Training in progress, epoch 1
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +4 -0
- README.md +73 -0
- added_tokens.json +3 -0
- all_results.json +26 -0
- eval_results.json +12 -0
- generated_predictions.csv +0 -0
- predict_results.json +11 -0
- prefix_tuning_par_bn_zwj/adapter_config.json +59 -0
- prefix_tuning_par_bn_zwj/head_config.json +15 -0
- prefix_tuning_par_bn_zwj/pytorch_adapter.bin +3 -0
- prefix_tuning_par_bn_zwj/pytorch_model_head.bin +3 -0
- prefix_tuning_par_bn_zwj_hiru/adapter_config.json +59 -0
- prefix_tuning_par_bn_zwj_hiru/head_config.json +15 -0
- prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin +3 -0
- prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin +3 -0
- run-0/checkpoint-144/added_tokens.json +3 -0
- run-0/checkpoint-144/optimizer.pt +3 -0
- run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/adapter_config.json +59 -0
- run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/head_config.json +15 -0
- run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin +3 -0
- run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin +3 -0
- run-0/checkpoint-144/rng_state.pth +3 -0
- run-0/checkpoint-144/scheduler.pt +3 -0
- run-0/checkpoint-144/sentencepiece.bpe.model +3 -0
- run-0/checkpoint-144/special_tokens_map.json +43 -0
- run-0/checkpoint-144/tokenizer.json +3 -0
- run-0/checkpoint-144/tokenizer_config.json +292 -0
- run-0/checkpoint-144/trainer_state.json +47 -0
- run-0/checkpoint-144/training_args.bin +3 -0
- run-0/checkpoint-216/added_tokens.json +3 -0
- run-0/checkpoint-216/optimizer.pt +3 -0
- run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/adapter_config.json +59 -0
- run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/head_config.json +15 -0
- run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin +3 -0
- run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin +3 -0
- run-0/checkpoint-216/rng_state.pth +3 -0
- run-0/checkpoint-216/scheduler.pt +3 -0
- run-0/checkpoint-216/sentencepiece.bpe.model +3 -0
- run-0/checkpoint-216/special_tokens_map.json +43 -0
- run-0/checkpoint-216/tokenizer.json +3 -0
- run-0/checkpoint-216/tokenizer_config.json +292 -0
- run-0/checkpoint-216/trainer_state.json +59 -0
- run-0/checkpoint-216/training_args.bin +3 -0
- run-0/checkpoint-72/added_tokens.json +3 -0
- run-0/checkpoint-72/optimizer.pt +3 -0
- run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/adapter_config.json +59 -0
- run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/head_config.json +15 -0
- run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin +3 -0
- run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin +3 -0
- run-0/checkpoint-72/rng_state.pth +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,7 @@ 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 |
+
run-0/checkpoint-144/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
run-0/checkpoint-216/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
run-0/checkpoint-72/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: facebook/mbart-large-cc25
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
model-index:
|
| 6 |
+
- name: test
|
| 7 |
+
results: []
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 11 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 12 |
+
|
| 13 |
+
# test
|
| 14 |
+
|
| 15 |
+
This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on an unknown dataset.
