 1,
|
| 200 |
+
"id2label": {
|
| 201 |
+
"0": "LABEL_0",
|
| 202 |
+
"1": "LABEL_1"
|
| 203 |
+
},
|
| 204 |
+
"is_decoder": false,
|
| 205 |
+
"is_encoder_decoder": false,
|
| 206 |
+
"label2id": {
|
| 207 |
+
"LABEL_0": 0,
|
| 208 |
+
"LABEL_1": 1
|
| 209 |
+
},
|
| 210 |
+
"length_penalty": 1.0,
|
| 211 |
+
"max_length": 20,
|
| 212 |
+
"min_length": 0,
|
| 213 |
+
"model_type": "siglip_visual_tokenizer",
|
| 214 |
+
"no_repeat_ngram_size": 0,
|
| 215 |
+
"num_beam_groups": 1,
|
| 216 |
+
"num_beams": 1,
|
| 217 |
+
"num_return_sequences": 1,
|
| 218 |
+
"output_attentions": false,
|
| 219 |
+
"output_hidden_states": false,
|
| 220 |
+
"output_scores": false,
|
| 221 |
+
"pad_token_id": null,
|
| 222 |
+
"prefix": null,
|
| 223 |
+
"problem_type": null,
|
| 224 |
+
"pruned_heads": {},
|
| 225 |
+
"remove_invalid_values": false,
|
| 226 |
+
"repetition_penalty": 1.0,
|
| 227 |
+
"return_dict": true,
|
| 228 |
+
"return_dict_in_generate": false,
|
| 229 |
+
"sep_token_id": null,
|
| 230 |
+
"suppress_tokens": null,
|
| 231 |
+
"task_specific_params": null,
|
| 232 |
+
"tau": 1.0,
|
| 233 |
+
"temperature": 1.0,
|
| 234 |
+
"tf_legacy_loss": false,
|
| 235 |
+
"tie_encoder_decoder": false,
|
| 236 |
+
"tie_word_embeddings": true,
|
| 237 |
+
"tokenize_function": "softmax",
|
| 238 |
+
"tokenizer_class": null,
|
| 239 |
+
"top_k": 50,
|
| 240 |
+
"top_p": 1.0,
|
| 241 |
+
"torch_dtype": null,
|
| 242 |
+
"torchscript": false,
|
| 243 |
+
"typical_p": 1.0,
|
| 244 |
+
"use_bfloat16": false,
|
| 245 |
+
"use_indicators": true,
|
| 246 |
+
"vocab_size": 131072
|
| 247 |
+
}
|
| 248 |
+
}
|
configuration_ovis.py
ADDED
|
@@ -0,0 +1,359 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import logging
|
| 2 |
+
from abc import ABC, abstractmethod
|
| 3 |
+
from typing import List, Dict, Union, Optional
|
| 4 |
+
|
| 5 |
+
import torch
|
| 6 |
+
from transformers import PretrainedConfig, AutoConfig
|
| 7 |
+
|
| 8 |
+
IGNORE_INDEX = -100
|
| 9 |
+
IMAGE_TOKEN_INDEX = -200
|
| 10 |
+
IMAGE_TOKEN = "<image>"
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
# ----------------------------------------------------------------------
|
| 14 |
+
# Visual Tokenizer Configuration
|
| 15 |
+
# ----------------------------------------------------------------------
|
| 16 |
+
class BaseVisualTokenizerConfig(PretrainedConfig):
|
| 17 |
+
def __init__(
|
| 18 |
+
self,
|
| 19 |
+
vocab_size=16384,
|
| 20 |
+
tokenize_function="softmax",
|
| 21 |
+
tau=1.0,
|
| 22 |
+
depths=None,
|
| 23 |
+
use_indicators=False,
|
| 24 |
+
drop_cls_token=False,
|
| 25 |
+
backbone_config: Optional[Union[PretrainedConfig, dict]] = None,
|
| 26 |
+
hidden_stride: int = 1,
|
| 27 |
+
hd_booster: Optional[str] = None,
|
| 28 |
+
**kwargs
|
| 29 |
+
):
|
| 30 |
+
super().__init__(**kwargs)
|
| 31 |
+
self.vocab_size = vocab_size
|
| 32 |
+
self.tokenize_function = tokenize_function
|
| 33 |
+
self.tau = tau
|
| 34 |
+
if isinstance(depths, str):
|
| 35 |
+
depths = [int(x) for x in depths.split('|')]
|
| 36 |
+
self.depths = depths
|
| 37 |
+
self.backbone_kwargs = {}
|
| 38 |
+
self.use_indicators = use_indicators
|
| 39 |
+
self.drop_cls_token = drop_cls_token
|
| 40 |
+
if backbone_config is not None:
|
| 41 |
+
assert isinstance(backbone_config, (PretrainedConfig, dict)), \
|
| 42 |
+
(f"expect `backbone_config` to be instance of PretrainedConfig or dict,"
|
| 43 |
+
f" but got {type(backbone_config)} type")
|
| 44 |
+
if not isinstance(backbone_config, PretrainedConfig):
|
| 45 |
+
model_type = backbone_config['model_type']
|
| 46 |
+
backbone_config.pop('model_type')
|
| 47 |
+
backbone_config = AutoConfig.for_model(model_type, **backbone_config)
|
| 48 |
+
self.backbone_config = backbone_config
|
| 49 |
+
self.hidden_stride = hidden_stride
|
| 50 |
+
self.hd_booster = hd_booster
|
| 51 |
+
|
| 52 |
+
|
| 53 |
+
class ClipVisualTokenizerConfig(BaseVisualTokenizerConfig):
|
| 54 |
+
model_type = "clip_visual_tokenizer"
|
| 55 |
+
|
| 56 |
+
def __init__(self, **kwargs):
|
| 57 |
+
super().__init__(**kwargs)
|
| 58 |
+
if self.depths:
|
| 59 |
+
assert len(self.depths) == 1
|
| 60 |
+
self.backbone_kwargs['num_hidden_layers'] = self.depths[0]
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
class SiglipVisualTokenizerConfig(BaseVisualTokenizerConfig):
|
| 64 |
+
model_type = "siglip_visual_tokenizer"
|
| 65 |
+
|
| 66 |
+
def __init__(self, **kwargs):
|
| 67 |
+
super().__init__(**kwargs)
|
| 68 |
+
if self.drop_cls_token:
|
| 69 |
+
logging.warning(
|
| 70 |
+
f'SiglipVisionModel has no cls token,'
|
| 71 |
+
f' so `drop_cls_token=True` is ignored and reset to `False`')
|
| 72 |
+
self.drop_cls_token = False
|
| 73 |
+
if self.depths:
|
| 74 |
+
assert len(self.depths) == 1
|
| 75 |
+
self.backbone_kwargs['num_hidden_layers'] = self.depths[0]
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
AutoConfig.register("clip_visual_tokenizer", ClipVisualTokenizerConfig)
|
| 79 |
+
AutoConfig.register("siglip_visual_tokenizer", SiglipVisualTokenizerConfig)
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
# ----------------------------------------------------------------------
|
| 83 |
+
# Ovis Configuration
|
| 84 |
+
# ----------------------------------------------------------------------
|
| 85 |
+
class OvisConfig(PretrainedConfig):
|
| 86 |
+
model_type = "ovis"
|
| 87 |
+
|
| 88 |
+
def __init__(
|
| 89 |
+
self,
|
| 90 |
+
llm_config: Optional[Union[PretrainedConfig, dict]] = None,
|
| 91 |
+
visual_tokenizer_config: Optional[Union[PretrainedConfig, dict]] = None,
|
| 92 |
+
multimodal_max_length=2048,
|
| 93 |
+
hidden_size=None,
|
| 94 |
+
conversation_formatter_class=None,
|
| 95 |
+
**kwargs
|
| 96 |
+
):
|
| 97 |
+
super().__init__(**kwargs)
|
| 98 |
+
if llm_config is not None:
|
| 99 |
+
assert isinstance(llm_config, (PretrainedConfig, dict)), \
|
| 100 |
+
(f"expect `llm_config` to be instance of PretrainedConfig or dict,"
|
| 101 |
+
f" but got {type(llm_config)} type")
|
| 102 |
+
if not isinstance(llm_config, PretrainedConfig):
|
| 103 |
+
model_type = llm_config['model_type']
|
| 104 |
+
llm_config.pop('model_type')
|
| 105 |
+
llm_config = AutoConfig.for_model(model_type, **llm_config)
|
| 106 |
+
self.llm_config = llm_config
|
| 107 |
+
if visual_tokenizer_config is not None:
|
| 108 |
+
assert isinstance(visual_tokenizer_config, (PretrainedConfig, dict)), \
|
| 109 |
+
(f"expect `visual_tokenizer_config` to be instance of PretrainedConfig or dict,"
|
| 110 |
+
f" but got {type(visual_tokenizer_config)} type")
|
| 111 |
+
if not isinstance(visual_tokenizer_config, PretrainedConfig):
|
| 112 |
+
model_type = visual_tokenizer_config['model_type']
|
| 113 |
+
visual_tokenizer_config.pop('model_type')
|
| 114 |
+
visual_tokenizer_config = AutoConfig.for_model(model_type, **visual_tokenizer_config)
|
| 115 |
+
self.visual_tokenizer_config = visual_tokenizer_config
|
| 116 |
+
self.multimodal_max_length = multimodal_max_length
|
| 117 |
+
self.hidden_size = hidden_size
|
| 118 |
+
self.conversation_formatter_class = conversation_formatter_class
|
| 119 |
+
|
| 120 |
+
|
| 121 |
+
# ----------------------------------------------------------------------
|
| 122 |
+
# Conversation Formatter
|
| 123 |
+
# ----------------------------------------------------------------------
|
| 124 |
+
class ConversationFormatter(ABC):
|
| 125 |
+
support_tokenizer_types = None
|
| 126 |
+
|
| 127 |
+
def __init__(self, tokenizer):
|
| 128 |
+
tokenizer_type = type(tokenizer).__name__
|
| 129 |
+
assert tokenizer_type in self.support_tokenizer_types, \
|
| 130 |
+
(f'Invalid tokenizer type, expected one from `{self.support_tokenizer_types}`,'
|
| 131 |
+
f' but got `{tokenizer_type}`')
|
| 132 |
+
self.tokenizer = tokenizer
|
| 133 |
+
self.image_symbol = IMAGE_TOKEN
|
| 134 |
+
self.image_token_index = IMAGE_TOKEN_INDEX
|
| 135 |
+
self.ignore_index = IGNORE_INDEX
|
| 136 |
+
|
| 137 |
+
def _tokenize_with_image_symbol(self, text):
|
| 138 |
+
text_chunks = [self.tokenizer(chunk, add_special_tokens=False).input_ids for chunk in
|
| 139 |
+
text.split(self.image_symbol)]
|
| 140 |
+
token_ids = []
|
| 141 |
+
num_chuck = len(text_chunks)
|
| 142 |
+
for i, chunk in enumerate(text_chunks):
|
| 143 |
+
token_ids.extend(chunk)
|
| 144 |
+
if i < num_chuck - 1:
|
| 145 |
+
token_ids.append(self.image_token_index)
|
| 146 |
+
return token_ids
|
| 147 |
+
|
| 148 |
+
@abstractmethod
|
| 149 |
+
def format(self, conversations: List[Dict], generation_preface=None):
|
| 150 |
+
pass
|
| 151 |
+
|
| 152 |
+
@abstractmethod
|
| 153 |
+
def format_query(self, query, generation_preface=""):
|
| 154 |
+
pass
|
| 155 |
+
|
| 156 |
+
|
| 157 |
+
class QwenConversationFormatter(ConversationFormatter):
|
| 158 |
+
support_tokenizer_types = ['QWenTokenizer', 'Qwen2TokenizerFast']
|
| 159 |
+
|
| 160 |
+
def __init__(self, tokenizer):
|
| 161 |
+
super().__init__(tokenizer)
|
| 162 |
+
self.from2role = {
|
| 163 |
+
"system": "<|im_start|>system\n",
|
| 164 |
+
"human": "<|im_start|>user\n",
|
| 165 |
+
"gpt": "<|im_start|>assistant\n",
|
| 166 |
+
}
|
| 167 |
+
self.gpt_token_num = None
|
| 168 |
+
self.im_end = "<|im_end|>\n"
|
| 169 |
+
self.default_system_prompt = "You are a helpful assistant."
|
| 170 |
+
|
| 171 |
+
def format(self, conversations: List[Dict], generation_preface=None):
|
| 172 |
+
if self.gpt_token_num is None:
|
| 173 |
+
self.gpt_token_num = len(
|
| 174 |
+
self.tokenizer(self.from2role["gpt"], add_special_tokens=False).input_ids)
|
| 175 |
+
|
| 176 |
+
if conversations[0]["from"] != "system":
|
| 177 |
+
conversations.insert(0, {
|
| 178 |
+
"from": "system",
|
| 179 |
+
"value": self.default_system_prompt
|
| 180 |
+
})
|
| 181 |
+
|
| 182 |
+
if generation_preface is not None:
|
| 183 |
+
conversations.append({
|
| 184 |
+
"from": "gpt",
|
| 185 |
+
"value": generation_preface
|
| 186 |
+
})
|
| 187 |
+
|
| 188 |
+
prompt = ""
|
| 189 |
+
input_ids = []
|
| 190 |
+
labels = []
|
| 191 |
+
num_conversation = len(conversations)
|
| 192 |
+
for i, conversation in enumerate(conversations):
|
| 193 |
+
frm = conversation["from"]
|
| 194 |
+
role = self.from2role[frm]
|
| 195 |
+
message = conversation["value"]
|
| 196 |
+
text = role + message
|
| 197 |
+
if i < num_conversation - 1 or generation_preface is None:
|
| 198 |
+
text += self.im_end
|
| 199 |
+
prompt += text
|
| 200 |
+
token_ids = self._tokenize_with_image_symbol(text)
|
| 201 |
+
input_ids.extend(token_ids)
|
| 202 |
+
label_ids = [self.ignore_index] * len(token_ids)
|
| 203 |
+
if frm == "gpt" and generation_preface is None:
|
| 204 |
+
# learning `\n` following `im_end` is meaningless, so the last `\n` token is ignored in label
|
| 205 |
+
label_ids[self.gpt_token_num:-1] = token_ids[self.gpt_token_num:-1]
|
| 206 |
+
labels.extend(label_ids)
|
| 207 |
+
|
| 208 |
+
assert self._tokenize_with_image_symbol(prompt) == input_ids
|
| 209 |
+
assert len(input_ids) == len(labels)
|
| 210 |
+
input_ids = torch.tensor(input_ids, dtype=torch.long)
|
| 211 |
+
labels = torch.tensor(labels, dtype=torch.long)
|
| 212 |
+
|
| 213 |
+
return prompt, input_ids, labels
|
| 214 |
+
|
| 215 |
+
def format_query(self, query, generation_preface=""):
|
| 216 |
+
prompt, input_ids, _ = self.format([{
|
| 217 |
+
"from": "human",
|
| 218 |
+
"value": query
|
| 219 |
+
}], generation_preface=generation_preface)
|
| 220 |
+
|
| 221 |
+
return prompt, input_ids
|
| 222 |
+
|
| 223 |
+
|
| 224 |
+
class Llama3ConversationFormatter(ConversationFormatter):
|
| 225 |
+
support_tokenizer_types = ['PreTrainedTokenizerFast']
|
| 226 |
+
|
| 227 |
+
def __init__(self, tokenizer):
|
| 228 |
+
super().__init__(tokenizer)
|
| 229 |
+
self.from2role = {
|
| 230 |
+
"system": "<|start_header_id|>system<|end_header_id|>\n\n",
|
| 231 |
+
"human": "<|start_header_id|>user<|end_header_id|>\n\n",
|
| 232 |
+
"gpt": "<|start_header_id|>assistant<|end_header_id|>\n\n",
|
| 233 |
+
}
|
| 234 |
+
self.gpt_token_num = None
|
| 235 |
+
self.im_end = "<|eot_id|>"
|
| 236 |
+
self.default_system_prompt = "You are a helpful and honest multimodal assistant."
