Upload checkpoints
Browse files- adapter_config.json +28 -0
- adapter_model.safetensors +3 -0
- config.json +41 -0
- non_lora_trainables.bin +3 -0
- trainer_state.json +0 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "susnato/phi-2",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 256,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 128,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"fc2",
|
| 23 |
+
"dense",
|
| 24 |
+
"fc1",
|
| 25 |
+
"query_key_value"
|
| 26 |
+
],
|
| 27 |
+
"task_type": "CAUSAL_LM"
|
| 28 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ed26367984c3d2a7996ca30cdc45645ca80d37007001252dd6049b7f3bd1f08
|
| 3 |
+
size 335579120
|
config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "BAAI/bunny-phi-2-siglip-lora",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"PhiForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "gelu_new",
|
| 12 |
+
"hidden_size": 2560,
|
| 13 |
+
"image_aspect_ratio": "pad",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 10240,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"mm_hidden_size": 1152,
|
| 19 |
+
"mm_projector_lr": 2e-05,
|
| 20 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 21 |
+
"mm_vision_select_feature": "patch",
|
| 22 |
+
"mm_vision_tower": "google/siglip-so400m-patch14-384",
|
| 23 |
+
"model_type": "bunny-phi",
|
| 24 |
+
"num_attention_heads": 32,
|
| 25 |
+
"num_hidden_layers": 32,
|
| 26 |
+
"partial_rotary_factor": 0.4,
|
| 27 |
+
"pretraining_tp": 1,
|
| 28 |
+
"qk_layernorm": false,
|
| 29 |
+
"resid_pdrop": 0.1,
|
| 30 |
+
"rope_scaling": null,
|
| 31 |
+
"rope_theta": 10000.0,
|
| 32 |
+
"tie_word_embeddings": false,
|
| 33 |
+
"tokenizer_model_max_length": 2048,
|
| 34 |
+
"tokenizer_padding_side": "right",
|
| 35 |
+
"torch_dtype": "float16",
|
| 36 |
+
"transformers_version": "4.36.2",
|
| 37 |
+
"tune_mm_mlp_adapter": false,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"use_mm_proj": true,
|
| 40 |
+
"vocab_size": 51200
|
| 41 |
+
}
|
non_lora_trainables.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9159cdbf2ec303e5929e7b46561607ebaecc337fcde60670351a8d047fb2b5b
|
| 3 |
+
size 19017904
|
trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|