|
| 237 |
+
self.bos_token = "<|begin_of_text|>"
|
| 238 |
+
self.bos_token_ids = None
|
| 239 |
+
|
| 240 |
+
def format(self, conversations: List[Dict], generation_preface=None):
|
| 241 |
+
if self.gpt_token_num is None:
|
| 242 |
+
self.gpt_token_num = len(
|
| 243 |
+
self.tokenizer(self.from2role["gpt"], add_special_tokens=False).input_ids)
|
| 244 |
+
|
| 245 |
+
if self.bos_token_ids is None:
|
| 246 |
+
self.bos_token_ids = self.tokenizer(self.bos_token, add_special_tokens=False).input_ids
|
| 247 |
+
|
| 248 |
+
if conversations[0]["from"] != "system":
|
| 249 |
+
conversations.insert(0, {
|
| 250 |
+
"from": "system",
|
| 251 |
+
"value": self.default_system_prompt
|
| 252 |
+
})
|
| 253 |
+
|
| 254 |
+
if generation_preface is not None:
|
| 255 |
+
conversations.append({
|
| 256 |
+
"from": "gpt",
|
| 257 |
+
"value": generation_preface
|
| 258 |
+
})
|
| 259 |
+
|
| 260 |
+
prompt = "" + self.bos_token
|
| 261 |
+
input_ids = [] + self.bos_token_ids
|
| 262 |
+
labels = [] + [IGNORE_INDEX] * len(input_ids)
|
| 263 |
+
num_conversation = len(conversations)
|
| 264 |
+
for i, conversation in enumerate(conversations):
|
| 265 |
+
frm = conversation["from"]
|
| 266 |
+
role = self.from2role[frm]
|
| 267 |
+
message = conversation["value"].strip()
|
| 268 |
+
text = role + message
|
| 269 |
+
if i < num_conversation - 1 or generation_preface is None:
|
| 270 |
+
text += self.im_end
|
| 271 |
+
prompt += text
|
| 272 |
+
token_ids = self._tokenize_with_image_symbol(text)
|
| 273 |
+
input_ids.extend(token_ids)
|
| 274 |
+
label_ids = [self.ignore_index] * len(token_ids)
|
| 275 |
+
if frm == "gpt":
|
| 276 |
+
label_ids[self.gpt_token_num:] = token_ids[self.gpt_token_num:]
|
| 277 |
+
labels.extend(label_ids)
|
| 278 |
+
|
| 279 |
+
assert self._tokenize_with_image_symbol(prompt) == input_ids
|
| 280 |
+
assert len(input_ids) == len(labels)
|
| 281 |
+
input_ids = torch.tensor(input_ids, dtype=torch.long)
|
| 282 |
+
labels = torch.tensor(labels, dtype=torch.long)
|
| 283 |
+
|
| 284 |
+
return prompt, input_ids, labels
|
| 285 |
+
|
| 286 |
+
def format_query(self, query, generation_preface=""):
|
| 287 |
+
prompt, input_ids, _ = self.format([{
|
| 288 |
+
"from": "human",
|
| 289 |
+
"value": query
|
| 290 |
+
}], generation_preface=generation_preface)
|
| 291 |
+
|
| 292 |
+
return prompt, input_ids
|
| 293 |
+
|
| 294 |
+
|
| 295 |
+
class GemmaConversationFormatter(ConversationFormatter):
|
| 296 |
+
support_tokenizer_types = ['GemmaTokenizer', 'GemmaTokenizerFast']
|
| 297 |
+
|
| 298 |
+
def __init__(self, tokenizer):
|
| 299 |
+
super().__init__(tokenizer)
|
| 300 |
+
# Gemma does not support system prompt
|
| 301 |
+
self.from2role = {
|
| 302 |
+
"human": "<start_of_turn>user\n",
|
| 303 |
+
"gpt": "<start_of_turn>model\n",
|
| 304 |
+
}
|
| 305 |
+
self.gpt_token_num = None
|
| 306 |
+
self.im_end = "<end_of_turn>\n"
|
| 307 |
+
self.bos_token = "<bos>"
|
| 308 |
+
self.bos_token_ids = None
|
| 309 |
+
|
| 310 |
+
def format(self, conversations: List[Dict], generation_preface=None):
|
| 311 |
+
if self.gpt_token_num is None:
|
| 312 |
+
self.gpt_token_num = len(self.tokenizer(self.from2role["gpt"], add_special_tokens=False).input_ids)
|
| 313 |
+
|
| 314 |
+
if self.bos_token_ids is None:
|
| 315 |
+
self.bos_token_ids = self.tokenizer(self.bos_token, add_special_tokens=False).input_ids
|
| 316 |
+
|
| 317 |
+
if conversations[0]["from"] == "system":
|
| 318 |
+
raise ValueError("Gemma does not support system prompt")
|
| 319 |
+
|
| 320 |
+
if generation_preface is not None:
|
| 321 |
+
conversations.append({
|
| 322 |
+
"from": "gpt",
|
| 323 |
+
"value": generation_preface
|
| 324 |
+
})
|
| 325 |
+
|
| 326 |
+
prompt = "" + self.bos_token
|
| 327 |
+
input_ids = [] + self.bos_token_ids
|
| 328 |
+
labels = [] + [IGNORE_INDEX] * len(input_ids)
|
| 329 |
+
num_conversation = len(conversations)
|
| 330 |
+
for i, conversation in enumerate(conversations):
|
| 331 |
+
frm = conversation["from"]
|
| 332 |
+
role = self.from2role[frm]
|
| 333 |
+
message = conversation["value"].strip()
|
| 334 |
+
text = role + message
|
| 335 |
+
if i < num_conversation - 1 or generation_preface is None:
|
| 336 |
+
text += self.im_end
|
| 337 |
+
prompt += text
|
| 338 |
+
token_ids = self._tokenize_with_image_symbol(text)
|
| 339 |
+
input_ids.extend(token_ids)
|
| 340 |
+
label_ids = [self.ignore_index] * len(token_ids)
|
| 341 |
+
if frm == "gpt":
|
| 342 |
+
# learning `\n` following `im_end` is meaningless, so the last `\n` token is ignored in label
|
| 343 |
+
label_ids[self.gpt_token_num:-1] = token_ids[self.gpt_token_num:-1]
|
| 344 |
+
labels.extend(label_ids)
|
| 345 |
+
|
| 346 |
+
assert self._tokenize_with_image_symbol(prompt) == input_ids
|
| 347 |
+
assert len(input_ids) == len(labels)
|
| 348 |
+
input_ids = torch.tensor(input_ids, dtype=torch.long)
|
| 349 |
+
labels = torch.tensor(labels, dtype=torch.long)
|
| 350 |
+
|
| 351 |
+
return prompt, input_ids, labels
|
| 352 |
+
|
| 353 |
+
def format_query(self, query, generation_preface=""):
|
| 354 |
+
prompt, input_ids, _ = self.format([{
|
| 355 |
+
"from": "human",
|
| 356 |
+
"value": query
|
| 357 |
+
}], generation_preface=generation_preface)
|
| 358 |
+
|
| 359 |
+
return prompt, input_ids
|
generation_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
1,
|
| 7 |
+
107
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"transformers_version": "4.43.2"
|
| 11 |
+
}
|
model-00001-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82e359e93af136382bc2709013f6a9e1e5650a686ee0ac17f8ee33f52bfb9f71
|
| 3 |
+
size 4903352248
|
model-00002-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fc6212bdccce1a9721d62f8e17c8043abc404544059e1082487f247cea18a85
|
| 3 |
+
size 4947571432
|
model-00003-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55e6655d231cf58a3c50a580bbca82b7a2a9e21f3caceae084aadec4d0fbc4e0
|
| 3 |
+
size 4962222000
|
model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4682e643748338edc4f3f162128515cfd8eb3897aa864b71fc110b1e0d3ca128
|
| 3 |
+
size 3670322624
|
model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:113f0c89991a2cc80ac0db110e3b2200dee3f4410b7393392f6b395ea762c6ff
|
| 3 |
+
size 4235546240
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,924 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 22718890104
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"llm.lm_head.weight": "model-00005-of-00005.safetensors",
|
| 7 |
+
"llm.model.embed_tokens.weight": "model-00001-of-00005.safetensors",
|
| 8 |
+
"llm.model.layers.0.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 9 |
+
"llm.model.layers.0.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 10 |
+
"llm.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 11 |
+
"llm.model.layers.0.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 12 |
+
"llm.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 13 |
+
"llm.model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 14 |
+
"llm.model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 15 |
+
"llm.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 16 |
+
"llm.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 17 |
+
"llm.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 18 |
+
"llm.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 19 |
+
"llm.model.layers.1.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 20 |
+
"llm.model.layers.1.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 21 |
+
"llm.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 22 |
+
"llm.model.layers.1.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 23 |
+
"llm.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 24 |
+
"llm.model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 25 |
+
"llm.model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 26 |
+
"llm.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 27 |
+
"llm.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 28 |
+
"llm.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 29 |
+
"llm.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 30 |
+
"llm.model.layers.10.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 31 |
+
"llm.model.layers.10.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 32 |
+
"llm.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 33 |
+
"llm.model.layers.10.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 34 |
+
"llm.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 35 |
+
"llm.model.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 36 |
+
"llm.model.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 37 |
+
"llm.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 38 |
+
"llm.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 39 |
+
"llm.model.layers.10.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 40 |
+
"llm.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 41 |
+
"llm.model.layers.11.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 42 |
+
"llm.model.layers.11.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 43 |
+
"llm.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 44 |
+
"llm.model.layers.11.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 45 |
+
"llm.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 46 |
+
"llm.model.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 47 |
+
"llm.model.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 48 |
+
"llm.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 49 |
+
"llm.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 50 |
+
"llm.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 51 |
+
"llm.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 52 |
+
"llm.model.layers.12.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 53 |
+
"llm.model.layers.12.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 54 |
+
"llm.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 55 |
+
"llm.model.layers.12.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 56 |
+
"llm.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 57 |
+
"llm.model.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 58 |
+
"llm.model.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 59 |
+
"llm.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 60 |
+
"llm.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 61 |
+
"llm.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 62 |
+
"llm.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 63 |
+
"llm.model.layers.13.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 64 |
+
"llm.model.layers.13.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 65 |
+
"llm.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 66 |
+
"llm.model.layers.13.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 67 |
+
"llm.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 68 |
+
"llm.model.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 69 |
+
"llm.model.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 70 |
+
"llm.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 71 |
+
"llm.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 72 |
+
"llm.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 73 |
+
"llm.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 74 |
+
"llm.model.layers.14.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 75 |
+
"llm.model.layers.14.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 76 |
+
"llm.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 77 |
+
"llm.model.layers.14.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 78 |
+
"llm.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 79 |
+
"llm.model.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 80 |
+
"llm.model.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 81 |
+
"llm.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 82 |
+
"llm.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 83 |
+
"llm.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 84 |
+
"llm.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 85 |
+
"llm.model.layers.15.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 86 |
+
"llm.model.layers.15.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 87 |
+
"llm.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 88 |
+
"llm.model.layers.15.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 89 |
+
"llm.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 90 |
+
"llm.model.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 91 |
+
"llm.model.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 92 |
+
"llm.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 93 |
+
"llm.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 94 |
+
"llm.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 95 |
+
"llm.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 96 |
+
"llm.model.layers.16.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 97 |
+
"llm.model.layers.16.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 98 |
+
"llm.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 99 |
+
"llm.model.layers.16.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 100 |
+
"llm.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 101 |
+
"llm.model.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 102 |
+
"llm.model.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 103 |
+
"llm.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 104 |
+
"llm.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 105 |
+
"llm.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 106 |
+
"llm.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 107 |
+
"llm.model.layers.17.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 108 |
+
"llm.model.layers.17.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 109 |
+
"llm.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 110 |
+
"llm.model.layers.17.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 111 |
+
"llm.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 112 |
+
"llm.model.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 113 |
+
"llm.model.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 114 |
+
"llm.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 115 |
+
"llm.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 116 |
+
"llm.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 117 |
+
"llm.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 118 |
+
"llm.model.layers.18.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 119 |
+
"llm.model.layers.18.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 120 |
+
"llm.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 121 |
+
"llm.model.layers.18.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 122 |
+
"llm.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 123 |
+
"llm.model.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 124 |
+
"llm.model.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 125 |
+
"llm.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 126 |
+
"llm.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 127 |
+
"llm.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 128 |
+
"llm.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 129 |
+
"llm.model.layers.19.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 130 |
+
"llm.model.layers.19.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 131 |
+
"llm.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 132 |
+
"llm.model.layers.19.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 133 |
+
"llm.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 134 |
+
"llm.model.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 135 |
+
"llm.model.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 136 |
+
"llm.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 137 |
+
"llm.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 138 |
+
"llm.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 139 |
+
"llm.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 140 |
+
"llm.model.layers.2.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 141 |
+
"llm.model.layers.2.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 142 |
+
"llm.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 143 |
+
"llm.model.layers.2.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 144 |
+
"llm.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 145 |
+
"llm.model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 146 |
+
"llm.model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 147 |
+
"llm.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 148 |
+
"llm.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 149 |
+
"llm.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 150 |
+
"llm.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 151 |
+
"llm.model.layers.20.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 152 |
+
"llm.model.layers.20.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 153 |
+
"llm.model.layers.20.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 154 |
+
"llm.model.layers.20.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 155 |
+
"llm.model.layers.20.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 156 |
+
"llm.model.layers.20.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 157 |
+
"llm.model.layers.20.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 158 |
+
"llm.model.layers.20.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 159 |
+
"llm.model.layers.20.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 160 |
+
"llm.model.layers.20.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 161 |
+
"llm.model.layers.20.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 162 |
+
"llm.model.layers.21.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 163 |
+
"llm.model.layers.21.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 164 |
+
"llm.model.layers.21.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 165 |
+
"llm.model.layers.21.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 166 |
+
"llm.model.layers.21.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 167 |
+
"llm.model.layers.21.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 168 |
+
"llm.model.layers.21.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 169 |
+
"llm.model.layers.21.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 170 |
+
"llm.model.layers.21.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 171 |
+
"llm.model.layers.21.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 172 |
+
"llm.model.layers.21.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 173 |
+
"llm.model.layers.22.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 174 |
+
"llm.model.layers.22.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 175 |
+
"llm.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 176 |
+
"llm.model.layers.22.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 177 |
+
"llm.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 178 |
+
"llm.model.layers.22.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 179 |
+
"llm.model.layers.22.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 180 |
+
"llm.model.layers.22.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 181 |
+
"llm.model.layers.22.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 182 |
+
"llm.model.layers.22.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 183 |
+
"llm.model.layers.22.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 184 |
+
"llm.model.layers.23.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 185 |
+
"llm.model.layers.23.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 186 |
+
"llm.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 187 |
+
"llm.model.layers.23.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 188 |
+
"llm.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 189 |
+
"llm.model.layers.23.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 190 |
+
"llm.model.layers.23.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 191 |
+
"llm.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 192 |
+
"llm.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 193 |
+
"llm.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 194 |
+
"llm.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 195 |
+
"llm.model.layers.24.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 196 |
+
"llm.model.layers.24.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 197 |
+
"llm.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 198 |
+
"llm.model.layers.24.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 199 |
+
"llm.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 200 |
+
"llm.model.layers.24.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 201 |
+
"llm.model.layers.24.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 202 |
+
"llm.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 203 |
+
"llm.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 204 |
+
"llm.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 205 |
+
"llm.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 206 |
+
"llm.model.layers.25.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 207 |
+
"llm.model.layers.25.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 208 |
+
"llm.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 209 |
+
"llm.model.layers.25.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 210 |
+
"llm.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 211 |
+
"llm.model.layers.25.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 212 |
+
"llm.model.layers.25.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 213 |
+
"llm.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 214 |
+
"llm.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 215 |
+
"llm.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 216 |
+
"llm.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 217 |
+
"llm.model.layers.26.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 218 |
+
"llm.model.layers.26.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 219 |
+
"llm.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 220 |
+
"llm.model.layers.26.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 221 |
+
"llm.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 222 |
+
"llm.model.layers.26.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 223 |
+
"llm.model.layers.26.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 224 |
+
"llm.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 225 |
+
"llm.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 226 |
+
"llm.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 227 |
+
"llm.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 228 |
+
"llm.model.layers.27.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 229 |
+
"llm.model.layers.27.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 230 |
+
"llm.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 231 |
+
"llm.model.layers.27.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 232 |
+
"llm.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 233 |
+
"llm.model.layers.27.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 234 |
+
"llm.model.layers.27.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 235 |
+
"llm.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 236 |
+
"llm.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 237 |
+
"llm.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 238 |
+
"llm.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 239 |
+
"llm.model.layers.28.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 240 |
+
"llm.model.layers.28.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 241 |
+
"llm.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 242 |
+
"llm.model.layers.28.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 243 |
+
"llm.model.layers.28.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 244 |
+
"llm.model.layers.28.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 245 |
+
"llm.model.layers.28.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 246 |
+
"llm.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 247 |
+
"llm.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 248 |
+
"llm.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 249 |
+
"llm.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 250 |
+
"llm.model.layers.29.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 251 |
+
"llm.model.layers.29.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 252 |
+
"llm.model.layers.29.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 253 |
+
"llm.model.layers.29.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 254 |
+
"llm.model.layers.29.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 255 |
+
"llm.model.layers.29.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 256 |
+
"llm.model.layers.29.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 257 |
+
"llm.model.layers.29.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 258 |
+
"llm.model.layers.29.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 259 |
+
"llm.model.layers.29.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 260 |
+
"llm.model.layers.29.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 261 |
+
"llm.model.layers.3.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 262 |
+
"llm.model.layers.3.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 263 |
+
"llm.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 264 |
+
"llm.model.layers.3.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 265 |
+
"llm.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 266 |
+
"llm.model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 267 |
+
"llm.model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 268 |
+
"llm.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 269 |
+
"llm.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 270 |
+
"llm.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 271 |
+
"llm.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 272 |
+
"llm.model.layers.30.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 273 |
+
"llm.model.layers.30.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 274 |
+
"llm.model.layers.30.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 275 |
+
"llm.model.layers.30.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 276 |
+
"llm.model.layers.30.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 277 |
+
"llm.model.layers.30.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 278 |
+
"llm.model.layers.30.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 279 |
+
"llm.model.layers.30.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 280 |
+
"llm.model.layers.30.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 281 |
+
"llm.model.layers.30.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 282 |
+
"llm.model.layers.30.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 283 |
+
"llm.model.layers.31.input_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 284 |
+
"llm.model.layers.31.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
|
| 285 |
+
"llm.model.layers.31.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 286 |
+
"llm.model.layers.31.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 287 |
+
"llm.model.layers.31.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 288 |
+
"llm.model.layers.31.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 289 |
+
"llm.model.layers.31.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
|
| 290 |
+
"llm.model.layers.31.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 291 |
+
"llm.model.layers.31.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 292 |
+
"llm.model.layers.31.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 293 |
+
"llm.model.layers.31.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 294 |
+
"llm.model.layers.32.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 295 |
+
"llm.model.layers.32.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 296 |
+
"llm.model.layers.32.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
|
| 297 |
+
"llm.model.layers.32.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
|
| 298 |
+
"llm.model.layers.32.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 299 |
+
"llm.model.layers.32.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 300 |
+
"llm.model.layers.32.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 301 |
+
"llm.model.layers.32.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
|
| 302 |
+
"llm.model.layers.32.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
|
| 303 |
+
"llm.model.layers.32.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
|
| 304 |
+
"llm.model.layers.32.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
|
| 305 |
+
"llm.model.layers.33.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 306 |
+
"llm.model.layers.33.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 307 |
+
"llm.model.layers.33.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 308 |
+
"llm.model.layers.33.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 309 |
+
"llm.model.layers.33.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 310 |
+
"llm.model.layers.33.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 311 |
+
"llm.model.layers.33.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 312 |
+
"llm.model.layers.33.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 313 |
+
"llm.model.layers.33.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 314 |
+
"llm.model.layers.33.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 315 |
+
"llm.model.layers.33.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 316 |
+
"llm.model.layers.34.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 317 |
+
"llm.model.layers.34.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 318 |
+
"llm.model.layers.34.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 319 |
+
"llm.model.layers.34.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 320 |
+
"llm.model.layers.34.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 321 |
+
"llm.model.layers.34.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 322 |
+
"llm.model.layers.34.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 323 |
+
"llm.model.layers.34.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 324 |
+
"llm.model.layers.34.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 325 |
+
"llm.model.layers.34.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 326 |
+
"llm.model.layers.34.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 327 |
+
"llm.model.layers.35.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 328 |
+
"llm.model.layers.35.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 329 |
+
"llm.model.layers.35.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 330 |
+
"llm.model.layers.35.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 331 |
+
"llm.model.layers.35.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 332 |
+
"llm.model.layers.35.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 333 |
+
"llm.model.layers.35.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 334 |
+
"llm.model.layers.35.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 335 |
+
"llm.model.layers.35.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 336 |
+
"llm.model.layers.35.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 337 |
+
"llm.model.layers.35.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 338 |
+
"llm.model.layers.36.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 339 |
+
"llm.model.layers.36.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 340 |
+
"llm.model.layers.36.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 341 |
+
"llm.model.layers.36.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 342 |
+
"llm.model.layers.36.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 343 |
+
"llm.model.layers.36.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 344 |
+
"llm.model.layers.36.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 345 |
+
"llm.model.layers.36.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 346 |
+
"llm.model.layers.36.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 347 |
+
"llm.model.layers.36.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 348 |
+
"llm.model.layers.36.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 349 |
+
"llm.model.layers.37.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 350 |
+
"llm.model.layers.37.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 351 |
+
"llm.model.layers.37.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 352 |
+
"llm.model.layers.37.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 353 |
+
"llm.model.layers.37.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 354 |
+
"llm.model.layers.37.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 355 |
+
"llm.model.layers.37.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 356 |
+
"llm.model.layers.37.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 357 |
+
"llm.model.layers.37.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 358 |
+
"llm.model.layers.37.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 359 |
+
"llm.model.layers.37.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 360 |
+
"llm.model.layers.38.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 361 |
+
"llm.model.layers.38.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 362 |
+
"llm.model.layers.38.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 363 |
+
"llm.model.layers.38.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 364 |
+
"llm.model.layers.38.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 365 |
+
"llm.model.layers.38.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 366 |
+
"llm.model.layers.38.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 367 |
+
"llm.model.layers.38.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 368 |
+
"llm.model.layers.38.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 369 |
+
"llm.model.layers.38.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 370 |
+
"llm.model.layers.38.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 371 |
+
"llm.model.layers.39.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 372 |
+
"llm.model.layers.39.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 373 |
+
"llm.model.layers.39.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 374 |
+
"llm.model.layers.39.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 375 |
+
"llm.model.layers.39.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 376 |
+
"llm.model.layers.39.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 377 |
+
"llm.model.layers.39.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 378 |
+
"llm.model.layers.39.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 379 |
+
"llm.model.layers.39.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 380 |
+
"llm.model.layers.39.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 381 |
+
"llm.model.layers.39.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 382 |
+
"llm.model.layers.4.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 383 |
+
"llm.model.layers.4.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 384 |
+
"llm.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 385 |
+
"llm.model.layers.4.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 386 |
+
"llm.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 387 |
+
"llm.model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 388 |
+
"llm.model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 389 |
+
"llm.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 390 |
+
"llm.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 391 |
+
"llm.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 392 |
+
"llm.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 393 |
+
"llm.model.layers.40.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 394 |
+
"llm.model.layers.40.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 395 |
+
"llm.model.layers.40.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 396 |
+
"llm.model.layers.40.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 397 |
+
"llm.model.layers.40.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 398 |
+
"llm.model.layers.40.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 399 |
+
"llm.model.layers.40.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 400 |
+
"llm.model.layers.40.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 401 |
+
"llm.model.layers.40.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 402 |
+
"llm.model.layers.40.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 403 |
+
"llm.model.layers.40.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 404 |
+
"llm.model.layers.41.input_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 405 |
+
"llm.model.layers.41.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
|
| 406 |
+
"llm.model.layers.41.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
|
| 407 |
+
"llm.model.layers.41.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
|
| 408 |
+
"llm.model.layers.41.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 409 |
+
"llm.model.layers.41.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 410 |
+
"llm.model.layers.41.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
|
| 411 |
+
"llm.model.layers.41.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
|
| 412 |
+
"llm.model.layers.41.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
|
| 413 |
+
"llm.model.layers.41.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
|
| 414 |
+
"llm.model.layers.41.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
|
| 415 |
+
"llm.model.layers.5.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 416 |
+
"llm.model.layers.5.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 417 |
+
"llm.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 418 |
+
"llm.model.layers.5.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 419 |
+
"llm.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 420 |
+
"llm.model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 421 |
+
"llm.model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 422 |
+
"llm.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 423 |
+
"llm.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 424 |
+
"llm.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 425 |
+
"llm.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 426 |
+
"llm.model.layers.6.input_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 427 |
+
"llm.model.layers.6.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
|
| 428 |
+
"llm.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 429 |
+
"llm.model.layers.6.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 430 |
+
"llm.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 431 |
+
"llm.model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 432 |
+
"llm.model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
|
| 433 |
+
"llm.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 434 |
+
"llm.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 435 |
+
"llm.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 436 |
+
"llm.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 437 |
+
"llm.model.layers.7.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 438 |
+
"llm.model.layers.7.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 439 |
+
"llm.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
|
| 440 |
+
"llm.model.layers.7.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
|
| 441 |
+
"llm.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 442 |
+
"llm.model.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 443 |
+
"llm.model.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 444 |
+
"llm.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
|
| 445 |
+
"llm.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
|
| 446 |
+
"llm.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
|
| 447 |
+
"llm.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
|
| 448 |
+
"llm.model.layers.8.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 449 |
+
"llm.model.layers.8.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 450 |
+
"llm.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 451 |
+
"llm.model.layers.8.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 452 |
+
"llm.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 453 |
+
"llm.model.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 454 |
+
"llm.model.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 455 |
+
"llm.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 456 |
+
"llm.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 457 |
+
"llm.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 458 |
+
"llm.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 459 |
+
"llm.model.layers.9.input_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 460 |
+
"llm.model.layers.9.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
|
| 461 |
+
"llm.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
|
| 462 |
+
"llm.model.layers.9.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
|
| 463 |
+
"llm.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 464 |
+
"llm.model.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 465 |
+
"llm.model.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
|
| 466 |
+
"llm.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
|
| 467 |
+
"llm.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
|
| 468 |
+
"llm.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
|
| 469 |
+
"llm.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
|
| 470 |
+
"llm.model.norm.weight": "model-00004-of-00005.safetensors",
|
| 471 |
+
"visual_tokenizer.backbone.vision_model.embeddings.patch_embedding.bias": "model-00005-of-00005.safetensors",
|
| 472 |
+
"visual_tokenizer.backbone.vision_model.embeddings.patch_embedding.weight": "model-00005-of-00005.safetensors",
|
| 473 |
+
"visual_tokenizer.backbone.vision_model.embeddings.position_embedding.weight": "model-00005-of-00005.safetensors",
|
| 474 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 475 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 476 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 477 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 478 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 479 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 480 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 481 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 482 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 483 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 484 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 485 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 486 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 487 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 488 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 489 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 490 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 491 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 492 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 493 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 494 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 495 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 496 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 497 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 498 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 499 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 500 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 501 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 502 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 503 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 504 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 505 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 506 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 507 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 508 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 509 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 510 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 511 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 512 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 513 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 514 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 515 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 516 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 517 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 518 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 519 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 520 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 521 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 522 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 523 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 524 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 525 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 526 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 527 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 528 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 529 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 530 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 531 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 532 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 533 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 534 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 535 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 536 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 537 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 538 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 539 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 540 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 541 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 542 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 543 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 544 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 545 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 546 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 547 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 548 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 549 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 550 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 551 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 552 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 553 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 554 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 555 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 556 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 557 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 558 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 559 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 560 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 561 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 562 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 563 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 564 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 565 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 566 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 567 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 568 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 569 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 570 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 571 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 572 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 573 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 574 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 575 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 576 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 577 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 578 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 579 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 580 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 581 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 582 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 583 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 584 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 585 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 586 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 587 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 588 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 589 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 590 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 591 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 592 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 593 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 594 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 595 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 596 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 597 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 598 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 599 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 600 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 601 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 602 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 603 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 604 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 605 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 606 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 607 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 608 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 609 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 610 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 611 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 612 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 613 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 614 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 615 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 616 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 617 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 618 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 619 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 620 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 621 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 622 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 623 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 624 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 625 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 626 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 627 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 628 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 629 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 630 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 631 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 632 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 633 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 634 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 635 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 636 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 637 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 638 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 639 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 640 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 641 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 642 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 643 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 644 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 645 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 646 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 647 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 648 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 649 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 650 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 651 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 652 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 653 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 654 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 655 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 656 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 657 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 658 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 659 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 660 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 661 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 662 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 663 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 664 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 665 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 666 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 667 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 668 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 669 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 670 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 671 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 672 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 673 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 674 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 675 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 676 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 677 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 678 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 679 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 680 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 681 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 682 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 683 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 684 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 685 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 686 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 687 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 688 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 689 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 690 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 691 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 692 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 693 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 694 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 695 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 696 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 697 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 698 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 699 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 700 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 701 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 702 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 703 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 704 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 705 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 706 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 707 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 708 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 709 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 710 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 711 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 712 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 713 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 714 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 715 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 716 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 717 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 718 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 719 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 720 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 721 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 722 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 723 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 724 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 725 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 726 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 727 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 728 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 729 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 730 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 731 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 732 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 733 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 734 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 735 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 736 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 737 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 738 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 739 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 740 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 741 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 742 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 743 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 744 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 745 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 746 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 747 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 748 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 749 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 750 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 751 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 752 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 753 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 754 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 755 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 756 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 757 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 758 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 759 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 760 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 761 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.24.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 762 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 763 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 764 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 765 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 766 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 767 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 768 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 769 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 770 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 771 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 772 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 773 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 774 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 775 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 776 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 777 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.25.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 778 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 779 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 780 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 781 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 782 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 783 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 784 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 785 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 786 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 787 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 788 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 789 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 790 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 791 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 792 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 793 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.26.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 794 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 795 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 796 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 797 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 798 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 799 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 800 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 801 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 802 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 803 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 804 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 805 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 806 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 807 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 808 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 809 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 810 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 811 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 812 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 813 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 814 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 815 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 816 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 817 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 818 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 819 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 820 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 821 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 822 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 823 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 824 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 825 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 826 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 827 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 828 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 829 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 830 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 831 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 832 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 833 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 834 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 835 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 836 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 837 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 838 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 839 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 840 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 841 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 842 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 843 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 844 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 845 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 846 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 847 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 848 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 849 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 850 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 851 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 852 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 853 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 854 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 855 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 856 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 857 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 858 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 859 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 860 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 861 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 862 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 863 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 864 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 865 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 866 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 867 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 868 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 869 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 870 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 871 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 872 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 873 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 874 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 875 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 876 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 877 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 878 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 879 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 880 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 881 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 882 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 883 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 884 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 885 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 886 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 887 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 888 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 889 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 890 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.layer_norm1.bias": "model-00005-of-00005.safetensors",
|
| 891 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.layer_norm1.weight": "model-00005-of-00005.safetensors",
|
| 892 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.layer_norm2.bias": "model-00005-of-00005.safetensors",
|
| 893 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.layer_norm2.weight": "model-00005-of-00005.safetensors",
|
| 894 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 895 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 896 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 897 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 898 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00005-of-00005.safetensors",
|
| 899 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
|
| 900 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 901 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 902 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00005-of-00005.safetensors",
|
| 903 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
|
| 904 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00005-of-00005.safetensors",
|
| 905 |
+
"visual_tokenizer.backbone.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
|
| 906 |
+
"visual_tokenizer.backbone.vision_model.head.attention.in_proj_bias": "model-00005-of-00005.safetensors",
|
| 907 |
+
"visual_tokenizer.backbone.vision_model.head.attention.in_proj_weight": "model-00005-of-00005.safetensors",
|
| 908 |
+
"visual_tokenizer.backbone.vision_model.head.attention.out_proj.bias": "model-00005-of-00005.safetensors",
|
| 909 |
+
"visual_tokenizer.backbone.vision_model.head.attention.out_proj.weight": "model-00005-of-00005.safetensors",
|
| 910 |
+
"visual_tokenizer.backbone.vision_model.head.layernorm.bias": "model-00005-of-00005.safetensors",
|
| 911 |
+
"visual_tokenizer.backbone.vision_model.head.layernorm.weight": "model-00005-of-00005.safetensors",
|
| 912 |
+
"visual_tokenizer.backbone.vision_model.head.mlp.fc1.bias": "model-00005-of-00005.safetensors",
|
| 913 |
+
"visual_tokenizer.backbone.vision_model.head.mlp.fc1.weight": "model-00005-of-00005.safetensors",
|
| 914 |
+
"visual_tokenizer.backbone.vision_model.head.mlp.fc2.bias": "model-00005-of-00005.safetensors",
|
| 915 |
+
"visual_tokenizer.backbone.vision_model.head.mlp.fc2.weight": "model-00005-of-00005.safetensors",
|
| 916 |
+
"visual_tokenizer.backbone.vision_model.head.probe": "model-00005-of-00005.safetensors",
|
| 917 |
+
"visual_tokenizer.backbone.vision_model.post_layernorm.bias": "model-00005-of-00005.safetensors",
|
| 918 |
+
"visual_tokenizer.backbone.vision_model.post_layernorm.weight": "model-00005-of-00005.safetensors",
|
| 919 |
+
"visual_tokenizer.head.0.weight": "model-00005-of-00005.safetensors",
|
| 920 |
+
"visual_tokenizer.head.1.bias": "model-00005-of-00005.safetensors",
|
| 921 |
+
"visual_tokenizer.head.1.weight": "model-00005-of-00005.safetensors",
|
| 922 |
+
"vte.weight": "model-00005-of-00005.safetensors"
|
| 923 |
+
}
|
| 924 |
+
}
|
modeling_ovis.py
ADDED
|
@@ -0,0 +1,691 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
from importlib import import_module
|
| 3 |
+
from typing import List, Callable, Union, Optional
|
| 4 |
+
|
| 5 |
+
import PIL.Image
|
| 6 |
+
import torch
|
| 7 |
+
import torch.nn.functional as F
|
| 8 |
+
from torch import LongTensor, IntTensor, Tensor
|
| 9 |
+
from transformers import CLIPImageProcessor, CLIPVisionModel, SiglipImageProcessor, SiglipVisionModel
|
| 10 |
+
from transformers import PreTrainedModel, AutoModel, AutoTokenizer, AutoModelForCausalLM, AutoImageProcessor
|
| 11 |
+
from transformers.generation.utils import GenerateOutput
|
| 12 |
+
from transformers.cache_utils import HybridCache
|
| 13 |
+
|
| 14 |
+
from .configuration_ovis import BaseVisualTokenizerConfig, ClipVisualTokenizerConfig, SiglipVisualTokenizerConfig
|
| 15 |
+
from .configuration_ovis import OvisConfig, ConversationFormatter, IGNORE_INDEX, IMAGE_TOKEN_INDEX
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
# ----------------------------------------------------------------------
|
| 19 |
+
# Visual Tokenizer
|
| 20 |
+
# ----------------------------------------------------------------------
|
| 21 |
+
class BaseVisualTokenizer(PreTrainedModel):
|
| 22 |
+
base_model_prefix = "backbone"
|
| 23 |
+
main_input_name = None
|
| 24 |
+
_image_processor_class = None
|
| 25 |
+
_image_processor_kwargs = {}
|
| 26 |
+
_backbone_class = None
|
| 27 |
+
_backbone_name_or_path = None
|
| 28 |
+
|
| 29 |
+
def __init__(self, config: BaseVisualTokenizerConfig, *inputs, **kwargs):
|
| 30 |
+
super().__init__(config, *inputs, **kwargs)
|
| 31 |
+
if kwargs.get('train_from_scratch'):
|
| 32 |
+
self.image_processor = self._image_processor_class.from_pretrained(
|
| 33 |
+
self._backbone_name_or_path, **self._image_processor_kwargs)
|
| 34 |
+
self.backbone = self._backbone_class.from_pretrained(
|
| 35 |
+
self._backbone_name_or_path, **self.config.backbone_kwargs)
|
| 36 |
+
self.config.backbone_config = self.backbone.config
|
| 37 |
+
else:
|
| 38 |
+
self.image_processor = AutoImageProcessor.from_pretrained(
|
| 39 |
+
kwargs['image_processor_name_or_path'])
|
| 40 |
+
self.backbone = AutoModel.from_config(self.config.backbone_config)
|
| 41 |
+
self.head = None
|
| 42 |
+
|
| 43 |
+
assert all((self.image_processor.do_resize,
|
| 44 |
+
not getattr(self.image_processor, 'do_center_crop', False),
|
| 45 |
+
self.image_processor.do_rescale,
|
| 46 |
+
self.image_processor.do_normalize
|
| 47 |
+
)), f"image_processor `{self.image_processor}` is not supported currently"
|
| 48 |
+
|
| 49 |
+
def get_backbone(self):
|
| 50 |
+
return self.backbone
|
| 51 |
+
|
| 52 |
+
def get_image_processor(self):
|
| 53 |
+
return self.image_processor
|
| 54 |
+
|
| 55 |
+
def get_zero_pixel_values(self, n=1):
|
| 56 |
+
height, width = self.get_image_size()
|
| 57 |
+
if self.config.hd_booster is None:
|
| 58 |
+
return torch.zeros(n, 3, height, width)
|
| 59 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 60 |
+
return torch.zeros(n, 3 * 5, height, width)
|
| 61 |
+
else:
|
| 62 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 63 |
+
|
| 64 |
+
def get_head(self):
|
| 65 |
+
return self.head
|
| 66 |
+
|
| 67 |
+
def get_image_size(self):
|
| 68 |
+
raise NotImplementedError
|
| 69 |
+
|
| 70 |
+
def preprocess_image(self, image: PIL.Image.Image, convert_to_rgb=True):
|
| 71 |
+
def _preprocess(img: PIL.Image.Image):
|
| 72 |
+
# first resize and preprocess
|
| 73 |
+
sides = self.get_image_size()
|
| 74 |
+
if sides[0] != sides[1]:
|
| 75 |
+
raise ValueError('get_image_size() returns non-square size')
|
| 76 |
+
side = sides[0]
|
| 77 |
+
|
| 78 |
+
w, h = img.size
|
| 79 |
+
if w == h:
|
| 80 |
+
new_width = new_height = side
|
| 81 |
+
elif w > h:
|
| 82 |
+
new_width = side
|
| 83 |
+
new_height = int(h / w * new_width)
|
| 84 |
+
else:
|
| 85 |
+
new_height = side
|
| 86 |
+
new_width = int(w / h * new_height)
|
| 87 |
+
new_size = dict(height=new_height, width=new_width)
|
| 88 |
+
pixel_values = self.image_processor.preprocess(
|
| 89 |
+
img, size=new_size, return_tensors='pt')['pixel_values']
|
| 90 |
+
|
| 91 |
+
# then pad to square
|
| 92 |
+
square_values = torch.zeros(
|
| 93 |
+
[1, 3, side, side], dtype=pixel_values.dtype, device=pixel_values.device)
|
| 94 |
+
new_height, new_width = pixel_values.shape[2:]
|
| 95 |
+
if new_height == new_width:
|
| 96 |
+
square_values[:, :, :, :] = pixel_values
|
| 97 |
+
elif new_height > new_width:
|
| 98 |
+
from_index = (side - new_width) // 2
|
| 99 |
+
square_values[:, :, :, from_index:from_index + new_width] = pixel_values
|
| 100 |
+
else:
|
| 101 |
+
from_index = (side - new_height) // 2
|
| 102 |
+
square_values[:, :, from_index:from_index + new_height, :] = pixel_values
|
| 103 |
+
|
| 104 |
+
return square_values
|
| 105 |
+
|
| 106 |
+
if convert_to_rgb and image.mode != 'RGB':
|
| 107 |
+
image = image.convert('RGB')
|
| 108 |
+
|
| 109 |
+
if self.config.hd_booster is None:
|
| 110 |
+
return _preprocess(image) # [1, 3, side, side]
|
| 111 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 112 |
+
width, height = image.size
|
| 113 |
+
is_low_resolution = (height < self.get_image_size()[0] * 1.5 or
|
| 114 |
+
width < self.get_image_size()[1] * 1.5)
|
| 115 |
+
if self.config.hd_booster == 's2wrapper-adaptive' and is_low_resolution:
|
| 116 |
+
values = self.get_zero_pixel_values() + torch.inf
|
| 117 |
+
values[0][:3] = _preprocess(image)[0]
|
| 118 |
+
else:
|
| 119 |
+
center_x, center_y = width // 2, height // 2
|
| 120 |
+
image_top_left = image.crop((0, 0, center_x, center_y))
|
| 121 |
+
image_top_right = image.crop((center_x, 0, width, center_y))
|
| 122 |
+
image_bottom_left = image.crop((0, center_y, center_x, height))
|
| 123 |
+
image_bottom_right = image.crop((center_x, center_y, width, height))
|
| 124 |
+
imgs = [image, image_top_left, image_top_right, image_bottom_left, image_bottom_right]
|
| 125 |
+
values = torch.cat([_preprocess(img) for img in imgs], dim=1)
|
| 126 |
+
return values # [1, 3*5, side, side]
|
| 127 |
+
else:
|
| 128 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 129 |
+
|
| 130 |
+
def get_backbone_layer(self, index):
|
| 131 |
+
return self.backbone.vision_model.encoder.layers[index]
|
| 132 |
+
|
| 133 |
+
def tokenize(self, logits):
|
| 134 |
+
def st_argmax(y_soft, dim): # straight-through softmax
|
| 135 |
+
index = y_soft.max(dim, keepdim=True)[1]
|
| 136 |
+
y_hard = torch.zeros_like(
|
| 137 |
+
y_soft, memory_format=torch.legacy_contiguous_format).scatter_(dim, index, 1.0)
|
| 138 |
+
ret = y_hard - y_soft.detach() + y_soft
|
| 139 |
+
return ret
|
| 140 |
+
|
| 141 |
+
if self.config.tokenize_function == 'softmax':
|
| 142 |
+
tokens = F.softmax(logits, dim=-1)
|
| 143 |
+
elif self.config.tokenize_function == 'gumbel_argmax':
|
| 144 |
+
tokens = F.gumbel_softmax(logits, tau=self.config.tau, hard=True)
|
| 145 |
+
elif self.config.tokenize_function == 'st_argmax':
|
| 146 |
+
tokens = st_argmax(logits, dim=-1)
|
| 147 |
+
else:
|
| 148 |
+
raise ValueError(
|
| 149 |
+
f'Invalid `max_type`, expected softmax or gumbel_argmax or st_argmax,'
|
| 150 |
+
f' but got {self.config.tokenize_function}')
|
| 151 |
+
return tokens
|
| 152 |
+
|
| 153 |
+
|
| 154 |
+
class ClipVisualTokenizer(BaseVisualTokenizer):
|
| 155 |
+
config_class = ClipVisualTokenizerConfig
|
| 156 |
+
supports_gradient_checkpointing = True
|
| 157 |
+
_no_split_modules = ["CLIPEncoderLayer"]
|
| 158 |
+
_image_processor_class = CLIPImageProcessor
|
| 159 |
+
_image_processor_kwargs = dict(do_center_crop=False)
|
| 160 |
+
_backbone_class = CLIPVisionModel
|
| 161 |
+
_backbone_name_or_path = "openai/clip-vit-large-patch14-336"
|
| 162 |
+
|
| 163 |
+
def __init__(self, config: ClipVisualTokenizerConfig = None, *inputs, **kwargs):
|
| 164 |
+
super().__init__(config, *inputs, **kwargs)
|
| 165 |
+
head_dim = self.config.vocab_size
|
| 166 |
+
if self.config.use_indicators:
|
| 167 |
+
head_dim -= 2 # reserved for two image indicator tokens
|
| 168 |
+
if self.config.hd_booster is None:
|
| 169 |
+
self.head = torch.nn.Sequential(
|
| 170 |
+
torch.nn.Linear(self.backbone.config.hidden_size, head_dim, bias=False),
|
| 171 |
+
torch.nn.LayerNorm(head_dim)
|
| 172 |
+
)
|
| 173 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 174 |
+
self.head = torch.nn.Sequential(
|
| 175 |
+
torch.nn.Linear(self.backbone.config.hidden_size * 2, head_dim, bias=False),
|
| 176 |
+
torch.nn.LayerNorm(head_dim)
|
| 177 |
+
)
|
| 178 |
+
else:
|
| 179 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 180 |
+
|
| 181 |
+
def get_image_size(self):
|
| 182 |
+
height = self.image_processor.crop_size["height"]
|
| 183 |
+
width = self.image_processor.crop_size["width"]
|
| 184 |
+
return height, width
|
| 185 |
+
|
| 186 |
+
def encode(self, pixel_values):
|
| 187 |
+
if self.config.hd_booster is None:
|
| 188 |
+
output = self.backbone(pixel_values, output_hidden_states=True, return_dict=True)
|
| 189 |
+
features = output.hidden_states[-1]
|
| 190 |
+
if self.config.drop_cls_token:
|
| 191 |
+
features = features[:, 1:, :]
|
| 192 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 193 |
+
n, c, side, _ = pixel_values.shape
|
| 194 |
+
if self.config.hd_booster == 's2wrapper-adaptive':
|
| 195 |
+
pixel_values_mask = torch.isinf(pixel_values) # [n, c, side, side]
|
| 196 |
+
pixel_values = torch.masked_fill(pixel_values, pixel_values_mask, 0.0)
|
| 197 |
+
pixel_values = pixel_values.reshape(n * 5, c // 5, side, side)
|
| 198 |
+
output = self.backbone(pixel_values, output_hidden_states=True, return_dict=True)
|
| 199 |
+
features = output.hidden_states[-1]
|
| 200 |
+
if self.config.drop_cls_token:
|
| 201 |
+
features = features[:, 1:, :]
|
| 202 |
+
_, l, d = features.shape
|
| 203 |
+
features = features.reshape(n, 5, l, d)
|
| 204 |
+
features_overall = features[:, 0, :, :] # [n, l, d]
|
| 205 |
+
features_parts = features[:, 1:, :, :] # [n, 4, l, d]
|
| 206 |
+
sqrt_l = int(l ** 0.5)
|
| 207 |
+
assert sqrt_l ** 2 == l, "The token sequence length should be a perfect square."
|
| 208 |
+
features_parts = features_parts.reshape(n, 4, sqrt_l, sqrt_l, d) # [n, 4, sqrt(l), sqrt(l), d]
|
| 209 |
+
features_top = torch.concat(
|
| 210 |
+
[features_parts[:, 0, :, :, :], features_parts[:, 1, :, :, :]], dim=-2) # [n, sqrt(l), sqrt(l)*2, d]
|
| 211 |
+
features_bottom = torch.concat(
|
| 212 |
+
[features_parts[:, 2, :, :, :], features_parts[:, 3, :, :, :]], dim=-2) # [n, sqrt(l), sqrt(l)*2, d]
|
| 213 |
+
features_merge = torch.concat([features_top, features_bottom], dim=-3) # [n, sqrt(l)*2, sqrt(l)*2, d]
|
| 214 |
+
features_pool = F.interpolate(
|
| 215 |
+
features_merge.permute(0, 3, 1, 2).to(torch.float32),
|
| 216 |
+
size=sqrt_l,
|
| 217 |
+
mode='area'
|
| 218 |
+
) # [n, d, sqrt_l, sqrt_l]
|
| 219 |
+
features_pool = features_pool.flatten(2).permute(0, 2, 1).to(features.dtype) # [n, l, d]
|
| 220 |
+
if self.config.hd_booster == 's2wrapper-adaptive':
|
| 221 |
+
features_pool_mask = torch.unsqueeze(
|
| 222 |
+
torch.unsqueeze(pixel_values_mask[:, -1, -1, -1], dim=-1), dim=-1) # [n, 1, 1]
|
| 223 |
+
features_pool = torch.masked_fill(features_pool, features_pool_mask, 0.0)
|
| 224 |
+
features = torch.cat([features_overall, features_pool], dim=-1) # [n, l, 2*d]
|
| 225 |
+
else:
|
| 226 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 227 |
+
return features
|
| 228 |
+
|
| 229 |
+
def forward(self, pixel_values) -> Tensor: # [BatchSize, ImageShape] -> [BatchSize, #Token, VocabSize]
|
| 230 |
+
features = self.encode(pixel_values)
|
| 231 |
+
logits = self.head(features)
|
| 232 |
+
tokens = self.tokenize(logits)
|
| 233 |
+
if self.config.use_indicators:
|
| 234 |
+
# tokens' shape is [BatchSize, #Token, VocabSize-2], so padding with [BatchSize, #Token, 2],
|
| 235 |
+
# after which, tokens' shape should become [BatchSize, #Token, VocabSize]
|
| 236 |
+
batch_size, token_len, _ = tokens.shape
|
| 237 |
+
padding_tensor = torch.zeros(
|
| 238 |
+
size=(batch_size, token_len, 2),
|
| 239 |
+
dtype=tokens.dtype,
|
| 240 |
+
device=tokens.device,
|
| 241 |
+
layout=tokens.layout,
|
| 242 |
+
requires_grad=False
|
| 243 |
+
)
|
| 244 |
+
tokens = torch.cat((tokens, padding_tensor), dim=2)
|
| 245 |
+
|
| 246 |
+
# adding indicator tokens, after which tokens' shape should become [BatchSize, 1+#Token+1, VocabSize]
|
| 247 |
+
begin_indicator = torch.zeros(
|
| 248 |
+
size=(batch_size, 1),
|
| 249 |
+
dtype=torch.long,
|
| 250 |
+
device=tokens.device,
|
| 251 |
+
requires_grad=False
|
| 252 |
+
) + self.config.vocab_size - 2
|
| 253 |
+
begin_indicator_token = F.one_hot(
|
| 254 |
+
begin_indicator, num_classes=self.config.vocab_size).to(dtype=tokens.dtype)
|
| 255 |
+
end_indicator = torch.zeros(
|
| 256 |
+
size=(batch_size, 1),
|
| 257 |
+
dtype=torch.long,
|
| 258 |
+
device=tokens.device,
|
| 259 |
+
requires_grad=False
|
| 260 |
+
) + self.config.vocab_size - 1
|
| 261 |
+
end_indicator_token = F.one_hot(
|
| 262 |
+
end_indicator, num_classes=self.config.vocab_size).to(dtype=tokens.dtype)
|
| 263 |
+
tokens = torch.cat((begin_indicator_token, tokens, end_indicator_token), dim=1)
|
| 264 |
+
return tokens
|
| 265 |
+
|
| 266 |
+
|
| 267 |
+
class SiglipVisualTokenizer(BaseVisualTokenizer):
|
| 268 |
+
config_class = SiglipVisualTokenizerConfig
|
| 269 |
+
supports_gradient_checkpointing = True
|
| 270 |
+
_no_split_modules = ["SiglipVisionTransformer"]
|
| 271 |
+
_image_processor_class = SiglipImageProcessor
|
| 272 |
+
_image_processor_kwargs = {}
|
| 273 |
+
_backbone_class = SiglipVisionModel
|
| 274 |
+
_backbone_name_or_path = "google/siglip-so400m-patch14-384"
|
| 275 |
+
|
| 276 |
+
def __init__(self, config: SiglipVisualTokenizerConfig = None, *inputs, **kwargs):
|
| 277 |
+
super().__init__(config, *inputs, **kwargs)
|
| 278 |
+
head_dim = self.config.vocab_size
|
| 279 |
+
if self.config.use_indicators:
|
| 280 |
+
head_dim -= 2 # reserved for two image indicator tokens
|
| 281 |
+
if self.config.hd_booster is None:
|
| 282 |
+
self.head = torch.nn.Sequential(
|
| 283 |
+
torch.nn.Linear(
|
| 284 |
+
self.backbone.config.hidden_size * self.config.hidden_stride * self.config.hidden_stride,
|
| 285 |
+
head_dim,
|
| 286 |
+
bias=False
|
| 287 |
+
),
|
| 288 |
+
torch.nn.LayerNorm(head_dim)
|
| 289 |
+
)
|
| 290 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 291 |
+
self.head = torch.nn.Sequential(
|
| 292 |
+
torch.nn.Linear(
|
| 293 |
+
self.backbone.config.hidden_size * self.config.hidden_stride * self.config.hidden_stride * 2,
|
| 294 |
+
head_dim,
|
| 295 |
+
bias=False
|
| 296 |
+
),
|
| 297 |
+
torch.nn.LayerNorm(head_dim)
|
| 298 |
+
)
|
| 299 |
+
else:
|
| 300 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 301 |
+
|
| 302 |
+
def get_image_size(self):
|
| 303 |
+
height = self.image_processor.size["height"]
|
| 304 |
+
width = self.image_processor.size["width"]
|
| 305 |
+
return height, width
|
| 306 |
+
|
| 307 |
+
def encode(self, pixel_values):
|
| 308 |
+
if self.config.hd_booster is None:
|
| 309 |
+
output = self.backbone(pixel_values, output_hidden_states=True, return_dict=True)
|
| 310 |
+
features = output.hidden_states[-1]
|
| 311 |
+
if self.config.drop_cls_token:
|
| 312 |
+
features = features[:, 1:, :]
|
| 313 |
+
elif self.config.hd_booster in ['s2wrapper', 's2wrapper-adaptive']:
|
| 314 |
+
n, c, side, _ = pixel_values.shape
|
| 315 |
+
if self.config.hd_booster == 's2wrapper-adaptive':
|
| 316 |
+
pixel_values_mask = torch.isinf(pixel_values) # [n, c, side, side]
|
| 317 |
+
pixel_values = torch.masked_fill(pixel_values, pixel_values_mask, 0.0)
|
| 318 |
+
pixel_values = pixel_values.reshape(n * 5, c // 5, side, side)
|
| 319 |
+
output = self.backbone(pixel_values, output_hidden_states=True, return_dict=True)
|
| 320 |
+
features = output.hidden_states[-1]
|
| 321 |
+
if self.config.drop_cls_token:
|
| 322 |
+
features = features[:, 1:, :]
|
| 323 |
+
_, l, d = features.shape
|
| 324 |
+
features = features.reshape(n, 5, l, d)
|
| 325 |
+
features_overall = features[:, 0, :, :] # [n, l, d]
|
| 326 |
+
features_parts = features[:, 1:, :, :] # [n, 4, l, d]
|
| 327 |
+
sqrt_l = int(l ** 0.5)
|
| 328 |
+
assert sqrt_l ** 2 == l, "The token sequence length should be a perfect square."
|
| 329 |
+
features_parts = features_parts.reshape(n, 4, sqrt_l, sqrt_l, d) # [n, 4, sqrt(l), sqrt(l), d]
|
| 330 |
+
features_top = torch.concat(
|
| 331 |
+
[features_parts[:, 0, :, :, :], features_parts[:, 1, :, :, :]], dim=-2) # [n, sqrt(l), sqrt(l)*2, d]
|
| 332 |
+
features_bottom = torch.concat(
|
| 333 |
+
[features_parts[:, 2, :, :, :], features_parts[:, 3, :, :, :]], dim=-2) # [n, sqrt(l), sqrt(l)*2, d]
|
| 334 |
+
features_merge = torch.concat([features_top, features_bottom], dim=-3) # [n, sqrt(l)*2, sqrt(l)*2, d]
|
| 335 |
+
features_pool = F.interpolate(
|
| 336 |
+
features_merge.permute(0, 3, 1, 2).to(torch.float32),
|
| 337 |
+
size=sqrt_l,
|
| 338 |
+
mode='area'
|
| 339 |
+
) # [n, d, sqrt_l, sqrt_l]
|
| 340 |
+
features_pool = features_pool.flatten(2).permute(0, 2, 1).to(features.dtype) # [n, l, d]
|
| 341 |
+
if self.config.hd_booster == 's2wrapper-adaptive':
|
| 342 |
+
features_pool_mask = torch.unsqueeze(
|
| 343 |
+
torch.unsqueeze(pixel_values_mask[:, -1, -1, -1], dim=-1), dim=-1) # [n, 1, 1]
|
| 344 |
+
features_pool = torch.masked_fill(features_pool, features_pool_mask, 0.0)
|
| 345 |
+
features = torch.cat([features_overall, features_pool], dim=-1) # [n, l, 2*d]
|
| 346 |
+
else:
|
| 347 |
+
raise ValueError(f'Unsupported hd_booster {self.config.hd_booster}')
|
| 348 |
+
|
| 349 |
+
# merge number of `hidden_stride * hidden_stride` hidden states together to reduce token sequence length
|
| 350 |
+
# e.g., for hidden_stride=3, this leads to a token length reduction: 729 -> 81
|
| 351 |
+
if self.config.hidden_stride > 1:
|
| 352 |
+
n, l, d = features.shape # this `d` maybe different from the above `d
|
| 353 |
+
sqrt_l = int(l ** 0.5)
|
| 354 |
+
assert sqrt_l ** 2 == l, "The token sequence length should be a perfect square."
|
| 355 |
+
assert l % (self.config.hidden_stride ** 2) == 0, \
|
| 356 |
+
"The token sequence length should be divisible by `hidden_stride**2`."
|
| 357 |
+
features = features.reshape(n, sqrt_l, sqrt_l, d)
|
| 358 |
+
features = features.reshape(n, sqrt_l // self.config.hidden_stride, self.config.hidden_stride,
|
| 359 |
+
sqrt_l // self.config.hidden_stride, self.config.hidden_stride, d)
|
| 360 |
+
features = features.permute(0, 1, 3, 2, 4, 5) # [n, sqrt_l/hs, sqrt_l/hs, hs, hs, d]
|
| 361 |
+
features = features.flatten(3) # [n, sqrt_l/hs, sqrt_l/hs, hs*hs*d]
|
| 362 |
+
features = features.reshape(n, l // (self.config.hidden_stride * self.config.hidden_stride),
|
| 363 |
+
self.config.hidden_stride * self.config.hidden_stride * d)
|
| 364 |
+
|
| 365 |
+
return features
|
| 366 |
+
|
| 367 |
+
def forward(self, pixel_values) -> Tensor: # [BatchSize, ImageShape] -> [BatchSize, #Token, VocabSize]
|
| 368 |
+
features = self.encode(pixel_values)
|
| 369 |
+
logits = self.head(features)
|
| 370 |
+
tokens = self.tokenize(logits)
|
| 371 |
+
if self.config.use_indicators:
|
| 372 |
+
# tokens' shape is [BatchSize, #Token, VocabSize-2], so padding with [BatchSize, #Token, 2], after
|
| 373 |
+
# which, tokens' shape should become [BatchSize, #Token, VocabSize]
|
| 374 |
+
batch_size, token_len, _ = tokens.shape
|
| 375 |
+
padding_tensor = torch.zeros(
|
| 376 |
+
size=(batch_size, token_len, 2),
|
| 377 |
+
dtype=tokens.dtype,
|
| 378 |
+
device=tokens.device,
|
| 379 |
+
layout=tokens.layout,
|
| 380 |
+
requires_grad=False
|
| 381 |
+
)
|
| 382 |
+
tokens = torch.cat((tokens, padding_tensor), dim=2)
|
| 383 |
+
|
| 384 |
+
# adding indicator tokens, after which tokens' shape should become [BatchSize, 1+#Token+1, VocabSize]
|
| 385 |
+
begin_indicator = torch.zeros(
|
| 386 |
+
size=(batch_size, 1),
|
| 387 |
+
dtype=torch.long,
|
| 388 |
+
device=tokens.device,
|
| 389 |
+
requires_grad=False
|
| 390 |
+
) + self.config.vocab_size - 2
|
| 391 |
+
begin_indicator_token = F.one_hot(
|
| 392 |
+
begin_indicator, num_classes=self.config.vocab_size).to(dtype=tokens.dtype)
|
| 393 |
+
end_indicator = torch.zeros(
|
| 394 |
+
size=(batch_size, 1),
|
| 395 |
+
dtype=torch.long,
|
| 396 |
+
device=tokens.device,
|
| 397 |
+
requires_grad=False
|
| 398 |
+
) + self.config.vocab_size - 1
|
| 399 |
+
end_indicator_token = F.one_hot(
|
| 400 |
+
end_indicator, num_classes=self.config.vocab_size).to(dtype=tokens.dtype)
|
| 401 |
+
tokens = torch.cat((begin_indicator_token, tokens, end_indicator_token), dim=1)
|
| 402 |
+
return tokens
|
| 403 |
+
|
| 404 |
+
|
| 405 |
+
AutoModel.register(ClipVisualTokenizerConfig, ClipVisualTokenizer)
|
| 406 |
+
AutoModel.register(SiglipVisualTokenizerConfig, SiglipVisualTokenizer)
|
| 407 |
+
|
| 408 |
+
|
| 409 |
+
# ----------------------------------------------------------------------
|
| 410 |
+
# Ovis
|
| 411 |
+
# ----------------------------------------------------------------------
|
| 412 |
+
class VisualEmbedding(torch.nn.Embedding):
|
| 413 |
+
def forward(self, input: Tensor) -> Tensor:
|
| 414 |
+
if any((isinstance(input, LongTensor), isinstance(input, IntTensor))):
|
| 415 |
+
return super().forward(input)
|
| 416 |
+
return torch.matmul(input, self.weight)
|
| 417 |
+
|
| 418 |
+
|
| 419 |
+
class OvisPreTrainedModel(PreTrainedModel):
|
| 420 |
+
config_class = OvisConfig
|
| 421 |
+
base_model_prefix = "ovis"
|
| 422 |
+
|
| 423 |
+
|
| 424 |
+
class Ovis(OvisPreTrainedModel):
|
| 425 |
+
|
| 426 |
+
def __init__(self, config: OvisConfig, *inputs, **kwargs):
|
| 427 |
+
super().__init__(config, *inputs, **kwargs)
|
| 428 |
+
self.llm = AutoModelForCausalLM.from_config(self.config.llm_config, attn_implementation="sdpa")
|
| 429 |
+
assert self.config.hidden_size == self.llm.config.hidden_size, "hidden size mismatch"
|
| 430 |
+
self.text_tokenizer = AutoTokenizer.from_pretrained(self.config.name_or_path)
|
| 431 |
+
self.visual_tokenizer = AutoModel.from_config(
|
| 432 |
+
self.config.visual_tokenizer_config,
|
| 433 |
+
image_processor_name_or_path=self.config.name_or_path
|
| 434 |
+
)
|
| 435 |
+
self.vte = VisualEmbedding(
|
| 436 |
+
self.config.visual_tokenizer_config.vocab_size,
|
| 437 |
+
self.config.hidden_size,
|
| 438 |
+
device=self.visual_tokenizer.device,
|
| 439 |
+
dtype=self.visual_tokenizer.dtype
|
| 440 |
+
)
|
| 441 |
+
|
| 442 |
+
def _merge_modules(modules_list: tuple):
|
| 443 |
+
merged_modules = []
|
| 444 |
+
for modules in modules_list:
|
| 445 |
+
merged_modules.extend(modules if modules else [])
|
| 446 |
+
return merged_modules
|
| 447 |
+
|
| 448 |
+
self._no_split_modules = _merge_modules(
|
| 449 |
+
(self.llm._no_split_modules, self.visual_tokenizer._no_split_modules))
|
| 450 |
+
self._skip_keys_device_placement = self.llm._skip_keys_device_placement
|
| 451 |
+
self._keep_in_fp32_modules = _merge_modules(
|
| 452 |
+
(self.llm._keep_in_fp32_modules, self.visual_tokenizer._keep_in_fp32_modules))
|
| 453 |
+
self.is_parallelizable = all((self.llm.is_parallelizable, self.visual_tokenizer.is_parallelizable))
|
| 454 |
+
self.supports_gradient_checkpointing = all(
|
| 455 |
+
(self.llm.supports_gradient_checkpointing, self.visual_tokenizer.supports_gradient_checkpointing))
|
| 456 |
+
self._supports_flash_attn_2 = all(
|
| 457 |
+
(self.llm._supports_flash_attn_2, self.visual_tokenizer._supports_flash_attn_2))
|
| 458 |
+
self._supports_sdpa = all((self.llm._supports_sdpa, self.visual_tokenizer._supports_sdpa))
|
| 459 |
+
|
| 460 |
+
def get_text_tokenizer(self):
|
| 461 |
+
return self.text_tokenizer
|
| 462 |
+
|
| 463 |
+
def get_visual_tokenizer(self):
|
| 464 |
+
return self.visual_tokenizer
|
| 465 |
+
|
| 466 |
+
def get_llm(self):
|
| 467 |
+
return self.llm
|
| 468 |
+
|
| 469 |
+
def get_vte(self):
|
| 470 |
+
return self.vte
|
| 471 |
+
|
| 472 |
+
def get_wte(self):
|
| 473 |
+
return self.llm.get_input_embeddings()
|
| 474 |
+
|
| 475 |
+
def get_conversation_formatter(self) -> ConversationFormatter:
|
| 476 |
+
if getattr(self, 'conversation_formatter', None) is None:
|
| 477 |
+
self.conversation_formatter = getattr(
|
| 478 |
+
import_module(".configuration_ovis", __package__),
|
| 479 |
+
self.config.conversation_formatter_class
|
| 480 |
+
)(self.text_tokenizer)
|
| 481 |
+
return self.conversation_formatter
|
| 482 |
+
|
| 483 |
+
def forward(
|
| 484 |
+
self,
|
| 485 |
+
input_ids: torch.Tensor,
|
| 486 |
+
attention_mask: torch.Tensor,
|
| 487 |
+
labels: Optional[torch.Tensor],
|
| 488 |
+
pixel_values: List[Optional[torch.Tensor]],
|
| 489 |
+
**kwargs
|
| 490 |
+
):
|
| 491 |
+
assert self.training, "`forward` can only be used in training. For inference, use `generate`."
|
| 492 |
+
_, inputs_embeds, labels, attention_mask = self.merge_multimodal(
|
| 493 |
+
text_input_ids=input_ids,
|
| 494 |
+
text_attention_masks=attention_mask,
|
| 495 |
+
text_labels=labels,
|
| 496 |
+
pixel_values=pixel_values
|
| 497 |
+
)
|
| 498 |
+
return self.llm(inputs_embeds=inputs_embeds, labels=labels, attention_mask=attention_mask, **kwargs)
|
| 499 |
+
|
| 500 |
+
def merge_multimodal(
|
| 501 |
+
self,
|
| 502 |
+
text_input_ids: torch.Tensor,
|
| 503 |
+
text_attention_masks: torch.Tensor,
|
| 504 |
+
text_labels: Optional[torch.Tensor],
|
| 505 |
+
pixel_values: List[Optional[torch.Tensor]]
|
| 506 |
+
):
|
| 507 |
+
input_device = text_input_ids.device
|
| 508 |
+
if self.training:
|
| 509 |
+
# When training, to be compatible with deepspeed zero, each sample has to include pixel_value tensor.
|
| 510 |
+
# For text-only sample, one can simply use a full zero tensor as pixel_value, which will be ignored
|
| 511 |
+
# (see below in this function); so, the gradient will not be affected.
|
| 512 |
+
num_images = [x.shape[0] for x in pixel_values]
|
| 513 |
+
visual_tokens = self.visual_tokenizer(torch.cat([x for x in pixel_values], dim=0))
|
| 514 |
+
visual_embeds = torch.split(
|
| 515 |
+
self.get_vte()(visual_tokens).to(dtype=self.dtype, device=input_device),
|
| 516 |
+
split_size_or_sections=num_images,
|
| 517 |
+
dim=0
|
| 518 |
+
)
|
| 519 |
+
visual_input_ids = torch.split(
|
| 520 |
+
torch.argmax(visual_tokens, dim=-1).to(device=input_device),
|
| 521 |
+
split_size_or_sections=num_images,
|
| 522 |
+
dim=0
|
| 523 |
+
)
|
| 524 |
+
visual_labels = [
|
| 525 |
+
torch.full(
|
| 526 |
+
x.shape, IGNORE_INDEX, dtype=torch.long, device=input_device
|
| 527 |
+
) for x in visual_input_ids
|
| 528 |
+
]
|
| 529 |
+
else:
|
| 530 |
+
# When inference, sample can include only text with `None` pixel_value
|
| 531 |
+
num_images = [x.shape[0] if x is not None else 0 for x in pixel_values]
|
| 532 |
+
if sum(num_images) > 0:
|
| 533 |
+
visual_tokens = self.visual_tokenizer(torch.cat([x for x in pixel_values if x is not None], dim=0))
|
| 534 |
+
visual_embeds = torch.split(
|
| 535 |
+
self.get_vte()(visual_tokens).to(dtype=self.dtype, device=input_device),
|
| 536 |
+
split_size_or_sections=num_images,
|
| 537 |
+
dim=0
|
| 538 |
+
)
|
| 539 |
+
visual_input_ids = torch.split(
|
| 540 |
+
torch.argmax(visual_tokens, dim=-1).to(device=input_device),
|
| 541 |
+
split_size_or_sections=num_images,
|
| 542 |
+
dim=0
|
| 543 |
+
)
|
| 544 |
+
visual_labels = [
|
| 545 |
+
torch.full(
|
| 546 |
+
x.shape, IGNORE_INDEX, dtype=torch.long, device=input_device
|
| 547 |
+
) for x in visual_input_ids
|
| 548 |
+
]
|
| 549 |
+
else:
|
| 550 |
+
# just placeholders
|
| 551 |
+
visual_embeds = [None] * len(num_images)
|
| 552 |
+
visual_input_ids = [None] * len(num_images)
|
| 553 |
+
visual_labels = [None] * len(num_images)
|
| 554 |
+
# just placeholders
|
| 555 |
+
text_labels = torch.full(text_input_ids.shape, IGNORE_INDEX, dtype=torch.long, device=input_device)
|
| 556 |
+
|
| 557 |
+
input_embeds = []
|
| 558 |
+
attention_masks = []
|
| 559 |
+
labels = []
|
| 560 |
+
for text_input_id, text_label, text_attention_mask, visual_embed, visual_input_id, visual_label in zip(
|
| 561 |
+
text_input_ids, text_labels, text_attention_masks, visual_embeds, visual_input_ids, visual_labels
|
| 562 |
+
):
|
| 563 |
+
image_token_mask = torch.eq(text_input_id, IMAGE_TOKEN_INDEX)
|
| 564 |
+
text_embed = self.get_wte()(torch.masked_fill(text_input_id, image_token_mask, 0))
|
| 565 |
+
image_token_positions = torch.where(image_token_mask)[0].tolist()
|
| 566 |
+
if len(image_token_positions) > 0:
|
| 567 |
+
input_embed_parts = []
|
| 568 |
+
attention_mask_parts = []
|
| 569 |
+
label_parts = []
|
| 570 |
+
prev_image_token_position = -1
|
| 571 |
+
for index, image_token_position in enumerate(image_token_positions):
|
| 572 |
+
input_embed_parts.append(
|
| 573 |
+
text_embed[prev_image_token_position + 1:image_token_position, :])
|
| 574 |
+
label_parts.append(
|
| 575 |
+
text_label[prev_image_token_position + 1:image_token_position])
|
| 576 |
+
attention_mask_parts.append(
|
| 577 |
+
text_attention_mask[prev_image_token_position + 1:image_token_position])
|
| 578 |
+
input_embed_parts.append(visual_embed[index])
|
| 579 |
+
attention_mask_parts.append(
|
| 580 |
+
torch.ones_like(visual_label[index], dtype=torch.bool))
|
| 581 |
+
label_parts.append(visual_label[index])
|
| 582 |
+
prev_image_token_position = image_token_position
|
| 583 |
+
if prev_image_token_position + 1 < text_input_id.shape[0]:
|
| 584 |
+
input_embed_parts.append(
|
| 585 |
+
text_embed[prev_image_token_position + 1:, :])
|
| 586 |
+
attention_mask_parts.append(
|
| 587 |
+
text_attention_mask[prev_image_token_position + 1:])
|
| 588 |
+
label_parts.append(
|
| 589 |
+
text_label[prev_image_token_position + 1:])
|
| 590 |
+
input_embed = torch.cat(input_embed_parts, dim=0)
|
| 591 |
+
attention_mask = torch.cat(attention_mask_parts, dim=0)
|
| 592 |
+
label = torch.cat(label_parts, dim=0)
|
| 593 |
+
else:
|
| 594 |
+
input_embed = text_embed
|
| 595 |
+
attention_mask = text_attention_mask
|
| 596 |
+
label = text_label
|
| 597 |
+
if self.training:
|
| 598 |
+
# Make visual_embed involved in the backward graph,
|
| 599 |
+
# to be compatible with deepspeed zero and ddp.
|
| 600 |
+
input_embed += torch.sum(visual_embed * 0.0)
|
| 601 |
+
input_embeds.append(input_embed)
|
| 602 |
+
attention_masks.append(attention_mask)
|
| 603 |
+
labels.append(label)
|
| 604 |
+
|
| 605 |
+
batch_input_embeds = torch.nn.utils.rnn.pad_sequence(
|
| 606 |
+
input_embeds, batch_first=True, padding_value=0.0)[:, :self.config.multimodal_max_length, :]
|
| 607 |
+
batch_attention_mask = torch.nn.utils.rnn.pad_sequence(
|
| 608 |
+
attention_masks, batch_first=True, padding_value=False)[:, :self.config.multimodal_max_length]
|
| 609 |
+
batch_labels = torch.nn.utils.rnn.pad_sequence(
|
| 610 |
+
labels, batch_first=True, padding_value=IGNORE_INDEX)[:, :self.config.multimodal_max_length]
|
| 611 |
+
|
| 612 |
+
return visual_input_ids, batch_input_embeds, batch_labels, batch_attention_mask
|
| 613 |
+
|
| 614 |
+
def save_pretrained(
|
| 615 |
+
self,
|
| 616 |
+
save_directory: Union[str, os.PathLike],
|
| 617 |
+
is_main_process: bool = True,
|
| 618 |
+
state_dict: Optional[dict] = None,
|
| 619 |
+
save_function: Callable = torch.save,
|
| 620 |
+
push_to_hub: bool = False,
|
| 621 |
+
max_shard_size: Union[int, str] = "5GB",
|
| 622 |
+
safe_serialization: bool = True,
|
| 623 |
+
variant: Optional[str] = None,
|
| 624 |
+
token: Optional[Union[str, bool]] = None,
|
| 625 |
+
save_peft_format: bool = True,
|
| 626 |
+
**kwargs
|
| 627 |
+
):
|
| 628 |
+
super().save_pretrained(save_directory,
|
| 629 |
+
is_main_process=is_main_process,
|
| 630 |
+
state_dict=state_dict,
|
| 631 |
+
save_function=save_function,
|
| 632 |
+
safe_serialization=safe_serialization)
|
| 633 |
+
self.get_text_tokenizer().save_pretrained(save_directory)
|
| 634 |
+
self.get_visual_tokenizer().get_image_processor().save_pretrained(save_directory)
|
| 635 |
+
|
| 636 |
+
# uncomment the following will additionally save a separate visual tokenizer
|
| 637 |
+
# visual_tokenizer_directory = os.path.join(save_directory, 'visual_tokenizer')
|
| 638 |
+
# self.get_visual_tokenizer().save_pretrained(visual_tokenizer_directory,
|
| 639 |
+
# is_main_process=is_main_process,
|
| 640 |
+
# state_dict=None,
|
| 641 |
+
# save_function=save_function,
|
| 642 |
+
# safe_serialization=safe_serialization)
|
| 643 |
+
# self.get_visual_tokenizer().get_image_processor().save_pretrained(visual_tokenizer_directory)
|
| 644 |
+
|
| 645 |
+
def _get_hybrid_cache_for_llm(self, max_batch_size: int, max_cache_len: int):
|
| 646 |
+
cache_cls = HybridCache
|
| 647 |
+
llm = self.get_llm()
|
| 648 |
+
|
| 649 |
+
need_new_cache = (
|
| 650 |
+
not hasattr(llm, "_cache")
|
| 651 |
+
or (not isinstance(llm._cache, cache_cls))
|
| 652 |
+
or llm._cache.max_batch_size != max_batch_size
|
| 653 |
+
or llm._cache.max_cache_len < max_cache_len
|
| 654 |
+
)
|
| 655 |
+
|
| 656 |
+
if need_new_cache:
|
| 657 |
+
if hasattr(llm.config, "_pre_quantization_dtype"):
|
| 658 |
+
cache_dtype = llm.config._pre_quantization_dtype
|
| 659 |
+
else:
|
| 660 |
+
cache_dtype = llm.dtype
|
| 661 |
+
llm._cache = cache_cls(
|
| 662 |
+
config=llm.config,
|
| 663 |
+
max_batch_size=max_batch_size,
|
| 664 |
+
max_cache_len=max_cache_len,
|
| 665 |
+
device=llm.device,
|
| 666 |
+
dtype=cache_dtype,
|
| 667 |
+
)
|
| 668 |
+
else:
|
| 669 |
+
llm._cache.reset()
|
| 670 |
+
return llm._cache
|
| 671 |
+
|
| 672 |
+
# TODO: support batch generation
|
| 673 |
+
def generate(
|
| 674 |
+
self,
|
| 675 |
+
inputs: Optional[torch.Tensor] = None,
|
| 676 |
+
**kwargs
|
| 677 |
+
) -> Union[GenerateOutput, torch.LongTensor]:
|
| 678 |
+
assert inputs.shape[0] == 1, 'Currently, only support `batch_size=1`'
|
| 679 |
+
_, inputs_embeds, labels, attention_mask = self.merge_multimodal(
|
| 680 |
+
text_input_ids=inputs,
|
| 681 |
+
text_attention_masks=kwargs.pop('attention_mask'),
|
| 682 |
+
text_labels=None,
|
| 683 |
+
pixel_values=kwargs.pop('pixel_values')
|
| 684 |
+
)
|
| 685 |
+
if getattr(self.generation_config, 'cache_implementation') == 'hybrid': # mainly for Gemma2
|
| 686 |
+
kwargs['past_key_values'] = self._get_hybrid_cache_for_llm(
|
| 687 |
+
getattr(kwargs, "num_beams", 1), kwargs['max_new_tokens'] + inputs_embeds.shape[-2])
|
| 688 |
+
self.get_llm()._supports_cache_class = True
|
| 689 |
+
kwargs['cache_implementation'] = None
|
| 690 |
+
|
| 691 |
+
return self.llm.generate(inputs=None, inputs_embeds=inputs_embeds, attention_mask=attention_mask, **kwargs)
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"processor_class": "SiglipProcessor",
|
| 18 |
+
"resample": 3,
|
| 19 |
+
"rescale_factor": 0.00392156862745098,
|
| 20 |
+
"size": {
|
| 21 |
+
"height": 384,
|
| 22 |
+
"width": 384
|
| 23 |
+
}
|
| 24 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<start_of_turn>",
|
| 4 |
+
"<end_of_turn>"
|
| 5 |
+
],
|
| 6 |
+
"bos_token": {
|
| 7 |
+
"content": "<bos>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"content": "<eos>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
},
|
| 20 |
+
"pad_token": {
|
| 21 |
+
"content": "<pad>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false
|
| 26 |
+
},
|
| 27 |
+
"unk_token": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false
|
| 33 |
+
}
|
| 34 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7da53ca29fb16f6b2489482fc0bc6a394162cdab14d12764a1755ebc583fea79
|
| 3 |
+
size 17518525
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
|
| 3 |
+
size 4241003
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,1757 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<pad>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<eos>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<bos>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"4": {
|
| 38 |
+
"content": "<mask>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": false
|
| 44 |
+
},
|
| 45 |
+
"5": {
|
| 46 |
+
"content": "<2mass>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": false
|
| 52 |
+
},
|
| 53 |
+
"6": {
|
| 54 |
+
"content": "[@BOS@]",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": false
|
| 60 |
+
},
|
| 61 |
+
"7": {
|
| 62 |
+
"content": "<unused0>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": false
|
| 68 |
+
},
|
| 69 |
+
"8": {
|
| 70 |
+
"content": "<unused1>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": false
|
| 76 |
+
},
|
| 77 |
+
"9": {
|
| 78 |
+
"content": "<unused2>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": false
|
| 84 |
+
},
|
| 85 |
+
"10": {
|
| 86 |
+
"content": "<unused3>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": false
|
| 92 |
+
},
|
| 93 |
+
"11": {
|
| 94 |
+
"content": "<unused4>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": false
|
| 100 |
+
},
|
| 101 |
+
"12": {
|
| 102 |
+
"content": "<unused5>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": false
|
| 108 |
+
},
|
| 109 |
+
"13": {
|
| 110 |
+
"content": "<unused6>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": false
|
| 116 |
+
},
|
| 117 |
+
"14": {
|
| 118 |
+
"content": "<unused7>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"15": {
|
| 126 |
+
"content": "<unused8>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"16": {
|
| 134 |
+
"content": "<unused9>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"17": {
|
| 142 |
+
"content": "<unused10>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"18": {
|
| 150 |
+
"content": "<unused11>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"19": {
|
| 158 |
+
"content": "<unused12>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"20": {
|
| 166 |
+
"content": "<unused13>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"21": {
|
| 174 |
+
"content": "<unused14>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"22": {
|
| 182 |
+
"content": "<unused15>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"23": {
|
| 190 |
+
"content": "<unused16>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"24": {
|
| 198 |
+
"content": "<unused17>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"25": {
|
| 206 |
+
"content": "<unused18>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
},
|
| 213 |
+
"26": {
|
| 214 |
+
"content": "<unused19>",
|
| 215 |
+
"lstrip": false,
|
| 216 |
+
"normalized": false,
|
| 217 |
+
"rstrip": false,
|
| 218 |
+
"single_word": false,
|
| 219 |
+
"special": false
|
| 220 |
+
},
|
| 221 |
+
"27": {
|
| 222 |
+
"content": "<unused20>",
|
| 223 |
+
"lstrip": false,
|
| 224 |
+
"normalized": false,
|
| 225 |
+
"rstrip": false,
|
| 226 |
+
"single_word": false,
|
| 227 |
+
"special": false
|
| 228 |
+
},
|
| 229 |
+
"28": {
|
| 230 |
+
"content": "<unused21>",
|
| 231 |
+
"lstrip": false,
|
| 232 |
+
"normalized": false,
|
| 233 |
+
"rstrip": false,
|
| 234 |
+
"single_word": false,
|
| 235 |
+
"special": false
|
| 236 |
+
},
|
| 237 |
+
"29": {
|
| 238 |
+
"content": "<unused22>",
|
| 239 |
+
"lstrip": false,
|
| 240 |
+
"normalized": false,
|
| 241 |
+
"rstrip": false,
|
| 242 |
+
"single_word": false,
|
| 243 |
+
"special": false
|
| 244 |
+
},
|
| 245 |
+
"30": {
|
| 246 |
+
"content": "<unused23>",
|
| 247 |
+
"lstrip": false,
|
| 248 |
+
"normalized": false,
|
| 249 |
+
"rstrip": false,
|
| 250 |
+
"single_word": false,
|
| 251 |
+
"special": false
|
| 252 |
+
},
|
| 253 |
+
"31": {
|
| 254 |
+
"content": "<unused24>",
|
| 255 |
+
"lstrip": false,
|
| 256 |
+
"normalized": false,
|
| 257 |
+
"rstrip": false,
|
| 258 |
+
"single_word": false,
|
| 259 |
+
"special": false
|
| 260 |
+
},
|
| 261 |
+
"32": {
|
| 262 |
+
"content": "<unused25>",
|
| 263 |
+
"lstrip": false,
|
| 264 |
+
"normalized": false,
|
| 265 |
+
"rstrip": false,
|
| 266 |
+
"single_word": false,
|
| 267 |
+
"special": false
|
| 268 |
+
},
|
| 269 |
+
"33": {
|
| 270 |
+
"content": "<unused26>",
|
| 271 |
+
"lstrip": false,
|
| 272 |
+
"normalized": false,
|
| 273 |
+
"rstrip": false,
|
| 274 |
+
"single_word": false,
|
| 275 |
+
"special": false
|
| 276 |
+
},
|
| 277 |
+
"34": {
|
| 278 |
+
"content": "<unused27>",
|
| 279 |
+
"lstrip": false,
|
| 280 |
+
"normalized": false,
|
| 281 |
+
"rstrip": false,
|
| 282 |
+
"single_word": false,
|
| 283 |
+
"special": false
|
| 284 |
+
},
|
| 285 |
+
"35": {
|
| 286 |
+
"content": "<unused28>",
|
| 287 |
+
"lstrip": false,
|
| 288 |
+
"normalized": false,
|
| 289 |
+
"rstrip": false,
|
| 290 |
+
"single_word": false,
|
| 291 |
+
"special": false
|
| 292 |
+
},
|
| 293 |
+
"36": {
|
| 294 |
+
"content": "<unused29>",
|
| 295 |
+
"lstrip": false,
|
| 296 |
+
"normalized": false,
|
| 297 |
+
"rstrip": false,
|
| 298 |
+
"single_word": false,
|
| 299 |
+
"special": false
|
| 300 |
+
},
|
| 301 |
+
"37": {
|
| 302 |
+
"content": "<unused30>",
|
| 303 |
+
"lstrip": false,
|
| 304 |
+
"normalized": false,
|
| 305 |
+
"rstrip": false,
|
| 306 |
+
"single_word": false,
|
| 307 |
+
"special": false
|
| 308 |
+
},
|
| 309 |
+
"38": {
|
| 310 |
+
"content": "<unused31>",
|
| 311 |
+
"lstrip": false,
|
| 312 |
+
"normalized": false,
|
| 313 |
+
"rstrip": false,
|
| 314 |
+
"single_word": false,
|
| 315 |
+
"special": false
|
| 316 |
+
},
|
| 317 |
+
"39": {
|
| 318 |
+
"content": "<unused32>",
|
| 319 |
+
"lstrip": false,
|
| 320 |
+
"normalized": false,
|
| 321 |
+
"rstrip": false,
|
| 322 |
+
"single_word": false,
|
| 323 |
+
"special": false
|
| 324 |
+
},
|
| 325 |
+
"40": {
|
| 326 |
+
"content": "<unused33>",
|
| 327 |
+
"lstrip": false,
|
| 328 |
+
"normalized": false,
|
| 329 |
+
"rstrip": false,
|
| 330 |
+
"single_word": false,
|
| 331 |
+
"special": false
|
| 332 |
+
},
|
| 333 |
+
"41": {
|
| 334 |
+
"content": "<unused34>",
|
| 335 |
+
"lstrip": false,
|
| 336 |
+
"normalized": false,
|
| 337 |
+
"rstrip": false,
|
| 338 |
+
"single_word": false,
|
| 339 |
+
"special": false
|
| 340 |
+
},
|
| 341 |
+
"42": {
|
| 342 |
+
"content": "<unused35>",
|
| 343 |
+
"lstrip": false,
|
| 344 |
+
"normalized": false,
|
| 345 |
+
"rstrip": false,
|
| 346 |
+
"single_word": false,
|
| 347 |
+
"special": false
|
| 348 |
+
},
|
| 349 |
+
"43": {
|
| 350 |
+
"content": "<unused36>",
|
| 351 |
+
"lstrip": false,
|
| 352 |
+
"normalized": false,
|
| 353 |
+
"rstrip": false,
|
| 354 |
+
"single_word": false,
|
| 355 |
+
"special": false
|
| 356 |
+
},
|
| 357 |
+
"44": {
|
| 358 |
+
"content": "<unused37>",
|
| 359 |
+
"lstrip": false,
|
| 360 |
+
"normalized": false,
|
| 361 |
+
"rstrip": false,
|
| 362 |
+
"single_word": false,
|
| 363 |
+
"special": false
|
| 364 |
+
},
|
| 365 |
+
"45": {
|
| 366 |
+
"content": "<unused38>",
|
| 367 |
+
"lstrip": false,
|
| 368 |
+
"normalized": false,
|
| 369 |
+
"rstrip": false,
|
| 370 |
+
"single_word": false,
|
| 371 |
+
"special": false
|
| 372 |
+
},
|
| 373 |
+
"46": {
|
| 374 |
+
"content": "<unused39>",
|
| 375 |
+
"lstrip": false,
|
| 376 |
+
"normalized": false,
|
| 377 |
+
"rstrip": false,
|
| 378 |
+
"single_word": false,
|
| 379 |
+
"special": false
|
| 380 |
+
},
|
| 381 |
+
"47": {
|
| 382 |
+
"content": "<unused40>",
|
| 383 |
+
"lstrip": false,
|
| 384 |
+
"normalized": false,
|
| 385 |
+
"rstrip": false,
|
| 386 |
+
"single_word": false,
|
| 387 |
+
"special": false
|
| 388 |
+
},
|
| 389 |
+
"48": {
|
| 390 |
+
"content": "<unused41>",
|
| 391 |
+
"lstrip": false,
|
| 392 |
+
"normalized": false,
|
| 393 |
+
"rstrip": false,
|
| 394 |
+
"single_word": false,
|
| 395 |
+
"special": false
|
| 396 |
+
},
|
| 397 |
+
"49": {
|
| 398 |
+
"content": "<unused42>",
|
| 399 |
+
"lstrip": false,
|
| 400 |
+
"normalized": false,
|
| 401 |
+
"rstrip": false,
|
| 402 |
+
"single_word": false,
|
| 403 |
+
"special": false
|
| 404 |
+
},
|
| 405 |
+
"50": {
|
| 406 |
+
"content": "<unused43>",
|
| 407 |
+
"lstrip": false,
|
| 408 |
+
"normalized": false,
|
| 409 |
+
"rstrip": false,
|
| 410 |
+
"single_word": false,
|
| 411 |
+
"special": false
|
| 412 |
+
},
|
| 413 |
+
"51": {
|
| 414 |
+
"content": "<unused44>",
|
| 415 |
+
"lstrip": false,
|
| 416 |
+
"normalized": false,
|
| 417 |
+
"rstrip": false,
|
| 418 |
+
"single_word": false,
|
| 419 |
+
"special": false
|
| 420 |
+
},
|
| 421 |
+
"52": {
|
| 422 |
+
"content": "<unused45>",
|
| 423 |
+
"lstrip": false,
|
| 424 |
+
"normalized": false,
|
| 425 |
+
"rstrip": false,
|
| 426 |
+
"single_word": false,
|
| 427 |
+
"special": false
|
| 428 |
+
},
|
| 429 |
+
"53": {
|
| 430 |
+
"content": "<unused46>",
|
| 431 |
+
"lstrip": false,
|
| 432 |
+
"normalized": false,
|
| 433 |
+
"rstrip": false,
|
| 434 |
+
"single_word": false,
|
| 435 |
+
"special": false
|
| 436 |
+
},
|
| 437 |
+
"54": {
|
| 438 |
+
"content": "<unused47>",
|
| 439 |
+
"lstrip": false,
|
| 440 |
+
"normalized": false,
|
| 441 |
+
"rstrip": false,
|
| 442 |
+
"single_word": false,
|
| 443 |
+
"special": false
|
| 444 |
+
},
|
| 445 |
+
"55": {
|
| 446 |
+
"content": "<unused48>",
|
| 447 |
+
"lstrip": false,
|
| 448 |
+
"normalized": false,
|
| 449 |
+
"rstrip": false,
|
| 450 |
+
"single_word": false,
|
| 451 |
+
"special": false
|
| 452 |
+
},
|
| 453 |
+
"56": {
|
| 454 |
+
"content": "<unused49>",
|
| 455 |
+
"lstrip": false,
|
| 456 |
+
"normalized": false,
|
| 457 |
+
"rstrip": false,
|
| 458 |
+
"single_word": false,
|
| 459 |
+
"special": false
|
| 460 |
+
},
|
| 461 |
+
"57": {
|
| 462 |
+
"content": "<unused50>",
|
| 463 |
+
"lstrip": false,
|
| 464 |
+
"normalized": false,
|
| 465 |
+
"rstrip": false,
|
| 466 |
+
"single_word": false,
|
| 467 |
+
"special": false
|
| 468 |
+
},
|
| 469 |
+
"58": {
|
| 470 |
+
"content": "<unused51>",
|
| 471 |
+
"lstrip": false,
|
| 472 |
+
"normalized": false,
|
| 473 |
+
"rstrip": false,
|
| 474 |
+
"single_word": false,
|
| 475 |
+
"special": false
|
| 476 |
+
},
|
| 477 |
+
"59": {
|
| 478 |
+
"content": "<unused52>",
|
| 479 |
+
"lstrip": false,
|
| 480 |
+
"normalized": false,
|
| 481 |
+
"rstrip": false,
|
| 482 |
+
"single_word": false,
|
| 483 |
+
"special": false
|
| 484 |
+
},
|
| 485 |
+
"60": {
|
| 486 |
+
"content": "<unused53>",
|
| 487 |
+
"lstrip": false,
|
| 488 |
+
"normalized": false,
|
| 489 |
+
"rstrip": false,
|
| 490 |
+
"single_word": false,
|
| 491 |
+
"special": false
|
| 492 |
+
},
|
| 493 |
+
"61": {
|
| 494 |
+
"content": "<unused54>",
|
| 495 |
+
"lstrip": false,
|
| 496 |
+
"normalized": false,
|
| 497 |
+
"rstrip": false,
|
| 498 |
+
"single_word": false,
|
| 499 |
+
"special": false
|
| 500 |
+
},
|
| 501 |
+
"62": {
|
| 502 |
+
"content": "<unused55>",
|
| 503 |
+
"lstrip": false,
|
| 504 |
+
"normalized": false,
|
| 505 |
+
"rstrip": false,
|
| 506 |
+
"single_word": false,
|
| 507 |
+
"special": false
|
| 508 |
+
},
|
| 509 |
+
"63": {
|
| 510 |
+
"content": "<unused56>",
|
| 511 |
+
"lstrip": false,
|
| 512 |
+
"normalized": false,
|
| 513 |
+
"rstrip": false,
|
| 514 |
+
"single_word": false,
|
| 515 |
+
"special": false
|
| 516 |
+
},
|
| 517 |
+
"64": {
|
| 518 |
+
"content": "<unused57>",
|
| 519 |
+
"lstrip": false,
|
| 520 |
+
"normalized": false,
|
| 521 |
+
"rstrip": false,
|
| 522 |
+
"single_word": false,
|
| 523 |
+
"special": false
|
| 524 |
+
},
|
| 525 |
+
"65": {
|
| 526 |
+
"content": "<unused58>",
|
| 527 |
+
"lstrip": false,
|
| 528 |
+
"normalized": false,
|
| 529 |
+
"rstrip": false,
|
| 530 |
+
"single_word": false,
|
| 531 |
+
"special": false
|
| 532 |
+
},
|
| 533 |
+
"66": {
|
| 534 |
+
"content": "<unused59>",
|
| 535 |
+
"lstrip": false,
|
| 536 |
+
"normalized": false,
|
| 537 |
+
"rstrip": false,
|
| 538 |
+
"single_word": false,
|
| 539 |
+
"special": false
|
| 540 |
+
},
|
| 541 |
+
"67": {
|
| 542 |
+
"content": "<unused60>",
|
| 543 |
+
"lstrip": false,
|
| 544 |
+
"normalized": false,
|
| 545 |
+
"rstrip": false,
|
| 546 |
+
"single_word": false,
|
| 547 |
+
"special": false
|
| 548 |
+
},
|
| 549 |
+
"68": {
|
| 550 |
+
"content": "<unused61>",
|
| 551 |
+
"lstrip": false,
|
| 552 |
+
"normalized": false,
|
| 553 |
+
"rstrip": false,
|
| 554 |
+
"single_word": false,
|
| 555 |
+
"special": false
|
| 556 |
+
},
|
| 557 |
+
"69": {
|
| 558 |
+
"content": "<unused62>",
|
| 559 |
+
"lstrip": false,
|
| 560 |
+
"normalized": false,
|
| 561 |
+
"rstrip": false,
|
| 562 |
+
"single_word": false,
|
| 563 |
+
"special": false
|
| 564 |
+
},
|
| 565 |
+
"70": {
|
| 566 |
+
"content": "<unused63>",
|
| 567 |
+
"lstrip": false,
|
| 568 |
+
"normalized": false,
|
| 569 |
+
"rstrip": false,
|
| 570 |
+
"single_word": false,
|
| 571 |
+
"special": false
|
| 572 |
+
},
|
| 573 |
+
"71": {
|
| 574 |
+
"content": "<unused64>",
|
| 575 |
+
"lstrip": false,
|
| 576 |
+
"normalized": false,
|
| 577 |
+
"rstrip": false,
|
| 578 |
+
"single_word": false,
|
| 579 |
+
"special": false
|
| 580 |
+
},
|
| 581 |
+
"72": {
|
| 582 |
+
"content": "<unused65>",
|
| 583 |
+
"lstrip": false,
|
| 584 |
+
"normalized": false,
|
| 585 |
+
"rstrip": false,
|
| 586 |
+
"single_word": false,
|
| 587 |
+
"special": false
|
| 588 |
+
},
|
| 589 |
+
"73": {
|
| 590 |
+
"content": "<unused66>",
|
| 591 |
+
"lstrip": false,
|
| 592 |
+
"normalized": false,
|
| 593 |
+
"rstrip": false,
|
| 594 |
+
"single_word": false,
|
| 595 |
+
"special": false
|
| 596 |
+
},
|
| 597 |
+
"74": {
|
| 598 |
+
"content": "<unused67>",
|
| 599 |
+
"lstrip": false,
|
| 600 |
+
"normalized": false,
|
| 601 |
+
"rstrip": false,
|
| 602 |
+
"single_word": false,
|
| 603 |
+
"special": false
|
| 604 |
+
},
|
| 605 |
+
"75": {
|
| 606 |
+
"content": "<unused68>",
|
| 607 |
+
"lstrip": false,
|
| 608 |
+
"normalized": false,
|
| 609 |
+
"rstrip": false,
|
| 610 |
+
"single_word": false,
|
| 611 |
+
"special": false
|
| 612 |
+
},
|
| 613 |
+
"76": {
|
| 614 |
+
"content": "<unused69>",
|
| 615 |
+
"lstrip": false,
|
| 616 |
+
"normalized": false,
|
| 617 |
+
"rstrip": false,
|
| 618 |
+
"single_word": false,
|
| 619 |
+
"special": false
|
| 620 |
+
},
|
| 621 |
+
"77": {
|
| 622 |
+
"content": "<unused70>",
|
| 623 |
+
"lstrip": false,
|
| 624 |
+
"normalized": false,
|
| 625 |
+
"rstrip": false,
|
| 626 |
+
"single_word": false,
|
| 627 |
+
"special": false
|
| 628 |
+
},
|
| 629 |
+
"78": {
|
| 630 |
+
"content": "<unused71>",
|
| 631 |
+
"lstrip": false,
|
| 632 |
+
"normalized": false,
|
| 633 |
+
"rstrip": false,
|
| 634 |
+
"single_word": false,
|
| 635 |
+
"special": false
|
| 636 |
+
},
|
| 637 |
+
"79": {
|
| 638 |
+
"content": "<unused72>",
|
| 639 |
+
"lstrip": false,
|
| 640 |
+
"normalized": false,
|
| 641 |
+
"rstrip": false,
|
| 642 |
+
"single_word": false,
|
| 643 |
+
"special": false
|
| 644 |
+
},
|
| 645 |
+
"80": {
|
| 646 |
+
"content": "<unused73>",
|
| 647 |
+
"lstrip": false,
|
| 648 |
+
"normalized": false,
|
| 649 |
+
"rstrip": false,
|
| 650 |
+
"single_word": false,
|
| 651 |
+
"special": false
|
| 652 |
+
},
|
| 653 |
+
"81": {
|
| 654 |
+
"content": "<unused74>",
|
| 655 |
+
"lstrip": false,
|
| 656 |
+
"normalized": false,
|
| 657 |
+
"rstrip": false,
|
| 658 |
+
"single_word": false,
|
| 659 |
+
"special": false
|
| 660 |
+
},
|
| 661 |
+
"82": {
|
| 662 |
+
"content": "<unused75>",
|
| 663 |
+
"lstrip": false,
|
| 664 |
+
"normalized": false,
|
| 665 |
+
"rstrip": false,
|
| 666 |
+
"single_word": false,
|
| 667 |
+
"special": false
|
| 668 |
+
},
|
| 669 |
+
"83": {
|
| 670 |
+
"content": "<unused76>",
|
| 671 |
+
"lstrip": false,
|
| 672 |
+
"normalized": false,
|
| 673 |
+
"rstrip": false,
|
| 674 |
+
"single_word": false,
|
| 675 |
+
"special": false
|
| 676 |
+
},
|
| 677 |
+
"84": {
|
| 678 |
+
"content": "<unused77>",
|
| 679 |
+
"lstrip": false,
|
| 680 |
+
"normalized": false,
|
| 681 |
+
"rstrip": false,
|
| 682 |
+
"single_word": false,
|
| 683 |
+
"special": false
|
| 684 |
+
},
|
| 685 |
+
"85": {
|
| 686 |
+
"content": "<unused78>",
|
| 687 |
+
"lstrip": false,
|
| 688 |
+
"normalized": false,
|
| 689 |
+
"rstrip": false,
|
| 690 |
+
"single_word": false,
|
| 691 |
+
"special": false
|
| 692 |
+
},
|
| 693 |
+
"86": {
|
| 694 |
+
"content": "<unused79>",
|
| 695 |
+
"lstrip": false,
|
| 696 |
+
"normalized": false,
|
| 697 |
+
"rstrip": false,
|
| 698 |
+
"single_word": false,
|
| 699 |
+
"special": false
|
| 700 |
+
},
|
| 701 |
+
"87": {
|
| 702 |
+
"content": "<unused80>",
|
| 703 |
+
"lstrip": false,
|
| 704 |
+
"normalized": false,
|
| 705 |
+
"rstrip": false,
|
| 706 |
+
"single_word": false,
|
| 707 |
+
"special": false
|
| 708 |
+
},
|
| 709 |
+
"88": {
|
| 710 |
+
"content": "<unused81>",
|
| 711 |
+
"lstrip": false,
|
| 712 |
+
"normalized": false,
|
| 713 |
+
"rstrip": false,
|
| 714 |
+
"single_word": false,
|
| 715 |
+
"special": false
|
| 716 |
+
},
|
| 717 |
+
"89": {
|
| 718 |
+
"content": "<unused82>",
|
| 719 |
+
"lstrip": false,
|
| 720 |
+
"normalized": false,
|
| 721 |
+
"rstrip": false,
|
| 722 |
+
"single_word": false,
|
| 723 |
+
"special": false
|
| 724 |
+
},
|
| 725 |
+
"90": {
|
| 726 |
+
"content": "<unused83>",
|
| 727 |
+
"lstrip": false,
|
| 728 |
+
"normalized": false,
|
| 729 |
+
"rstrip": false,
|
| 730 |
+
"single_word": false,
|
| 731 |
+
"special": false
|
| 732 |
+
},
|
| 733 |
+
"91": {
|
| 734 |
+
"content": "<unused84>",
|
| 735 |
+
"lstrip": false,
|
| 736 |
+
"normalized": false,
|
| 737 |
+
"rstrip": false,
|
| 738 |
+
"single_word": false,
|
| 739 |
+
"special": false
|
| 740 |
+
},
|
| 741 |
+
"92": {
|
| 742 |
+
"content": "<unused85>",
|
| 743 |
+
"lstrip": false,
|
| 744 |
+
"normalized": false,
|
| 745 |
+
"rstrip": false,
|
| 746 |
+
"single_word": false,
|
| 747 |
+
"special": false
|
| 748 |
+
},
|
| 749 |
+
"93": {
|
| 750 |
+
"content": "<unused86>",
|
| 751 |
+
"lstrip": false,
|
| 752 |
+
"normalized": false,
|
| 753 |
+
"rstrip": false,
|
| 754 |
+
"single_word": false,
|
| 755 |
+
"special": false
|
| 756 |
+
},
|
| 757 |
+
"94": {
|
| 758 |
+
"content": "<unused87>",
|
| 759 |
+
"lstrip": false,
|
| 760 |
+
"normalized": false,
|
| 761 |
+
"rstrip": false,
|
| 762 |
+
"single_word": false,
|
| 763 |
+
"special": false
|
| 764 |
+
},
|
| 765 |
+
"95": {
|
| 766 |
+
"content": "<unused88>",
|
| 767 |
+
"lstrip": false,
|
| 768 |
+
"normalized": false,
|
| 769 |
+
"rstrip": false,
|
| 770 |
+
"single_word": false,
|
| 771 |
+
"special": false
|
| 772 |
+
},
|
| 773 |
+
"96": {
|
| 774 |
+
"content": "<unused89>",
|
| 775 |
+
"lstrip": false,
|
| 776 |
+
"normalized": false,
|
| 777 |
+
"rstrip": false,
|
| 778 |
+
"single_word": false,
|
| 779 |
+
"special": false
|
| 780 |
+
},
|
| 781 |
+
"97": {
|
| 782 |
+
"content": "<unused90>",
|
| 783 |
+
"lstrip": false,
|
| 784 |
+
"normalized": false,
|
| 785 |
+
"rstrip": false,
|
| 786 |
+
"single_word": false,
|
| 787 |
+
"special": false
|
| 788 |
+
},
|
| 789 |
+
"98": {
|
| 790 |
+
"content": "<unused91>",
|
| 791 |
+
"lstrip": false,
|
| 792 |
+
"normalized": false,
|
| 793 |
+
"rstrip": false,
|
| 794 |
+
"single_word": false,
|
| 795 |
+
"special": false
|
| 796 |
+
},
|
| 797 |
+
"99": {
|
| 798 |
+
"content": "<unused92>",
|
| 799 |
+
"lstrip": false,
|
| 800 |
+
"normalized": false,
|
| 801 |
+
"rstrip": false,
|
| 802 |
+
"single_word": false,
|
| 803 |
+
"special": false
|
| 804 |
+
},
|
| 805 |
+
"100": {
|
| 806 |
+
"content": "<unused93>",
|
| 807 |
+
"lstrip": false,
|
| 808 |
+
"normalized": false,
|
| 809 |
+
"rstrip": false,
|
| 810 |
+
"single_word": false,
|
| 811 |
+
"special": false
|
| 812 |
+
},
|
| 813 |
+
"101": {
|
| 814 |
+
"content": "<unused94>",
|
| 815 |
+
"lstrip": false,
|
| 816 |
+
"normalized": false,
|
| 817 |
+
"rstrip": false,
|
| 818 |
+
"single_word": false,
|
| 819 |
+
"special": false
|
| 820 |
+
},
|
| 821 |
+
"102": {
|
| 822 |
+
"content": "<unused95>",
|
| 823 |
+
"lstrip": false,
|
| 824 |
+
"normalized": false,
|
| 825 |
+
"rstrip": false,
|
| 826 |
+
"single_word": false,
|
| 827 |
+
"special": false
|
| 828 |
+
},
|
| 829 |
+
"103": {
|
| 830 |
+
"content": "<unused96>",
|
| 831 |
+
"lstrip": false,
|
| 832 |
+
"normalized": false,
|
| 833 |
+
"rstrip": false,
|
| 834 |
+
"single_word": false,
|
| 835 |
+
"special": false
|
| 836 |
+
},
|
| 837 |
+
"104": {
|
| 838 |
+
"content": "<unused97>",
|
| 839 |
+
"lstrip": false,
|
| 840 |
+
"normalized": false,
|
| 841 |
+
"rstrip": false,
|
| 842 |
+
"single_word": false,
|
| 843 |
+
"special": false
|
| 844 |
+
},
|
| 845 |
+
"105": {
|
| 846 |
+
"content": "<unused98>",
|
| 847 |
+
"lstrip": false,
|
| 848 |
+
"normalized": false,
|
| 849 |
+
"rstrip": false,
|
| 850 |
+
"single_word": false,
|
| 851 |
+
"special": false
|
| 852 |
+
},
|
| 853 |
+
"106": {
|
| 854 |
+
"content": "<start_of_turn>",
|
| 855 |
+
"lstrip": false,
|
| 856 |
+
"normalized": false,
|
| 857 |
+
"rstrip": false,
|
| 858 |
+
"single_word": false,
|
| 859 |
+
"special": true
|
| 860 |
+
},
|
| 861 |
+
"107": {
|
| 862 |
+
"content": "<end_of_turn>",
|
| 863 |
+
"lstrip": false,
|
| 864 |
+
"normalized": false,
|
| 865 |
+
"rstrip": false,
|
| 866 |
+
"single_word": false,
|
| 867 |
+
"special": true
|
| 868 |
+
},
|
| 869 |
+
"108": {
|
| 870 |
+
"content": "\n",
|
| 871 |
+
"lstrip": false,
|
| 872 |
+
"normalized": false,
|
| 873 |
+
"rstrip": false,
|
| 874 |
+
"single_word": false,
|
| 875 |
+
"special": false
|
| 876 |
+
},
|
| 877 |
+
"109": {
|
| 878 |
+
"content": "\n\n",
|
| 879 |
+
"lstrip": false,
|
| 880 |
+
"normalized": false,
|
| 881 |
+
"rstrip": false,
|
| 882 |
+
"single_word": false,
|
| 883 |
+
"special": false
|
| 884 |
+
},
|
| 885 |
+
"110": {
|
| 886 |
+
"content": "\n\n\n",
|
| 887 |
+
"lstrip": false,
|
| 888 |
+
"normalized": false,
|
| 889 |
+
"rstrip": false,
|
| 890 |
+
"single_word": false,
|
| 891 |
+
"special": false
|
| 892 |
+
},
|
| 893 |
+
"111": {
|
| 894 |
+
"content": "\n\n\n\n",
|
| 895 |
+
"lstrip": false,
|
| 896 |
+
"normalized": false,
|
| 897 |
+
"rstrip": false,
|
| 898 |
+
"single_word": false,
|
| 899 |
+
"special": false
|
| 900 |
+
},
|
| 901 |
+
"112": {
|
| 902 |
+
"content": "\n\n\n\n\n",
|
| 903 |
+
"lstrip": false,
|
| 904 |
+
"normalized": false,
|
| 905 |
+
"rstrip": false,
|
| 906 |
+
"single_word": false,
|
| 907 |
+
"special": false
|
| 908 |
+
},
|
| 909 |
+
"113": {
|
| 910 |
+
"content": "\n\n\n\n\n\n",
|
| 911 |
+
"lstrip": false,
|
| 912 |
+
"normalized": false,
|
| 913 |
+
"rstrip": false,
|
| 914 |
+
"single_word": false,
|
| 915 |
+
"special": false
|
| 916 |
+
},
|
| 917 |
+
"114": {
|
| 918 |
+
"content": "\n\n\n\n\n\n\n",
|
| 919 |
+
"lstrip": false,
|
| 920 |
+
"normalized": false,
|
| 921 |
+
"rstrip": false,
|
| 922 |
+
"single_word": false,
|
| 923 |
+
"special": false
|
| 924 |
+
},
|
| 925 |
+
"115": {
|
| 926 |
+
"content": "\n\n\n\n\n\n\n\n",
|
| 927 |
+
"lstrip": false,
|
| 928 |
+
"normalized": false,
|
| 929 |
+
"rstrip": false,
|
| 930 |
+
"single_word": false,
|
| 931 |
+
"special": false
|
| 932 |
+
},
|
| 933 |
+
"116": {
|
| 934 |
+
"content": "\n\n\n\n\n\n\n\n\n",
|
| 935 |
+
"lstrip": false,
|
| 936 |
+
"normalized": false,
|
| 937 |
+
"rstrip": false,
|
| 938 |
+
"single_word": false,
|
| 939 |
+
"special": false
|
| 940 |
+
},
|
| 941 |
+
"117": {
|
| 942 |
+
"content": "\n\n\n\n\n\n\n\n\n\n",
|
| 943 |
+
"lstrip": false,
|
| 944 |
+
"normalized": false,
|
| 945 |
+
"rstrip": false,
|
| 946 |
+
"single_word": false,
|
| 947 |
+
"special": false
|
| 948 |
+
},
|
| 949 |
+
"118": {
|
| 950 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n",
|
| 951 |
+
"lstrip": false,
|
| 952 |
+
"normalized": false,
|
| 953 |
+
"rstrip": false,
|
| 954 |
+
"single_word": false,
|
| 955 |
+
"special": false
|
| 956 |
+
},
|
| 957 |
+
"119": {
|
| 958 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 959 |
+
"lstrip": false,
|
| 960 |
+
"normalized": false,
|
| 961 |
+
"rstrip": false,
|
| 962 |
+
"single_word": false,
|
| 963 |
+
"special": false
|
| 964 |
+
},
|
| 965 |
+
"120": {
|
| 966 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 967 |
+
"lstrip": false,
|
| 968 |
+
"normalized": false,
|
| 969 |
+
"rstrip": false,
|
| 970 |
+
"single_word": false,
|
| 971 |
+
"special": false
|
| 972 |
+
},
|
| 973 |
+
"121": {
|
| 974 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 975 |
+
"lstrip": false,
|
| 976 |
+
"normalized": false,
|
| 977 |
+
"rstrip": false,
|
| 978 |
+
"single_word": false,
|
| 979 |
+
"special": false
|
| 980 |
+
},
|
| 981 |
+
"122": {
|
| 982 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 983 |
+
"lstrip": false,
|
| 984 |
+
"normalized": false,
|
| 985 |
+
"rstrip": false,
|
| 986 |
+
"single_word": false,
|
| 987 |
+
"special": false
|
| 988 |
+
},
|
| 989 |
+
"123": {
|
| 990 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 991 |
+
"lstrip": false,
|
| 992 |
+
"normalized": false,
|
| 993 |
+
"rstrip": false,
|
| 994 |
+
"single_word": false,
|
| 995 |
+
"special": false
|
| 996 |
+
},
|
| 997 |
+
"124": {
|
| 998 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 999 |
+
"lstrip": false,
|
| 1000 |
+
"normalized": false,
|
| 1001 |
+
"rstrip": false,
|
| 1002 |
+
"single_word": false,
|
| 1003 |
+
"special": false
|
| 1004 |
+
},
|
| 1005 |
+
"125": {
|
| 1006 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1007 |
+
"lstrip": false,
|
| 1008 |
+
"normalized": false,
|
| 1009 |
+
"rstrip": false,
|
| 1010 |
+
"single_word": false,
|
| 1011 |
+
"special": false
|
| 1012 |
+
},
|
| 1013 |
+
"126": {
|
| 1014 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1015 |
+
"lstrip": false,
|
| 1016 |
+
"normalized": false,
|
| 1017 |
+
"rstrip": false,
|
| 1018 |
+
"single_word": false,
|
| 1019 |
+
"special": false
|
| 1020 |
+
},
|
| 1021 |
+
"127": {
|
| 1022 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1023 |
+
"lstrip": false,
|
| 1024 |
+
"normalized": false,
|
| 1025 |
+
"rstrip": false,
|
| 1026 |
+
"single_word": false,
|
| 1027 |
+
"special": false
|
| 1028 |
+
},
|
| 1029 |
+
"128": {
|
| 1030 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1031 |
+
"lstrip": false,
|
| 1032 |
+
"normalized": false,
|
| 1033 |
+
"rstrip": false,
|
| 1034 |
+
"single_word": false,
|
| 1035 |
+
"special": false
|
| 1036 |
+
},
|
| 1037 |
+
"129": {
|
| 1038 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1039 |
+
"lstrip": false,
|
| 1040 |
+
"normalized": false,
|
| 1041 |
+
"rstrip": false,
|
| 1042 |
+
"single_word": false,
|
| 1043 |
+
"special": false
|
| 1044 |
+
},
|
| 1045 |
+
"130": {
|
| 1046 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1047 |
+
"lstrip": false,
|
| 1048 |
+
"normalized": false,
|
| 1049 |
+
"rstrip": false,
|
| 1050 |
+
"single_word": false,
|
| 1051 |
+
"special": false
|
| 1052 |
+
},
|
| 1053 |
+
"131": {
|
| 1054 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1055 |
+
"lstrip": false,
|
| 1056 |
+
"normalized": false,
|
| 1057 |
+
"rstrip": false,
|
| 1058 |
+
"single_word": false,
|
| 1059 |
+
"special": false
|
| 1060 |
+
},
|
| 1061 |
+
"132": {
|
| 1062 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1063 |
+
"lstrip": false,
|
| 1064 |
+
"normalized": false,
|
| 1065 |
+
"rstrip": false,
|
| 1066 |
+
"single_word": false,
|
| 1067 |
+
"special": false
|
| 1068 |
+
},
|
| 1069 |
+
"133": {
|
| 1070 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1071 |
+
"lstrip": false,
|
| 1072 |
+
"normalized": false,
|
| 1073 |
+
"rstrip": false,
|
| 1074 |
+
"single_word": false,
|
| 1075 |
+
"special": false
|
| 1076 |
+
},
|
| 1077 |
+
"134": {
|
| 1078 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1079 |
+
"lstrip": false,
|
| 1080 |
+
"normalized": false,
|
| 1081 |
+
"rstrip": false,
|
| 1082 |
+
"single_word": false,
|
| 1083 |
+
"special": false
|
| 1084 |
+
},
|
| 1085 |
+
"135": {
|
| 1086 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1087 |
+
"lstrip": false,
|
| 1088 |
+
"normalized": false,
|
| 1089 |
+
"rstrip": false,
|
| 1090 |
+
"single_word": false,
|
| 1091 |
+
"special": false
|
| 1092 |
+
},
|
| 1093 |
+
"136": {
|
| 1094 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1095 |
+
"lstrip": false,
|
| 1096 |
+
"normalized": false,
|
| 1097 |
+
"rstrip": false,
|
| 1098 |
+
"single_word": false,
|
| 1099 |
+
"special": false
|
| 1100 |
+
},
|
| 1101 |
+
"137": {
|
| 1102 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1103 |
+
"lstrip": false,
|
| 1104 |
+
"normalized": false,
|
| 1105 |
+
"rstrip": false,
|
| 1106 |
+
"single_word": false,
|
| 1107 |
+
"special": false
|
| 1108 |
+
},
|
| 1109 |
+
"138": {
|
| 1110 |
+
"content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
|
| 1111 |
+
"lstrip": false,
|
| 1112 |
+
"normalized": false,
|
| 1113 |
+
"rstrip": false,
|
| 1114 |
+
"single_word": false,
|
| 1115 |
+
"special": false
|
| 1116 |
+
},
|
| 1117 |
+
"139": {
|
| 1118 |
+
"content": "▁▁",
|
| 1119 |
+
"lstrip": false,
|
| 1120 |
+
"normalized": false,
|
| 1121 |
+
"rstrip": false,
|
| 1122 |
+
"single_word": false,
|
| 1123 |
+
"special": false
|
| 1124 |
+
},
|
| 1125 |
+
"140": {
|
| 1126 |
+
"content": "▁▁▁",
|
| 1127 |
+
"lstrip": false,
|
| 1128 |
+
"normalized": false,
|
| 1129 |
+
"rstrip": false,
|
| 1130 |
+
"single_word": false,
|
| 1131 |
+
"special": false
|
| 1132 |
+
},
|
| 1133 |
+
"141": {
|
| 1134 |
+
"content": "▁▁▁▁",
|
| 1135 |
+
"lstrip": false,
|
| 1136 |
+
"normalized": false,
|
| 1137 |
+
"rstrip": false,
|
| 1138 |
+
"single_word": false,
|
| 1139 |
+
"special": false
|
| 1140 |
+
},
|
| 1141 |
+
"142": {
|
| 1142 |
+
"content": "▁▁▁▁▁",
|
| 1143 |
+
"lstrip": false,
|
| 1144 |
+
"normalized": false,
|
| 1145 |
+
"rstrip": false,
|
| 1146 |
+
"single_word": false,
|
| 1147 |
+
"special": false
|
| 1148 |
+
},
|
| 1149 |
+
"143": {
|
| 1150 |
+
"content": "▁▁▁▁▁▁",
|
| 1151 |
+
"lstrip": false,
|
| 1152 |
+
"normalized": false,
|
| 1153 |
+
"rstrip": false,
|
| 1154 |
+
"single_word": false,
|
| 1155 |
+
"special": false
|
| 1156 |
+
},
|
| 1157 |
+
"144": {
|
| 1158 |
+
"content": "▁▁▁▁▁▁▁",
|
| 1159 |
+
"lstrip": false,
|
| 1160 |
+
"normalized": false,
|
| 1161 |
+
"rstrip": false,
|
| 1162 |
+
"single_word": false,
|
| 1163 |
+
"special": false
|
| 1164 |
+
},
|
| 1165 |
+
"145": {
|
| 1166 |
+
"content": "▁▁▁▁▁▁▁▁",
|
| 1167 |
+
"lstrip": false,
|
| 1168 |
+
"normalized": false,
|
| 1169 |
+
"rstrip": false,
|
| 1170 |
+
"single_word": false,
|
| 1171 |
+
"special": false
|
| 1172 |
+
},
|
| 1173 |
+
"146": {
|
| 1174 |
+
"content": "▁▁▁▁▁▁▁▁▁",
|
| 1175 |
+
"lstrip": false,
|
| 1176 |
+
"normalized": false,
|
| 1177 |
+
"rstrip": false,
|
| 1178 |
+
"single_word": false,
|
| 1179 |
+
"special": false
|
| 1180 |
+
},
|
| 1181 |
+
"147": {
|
| 1182 |
+
"content": "▁▁▁▁▁▁▁▁▁▁",
|
| 1183 |
+
"lstrip": false,
|
| 1184 |
+
"normalized": false,
|
| 1185 |
+
"rstrip": false,
|
| 1186 |
+
"single_word": false,
|
| 1187 |
+
"special": false
|
| 1188 |
+
},
|
| 1189 |
+
"148": {
|
| 1190 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁",
|
| 1191 |
+
"lstrip": false,
|
| 1192 |
+
"normalized": false,
|
| 1193 |
+
"rstrip": false,
|
| 1194 |
+
"single_word": false,
|
| 1195 |
+
"special": false
|
| 1196 |
+
},
|
| 1197 |
+
"149": {
|
| 1198 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1199 |
+
"lstrip": false,
|
| 1200 |
+
"normalized": false,
|
| 1201 |
+
"rstrip": false,
|
| 1202 |
+
"single_word": false,
|
| 1203 |
+
"special": false
|
| 1204 |
+
},
|
| 1205 |
+
"150": {
|
| 1206 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1207 |
+
"lstrip": false,
|
| 1208 |
+
"normalized": false,
|
| 1209 |
+
"rstrip": false,
|
| 1210 |
+
"single_word": false,
|
| 1211 |
+
"special": false
|
| 1212 |
+
},
|
| 1213 |
+
"151": {
|
| 1214 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1215 |
+
"lstrip": false,
|
| 1216 |
+
"normalized": false,
|
| 1217 |
+
"rstrip": false,
|
| 1218 |
+
"single_word": false,
|
| 1219 |
+
"special": false
|
| 1220 |
+
},
|
| 1221 |
+
"152": {
|
| 1222 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1223 |
+
"lstrip": false,
|
| 1224 |
+
"normalized": false,
|
| 1225 |
+
"rstrip": false,
|
| 1226 |
+
"single_word": false,
|
| 1227 |
+
"special": false
|
| 1228 |
+
},
|
| 1229 |
+
"153": {
|
| 1230 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1231 |
+
"lstrip": false,
|
| 1232 |
+
"normalized": false,
|
| 1233 |
+
"rstrip": false,
|
| 1234 |
+
"single_word": false,
|
| 1235 |
+
"special": false
|
| 1236 |
+
},
|
| 1237 |
+
"154": {
|
| 1238 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1239 |
+
"lstrip": false,
|
| 1240 |
+
"normalized": false,
|
| 1241 |
+
"rstrip": false,
|
| 1242 |
+
"single_word": false,
|
| 1243 |
+
"special": false
|
| 1244 |
+
},
|
| 1245 |
+
"155": {
|
| 1246 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1247 |
+
"lstrip": false,
|
| 1248 |
+
"normalized": false,
|
| 1249 |
+
"rstrip": false,
|
| 1250 |
+
"single_word": false,
|
| 1251 |
+
"special": false
|
| 1252 |
+
},
|
| 1253 |
+
"156": {
|
| 1254 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1255 |
+
"lstrip": false,
|
| 1256 |
+
"normalized": false,
|
| 1257 |
+
"rstrip": false,
|
| 1258 |
+
"single_word": false,
|
| 1259 |
+
"special": false
|
| 1260 |
+
},
|
| 1261 |
+
"157": {
|
| 1262 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1263 |
+
"lstrip": false,
|
| 1264 |
+
"normalized": false,
|
| 1265 |
+
"rstrip": false,
|
| 1266 |
+
"single_word": false,
|
| 1267 |
+
"special": false
|
| 1268 |
+
},
|
| 1269 |
+
"158": {
|
| 1270 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1271 |
+
"lstrip": false,
|
| 1272 |
+
"normalized": false,
|
| 1273 |
+
"rstrip": false,
|
| 1274 |
+
"single_word": false,
|
| 1275 |
+
"special": false
|
| 1276 |
+
},
|
| 1277 |
+
"159": {
|
| 1278 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1279 |
+
"lstrip": false,
|
| 1280 |
+
"normalized": false,
|
| 1281 |
+
"rstrip": false,
|
| 1282 |
+
"single_word": false,
|
| 1283 |
+
"special": false
|
| 1284 |
+
},
|
| 1285 |
+
"160": {
|
| 1286 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1287 |
+
"lstrip": false,
|
| 1288 |
+
"normalized": false,
|
| 1289 |
+
"rstrip": false,
|
| 1290 |
+
"single_word": false,
|
| 1291 |
+
"special": false
|
| 1292 |
+
},
|
| 1293 |
+
"161": {
|
| 1294 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1295 |
+
"lstrip": false,
|
| 1296 |
+
"normalized": false,
|
| 1297 |
+
"rstrip": false,
|
| 1298 |
+
"single_word": false,
|
| 1299 |
+
"special": false
|
| 1300 |
+
},
|
| 1301 |
+
"162": {
|
| 1302 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1303 |
+
"lstrip": false,
|
| 1304 |
+
"normalized": false,
|
| 1305 |
+
"rstrip": false,
|
| 1306 |
+
"single_word": false,
|
| 1307 |
+
"special": false
|
| 1308 |
+
},
|
| 1309 |
+
"163": {
|
| 1310 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1311 |
+
"lstrip": false,
|
| 1312 |
+
"normalized": false,
|
| 1313 |
+
"rstrip": false,
|
| 1314 |
+
"single_word": false,
|
| 1315 |
+
"special": false
|
| 1316 |
+
},
|
| 1317 |
+
"164": {
|
| 1318 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1319 |
+
"lstrip": false,
|
| 1320 |
+
"normalized": false,
|
| 1321 |
+
"rstrip": false,
|
| 1322 |
+
"single_word": false,
|
| 1323 |
+
"special": false
|
| 1324 |
+
},
|
| 1325 |
+
"165": {
|
| 1326 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1327 |
+
"lstrip": false,
|
| 1328 |
+
"normalized": false,
|
| 1329 |
+
"rstrip": false,
|
| 1330 |
+
"single_word": false,
|
| 1331 |
+
"special": false
|
| 1332 |
+
},
|
| 1333 |
+
"166": {
|
| 1334 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1335 |
+
"lstrip": false,
|
| 1336 |
+
"normalized": false,
|
| 1337 |
+
"rstrip": false,
|
| 1338 |
+
"single_word": false,
|
| 1339 |
+
"special": false
|
| 1340 |
+
},
|
| 1341 |
+
"167": {
|
| 1342 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1343 |
+
"lstrip": false,
|
| 1344 |
+
"normalized": false,
|
| 1345 |
+
"rstrip": false,
|
| 1346 |
+
"single_word": false,
|
| 1347 |
+
"special": false
|
| 1348 |
+
},
|
| 1349 |
+
"168": {
|
| 1350 |
+
"content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
|
| 1351 |
+
"lstrip": false,
|
| 1352 |
+
"normalized": false,
|
| 1353 |
+
"rstrip": false,
|
| 1354 |
+
"single_word": false,
|
| 1355 |
+
"special": false
|
| 1356 |
+
},
|
| 1357 |
+
"169": {
|
| 1358 |
+
"content": "<table>",
|
| 1359 |
+
"lstrip": false,
|
| 1360 |
+
"normalized": false,
|
| 1361 |
+
"rstrip": false,
|
| 1362 |
+
"single_word": false,
|
| 1363 |
+
"special": false
|
| 1364 |
+
},
|
| 1365 |
+
"170": {
|
| 1366 |
+
"content": "<caption>",
|
| 1367 |
+
"lstrip": false,
|
| 1368 |
+
"normalized": false,
|
| 1369 |
+
"rstrip": false,
|
| 1370 |
+
"single_word": false,
|
| 1371 |
+
"special": false
|
| 1372 |
+
},
|
| 1373 |
+
"171": {
|
| 1374 |
+
"content": "<thead>",
|
| 1375 |
+
"lstrip": false,
|
| 1376 |
+
"normalized": false,
|
| 1377 |
+
"rstrip": false,
|
| 1378 |
+
"single_word": false,
|
| 1379 |
+
"special": false
|
| 1380 |
+
},
|
| 1381 |
+
"172": {
|
| 1382 |
+
"content": "<tbody>",
|
| 1383 |
+
"lstrip": false,
|
| 1384 |
+
"normalized": false,
|
| 1385 |
+
"rstrip": false,
|
| 1386 |
+
"single_word": false,
|
| 1387 |
+
"special": false
|
| 1388 |
+
},
|
| 1389 |
+
"173": {
|
| 1390 |
+
"content": "<tfoot>",
|
| 1391 |
+
"lstrip": false,
|
| 1392 |
+
"normalized": false,
|
| 1393 |
+
"rstrip": false,
|
| 1394 |
+
"single_word": false,
|
| 1395 |
+
"special": false
|
| 1396 |
+
},
|
| 1397 |
+
"174": {
|
| 1398 |
+
"content": "<tr>",
|
| 1399 |
+
"lstrip": false,
|
| 1400 |
+
"normalized": false,
|
| 1401 |
+
"rstrip": false,
|
| 1402 |
+
"single_word": false,
|
| 1403 |
+
"special": false
|
| 1404 |
+
},
|
| 1405 |
+
"175": {
|
| 1406 |
+
"content": "<th>",
|
| 1407 |
+
"lstrip": false,
|
| 1408 |
+
"normalized": false,
|
| 1409 |
+
"rstrip": false,
|
| 1410 |
+
"single_word": false,
|
| 1411 |
+
"special": false
|
| 1412 |
+
},
|
| 1413 |
+
"176": {
|
| 1414 |
+
"content": "<td>",
|
| 1415 |
+
"lstrip": false,
|
| 1416 |
+
"normalized": false,
|
| 1417 |
+
"rstrip": false,
|
| 1418 |
+
"single_word": false,
|
| 1419 |
+
"special": false
|
| 1420 |
+
},
|
| 1421 |
+
"177": {
|
| 1422 |
+
"content": "</table>",
|
| 1423 |
+
"lstrip": false,
|
| 1424 |
+
"normalized": false,
|
| 1425 |
+
"rstrip": false,
|
| 1426 |
+
"single_word": false,
|
| 1427 |
+
"special": false
|
| 1428 |
+
},
|
| 1429 |
+
"178": {
|
| 1430 |
+
"content": "</caption>",
|
| 1431 |
+
"lstrip": false,
|
| 1432 |
+
"normalized": false,
|
| 1433 |
+
"rstrip": false,
|
| 1434 |
+
"single_word": false,
|
| 1435 |
+
"special": false
|
| 1436 |
+
},
|
| 1437 |
+
"179": {
|
| 1438 |
+
"content": "</thead>",
|
| 1439 |
+
"lstrip": false,
|
| 1440 |
+
"normalized": false,
|
| 1441 |
+
"rstrip": false,
|
| 1442 |
+
"single_word": false,
|
| 1443 |
+
"special": false
|
| 1444 |
+
},
|
| 1445 |
+
"180": {
|
| 1446 |
+
"content": "</tbody>",
|
| 1447 |
+
"lstrip": false,
|
| 1448 |
+
"normalized": false,
|
| 1449 |
+
"rstrip": false,
|
| 1450 |
+
"single_word": false,
|
| 1451 |
+
"special": false
|
| 1452 |
+
},
|
| 1453 |
+
"181": {
|
| 1454 |
+
"content": "</tfoot>",
|
| 1455 |
+
"lstrip": false,
|
| 1456 |
+
"normalized": false,
|
| 1457 |
+
"rstrip": false,
|
| 1458 |
+
"single_word": false,
|
| 1459 |
+
"special": false
|
| 1460 |
+
},
|
| 1461 |
+
"182": {
|
| 1462 |
+
"content": "</tr>",
|
| 1463 |
+
"lstrip": false,
|
| 1464 |
+
"normalized": false,
|
| 1465 |
+
"rstrip": false,
|
| 1466 |
+
"single_word": false,
|
| 1467 |
+
"special": false
|
| 1468 |
+
},
|
| 1469 |
+
"183": {
|
| 1470 |
+
"content": "</th>",
|
| 1471 |
+
"lstrip": false,
|
| 1472 |
+
"normalized": false,
|
| 1473 |
+
"rstrip": false,
|
| 1474 |
+
"single_word": false,
|
| 1475 |
+
"special": false
|
| 1476 |
+
},
|
| 1477 |
+
"184": {
|
| 1478 |
+
"content": "</td>",
|
| 1479 |
+
"lstrip": false,
|
| 1480 |
+
"normalized": false,
|
| 1481 |
+
"rstrip": false,
|
| 1482 |
+
"single_word": false,
|
| 1483 |
+
"special": false
|
| 1484 |
+
},
|
| 1485 |
+
"185": {
|
| 1486 |
+
"content": "<h1>",
|
| 1487 |
+
"lstrip": false,
|
| 1488 |
+
"normalized": false,
|
| 1489 |
+
"rstrip": false,
|
| 1490 |
+
"single_word": false,
|
| 1491 |
+
"special": false
|
| 1492 |
+
},
|
| 1493 |
+
"186": {
|
| 1494 |
+
"content": "<h2>",
|
| 1495 |
+
"lstrip": false,
|
| 1496 |
+
"normalized": false,
|
| 1497 |
+
"rstrip": false,
|
| 1498 |
+
"single_word": false,
|
| 1499 |
+
"special": false
|
| 1500 |
+
},
|
| 1501 |
+
"187": {
|
| 1502 |
+
"content": "<h3>",
|
| 1503 |
+
"lstrip": false,
|
| 1504 |
+
"normalized": false,
|
| 1505 |
+
"rstrip": false,
|
| 1506 |
+
"single_word": false,
|
| 1507 |
+
"special": false
|
| 1508 |
+
},
|
| 1509 |
+
"188": {
|
| 1510 |
+
"content": "<h4>",
|
| 1511 |
+
"lstrip": false,
|
| 1512 |
+
"normalized": false,
|
| 1513 |
+
"rstrip": false,
|
| 1514 |
+
"single_word": false,
|
| 1515 |
+
"special": false
|
| 1516 |
+
},
|
| 1517 |
+
"189": {
|
| 1518 |
+
"content": "<h5>",
|
| 1519 |
+
"lstrip": false,
|
| 1520 |
+
"normalized": false,
|
| 1521 |
+
"rstrip": false,
|
| 1522 |
+
"single_word": false,
|
| 1523 |
+
"special": false
|
| 1524 |
+
},
|
| 1525 |
+
"190": {
|
| 1526 |
+
"content": "<h6>",
|
| 1527 |
+
"lstrip": false,
|
| 1528 |
+
"normalized": false,
|
| 1529 |
+
"rstrip": false,
|
| 1530 |
+
"single_word": false,
|
| 1531 |
+
"special": false
|
| 1532 |
+
},
|
| 1533 |
+
"191": {
|
| 1534 |
+
"content": "<blockquote>",
|
| 1535 |
+
"lstrip": false,
|
| 1536 |
+
"normalized": false,
|
| 1537 |
+
"rstrip": false,
|
| 1538 |
+
"single_word": false,
|
| 1539 |
+
"special": false
|
| 1540 |
+
},
|
| 1541 |
+
"192": {
|
| 1542 |
+
"content": "</h1>",
|
| 1543 |
+
"lstrip": false,
|
| 1544 |
+
"normalized": false,
|
| 1545 |
+
"rstrip": false,
|
| 1546 |
+
"single_word": false,
|
| 1547 |
+
"special": false
|
| 1548 |
+
},
|
| 1549 |
+
"193": {
|
| 1550 |
+
"content": "</h2>",
|
| 1551 |
+
"lstrip": false,
|
| 1552 |
+
"normalized": false,
|
| 1553 |
+
"rstrip": false,
|
| 1554 |
+
"single_word": false,
|
| 1555 |
+
"special": false
|
| 1556 |
+
},
|
| 1557 |
+
"194": {
|
| 1558 |
+
"content": "</h3>",
|
| 1559 |
+
"lstrip": false,
|
| 1560 |
+
"normalized": false,
|
| 1561 |
+
"rstrip": false,
|
| 1562 |
+
"single_word": false,
|
| 1563 |
+
"special": false
|
| 1564 |
+
},
|
| 1565 |
+
"195": {
|
| 1566 |
+
"content": "</h4>",
|
| 1567 |
+
"lstrip": false,
|
| 1568 |
+
"normalized": false,
|
| 1569 |
+
"rstrip": false,
|
| 1570 |
+
"single_word": false,
|
| 1571 |
+
"special": false
|
| 1572 |
+
},
|
| 1573 |
+
"196": {
|
| 1574 |
+
"content": "</h5>",
|
| 1575 |
+
"lstrip": false,
|
| 1576 |
+
"normalized": false,
|
| 1577 |
+
"rstrip": false,
|
| 1578 |
+
"single_word": false,
|
| 1579 |
+
"special": false
|
| 1580 |
+
},
|
| 1581 |
+
"197": {
|
| 1582 |
+
"content": "</h6>",
|
| 1583 |
+
"lstrip": false,
|
| 1584 |
+
"normalized": false,
|
| 1585 |
+
"rstrip": false,
|
| 1586 |
+
"single_word": false,
|
| 1587 |
+
"special": false
|
| 1588 |
+
},
|
| 1589 |
+
"198": {
|
| 1590 |
+
"content": "</blockquote>",
|
| 1591 |
+
"lstrip": false,
|
| 1592 |
+
"normalized": false,
|
| 1593 |
+
"rstrip": false,
|
| 1594 |
+
"single_word": false,
|
| 1595 |
+
"special": false
|
| 1596 |
+
},
|
| 1597 |
+
"199": {
|
| 1598 |
+
"content": "<strong>",
|
| 1599 |
+
"lstrip": false,
|
| 1600 |
+
"normalized": false,
|
| 1601 |
+
"rstrip": false,
|
| 1602 |
+
"single_word": false,
|
| 1603 |
+
"special": false
|
| 1604 |
+
},
|
| 1605 |
+
"200": {
|
| 1606 |
+
"content": "<em>",
|
| 1607 |
+
"lstrip": false,
|
| 1608 |
+
"normalized": false,
|
| 1609 |
+
"rstrip": false,
|
| 1610 |
+
"single_word": false,
|
| 1611 |
+
"special": false
|
| 1612 |
+
},
|
| 1613 |
+
"201": {
|
| 1614 |
+
"content": "<b>",
|
| 1615 |
+
"lstrip": false,
|
| 1616 |
+
"normalized": false,
|
| 1617 |
+
"rstrip": false,
|
| 1618 |
+
"single_word": false,
|
| 1619 |
+
"special": false
|
| 1620 |
+
},
|
| 1621 |
+
"202": {
|
| 1622 |
+
"content": "<i>",
|
| 1623 |
+
"lstrip": false,
|
| 1624 |
+
"normalized": false,
|
| 1625 |
+
"rstrip": false,
|
| 1626 |
+
"single_word": false,
|
| 1627 |
+
"special": false
|
| 1628 |
+
},
|
| 1629 |
+
"203": {
|
| 1630 |
+
"content": "<u>",
|
| 1631 |
+
"lstrip": false,
|
| 1632 |
+
"normalized": false,
|
| 1633 |
+
"rstrip": false,
|
| 1634 |
+
"single_word": false,
|
| 1635 |
+
"special": false
|
| 1636 |
+
},
|
| 1637 |
+
"204": {
|
| 1638 |
+
"content": "<s>",
|
| 1639 |
+
"lstrip": false,
|
| 1640 |
+
"normalized": false,
|
| 1641 |
+
"rstrip": false,
|
| 1642 |
+
"single_word": false,
|
| 1643 |
+
"special": false
|
| 1644 |
+
},
|
| 1645 |
+
"205": {
|
| 1646 |
+
"content": "<sub>",
|
| 1647 |
+
"lstrip": false,
|
| 1648 |
+
"normalized": false,
|
| 1649 |
+
"rstrip": false,
|
| 1650 |
+
"single_word": false,
|
| 1651 |
+
"special": false
|
| 1652 |
+
},
|
| 1653 |
+
"206": {
|
| 1654 |
+
"content": "<sup>",
|
| 1655 |
+
"lstrip": false,
|
| 1656 |
+
"normalized": false,
|
| 1657 |
+
"rstrip": false,
|
| 1658 |
+
"single_word": false,
|
| 1659 |
+
"special": false
|
| 1660 |
+
},
|
| 1661 |
+
"207": {
|
| 1662 |
+
"content": "<code>",
|
| 1663 |
+
"lstrip": false,
|
| 1664 |
+
"normalized": false,
|
| 1665 |
+
"rstrip": false,
|
| 1666 |
+
"single_word": false,
|
| 1667 |
+
"special": false
|
| 1668 |
+
},
|
| 1669 |
+
"208": {
|
| 1670 |
+
"content": "</strong>",
|
| 1671 |
+
"lstrip": false,
|
| 1672 |
+
"normalized": false,
|
| 1673 |
+
"rstrip": false,
|
| 1674 |
+
"single_word": false,
|
| 1675 |
+
"special": false
|
| 1676 |
+
},
|
| 1677 |
+
"209": {
|
| 1678 |
+
"content": "</em>",
|
| 1679 |
+
"lstrip": false,
|
| 1680 |
+
"normalized": false,
|
| 1681 |
+
"rstrip": false,
|
| 1682 |
+
"single_word": false,
|
| 1683 |
+
"special": false
|
| 1684 |
+
},
|
| 1685 |
+
"210": {
|
| 1686 |
+
"content": "</b>",
|
| 1687 |
+
"lstrip": false,
|
| 1688 |
+
"normalized": false,
|
| 1689 |
+
"rstrip": false,
|
| 1690 |
+
"single_word": false,
|
| 1691 |
+
"special": false
|
| 1692 |
+
},
|
| 1693 |
+
"211": {
|
| 1694 |
+
"content": "</i>",
|
| 1695 |
+
"lstrip": false,
|
| 1696 |
+
"normalized": false,
|
| 1697 |
+
"rstrip": false,
|
| 1698 |
+
"single_word": false,
|
| 1699 |
+
"special": false
|
| 1700 |
+
},
|
| 1701 |
+
"212": {
|
| 1702 |
+
"content": "</u>",
|
| 1703 |
+
"lstrip": false,
|
| 1704 |
+
"normalized": false,
|
| 1705 |
+
"rstrip": false,
|
| 1706 |
+
"single_word": false,
|
| 1707 |
+
"special": false
|
| 1708 |
+
},
|
| 1709 |
+
"213": {
|
| 1710 |
+
"content": "</s>",
|
| 1711 |
+
"lstrip": false,
|
| 1712 |
+
"normalized": false,
|
| 1713 |
+
"rstrip": false,
|
| 1714 |
+
"single_word": false,
|
| 1715 |
+
"special": false
|
| 1716 |
+
},
|
| 1717 |
+
"214": {
|
| 1718 |
+
"content": "</sub>",
|
| 1719 |
+
"lstrip": false,
|
| 1720 |
+
"normalized": false,
|
| 1721 |
+
"rstrip": false,
|
| 1722 |
+
"single_word": false,
|
| 1723 |
+
"special": false
|
| 1724 |
+
},
|
| 1725 |
+
"215": {
|
| 1726 |
+
"content": "</sup>",
|
| 1727 |
+
"lstrip": false,
|
| 1728 |
+
"normalized": false,
|
| 1729 |
+
"rstrip": false,
|
| 1730 |
+
"single_word": false,
|
| 1731 |
+
"special": false
|
| 1732 |
+
},
|
| 1733 |
+
"216": {
|
| 1734 |
+
"content": "</code>",
|
| 1735 |
+
"lstrip": false,
|
| 1736 |
+
"normalized": false,
|
| 1737 |
+
"rstrip": false,
|
| 1738 |
+
"single_word": false,
|
| 1739 |
+
"special": false
|
| 1740 |
+
}
|
| 1741 |
+
},
|
| 1742 |
+
"additional_special_tokens": [
|
| 1743 |
+
"<start_of_turn>",
|
| 1744 |
+
"<end_of_turn>"
|
| 1745 |
+
],
|
| 1746 |
+
"bos_token": "<bos>",
|
| 1747 |
+
"chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
|
| 1748 |
+
"clean_up_tokenization_spaces": false,
|
| 1749 |
+
"eos_token": "<eos>",
|
| 1750 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 1751 |
+
"pad_token": "<pad>",
|
| 1752 |
+
"sp_model_kwargs": {},
|
| 1753 |
+
"spaces_between_special_tokens": false,
|
| 1754 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 1755 |
+
"unk_token": "<unk>",
|
| 1756 |
+
"use_default_system_prompt": false
|
| 1757 |
+
}
|