|
| 16 |
+
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 3.2042
|
| 18 |
+
- Gen Len: 18.5844
|
| 19 |
+
- Rouge-1: 34.6893
|
| 20 |
+
- Rouge-2: 17.6676
|
| 21 |
+
- Rouge-l: 34.0228
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0001
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: polynomial
|
| 46 |
+
- lr_scheduler_warmup_steps: 1000
|
| 47 |
+
- num_epochs: 50
|
| 48 |
+
- label_smoothing_factor: 0.1
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Gen Len | Rouge-1 | Rouge-2 | Rouge-l |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:-------:|
|
| 54 |
+
| No log | 1.0 | 642 | 3.6377 | 22.68 | 23.5122 | 9.8274 | 22.6444 |
|
| 55 |
+
| No log | 2.0 | 1284 | 3.4690 | 24.1111 | 26.1024 | 11.7671 | 25.2784 |
|
| 56 |
+
| No log | 3.0 | 1926 | 3.2594 | 23.4756 | 30.4311 | 14.7772 | 29.7887 |
|
| 57 |
+
| No log | 4.0 | 2568 | 3.2171 | 21.3644 | 33.2453 | 16.8783 | 32.3833 |
|
| 58 |
+
| No log | 5.0 | 3210 | 3.1916 | 19.3667 | 33.1313 | 15.8622 | 32.4001 |
|
| 59 |
+
| No log | 6.0 | 3852 | 3.1366 | 21.8133 | 32.3945 | 16.0472 | 31.6275 |
|
| 60 |
+
| No log | 7.0 | 4494 | 3.1354 | 20.72 | 33.2417 | 16.5622 | 32.4859 |
|
| 61 |
+
| 3.4117 | 8.0 | 5136 | 3.1412 | 19.2578 | 34.706 | 17.7827 | 33.877 |
|
| 62 |
+
| 3.4117 | 9.0 | 5778 | 3.1538 | 18.1556 | 32.9258 | 17.2139 | 32.2229 |
|
| 63 |
+
| 3.4117 | 10.0 | 6420 | 3.1681 | 19.1867 | 34.1901 | 17.269 | 33.3286 |
|
| 64 |
+
| 3.4117 | 11.0 | 7062 | 3.1708 | 18.2978 | 34.4327 | 17.98 | 33.6562 |
|
| 65 |
+
| 3.4117 | 12.0 | 7704 | 3.2042 | 18.5844 | 34.6893 | 17.6676 | 34.0228 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.35.2
|
| 71 |
+
- Pytorch 2.2.1+cu121
|
| 72 |
+
- Datasets 2.18.0
|
| 73 |
+
- Tokenizers 0.15.2
|
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"": 250027
|
| 3 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"eval_gen_len": 20.72,
|
| 4 |
+
"eval_loss": 3.1354117393493652,
|
| 5 |
+
"eval_rouge-1": 33.2417,
|
| 6 |
+
"eval_rouge-2": 16.5622,
|
| 7 |
+
"eval_rouge-l": 32.4859,
|
| 8 |
+
"eval_runtime": 78.3041,
|
| 9 |
+
"eval_samples": 450,
|
| 10 |
+
"eval_samples_per_second": 5.747,
|
| 11 |
+
"eval_steps_per_second": 0.728,
|
| 12 |
+
"predict_gen_len": 20.9544,
|
| 13 |
+
"predict_loss": 3.1332778930664062,
|
| 14 |
+
"predict_rouge-1": 32.4791,
|
| 15 |
+
"predict_rouge-2": 16.6611,
|
| 16 |
+
"predict_rouge-l": 31.8555,
|
| 17 |
+
"predict_runtime": 73.72,
|
| 18 |
+
"predict_samples": 417,
|
| 19 |
+
"predict_samples_per_second": 5.657,
|
| 20 |
+
"predict_steps_per_second": 0.719,
|
| 21 |
+
"train_loss": 3.152212172653816,
|
| 22 |
+
"train_runtime": 2415.933,
|
| 23 |
+
"train_samples": 5133,
|
| 24 |
+
"train_samples_per_second": 106.232,
|
| 25 |
+
"train_steps_per_second": 13.287
|
| 26 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"eval_gen_len": 20.72,
|
| 4 |
+
"eval_loss": 3.1354117393493652,
|
| 5 |
+
"eval_rouge-1": 33.2417,
|
| 6 |
+
"eval_rouge-2": 16.5622,
|
| 7 |
+
"eval_rouge-l": 32.4859,
|
| 8 |
+
"eval_runtime": 78.3041,
|
| 9 |
+
"eval_samples": 450,
|
| 10 |
+
"eval_samples_per_second": 5.747,
|
| 11 |
+
"eval_steps_per_second": 0.728
|
| 12 |
+
}
|
generated_predictions.csv
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
predict_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"predict_gen_len": 20.9544,
|
| 3 |
+
"predict_loss": 3.1332778930664062,
|
| 4 |
+
"predict_rouge-1": 32.4791,
|
| 5 |
+
"predict_rouge-2": 16.6611,
|
| 6 |
+
"predict_rouge-l": 31.8555,
|
| 7 |
+
"predict_runtime": 73.72,
|
| 8 |
+
"predict_samples": 417,
|
| 9 |
+
"predict_samples_per_second": 5.657,
|
| 10 |
+
"predict_steps_per_second": 0.719
|
| 11 |
+
}
|
prefix_tuning_par_bn_zwj/adapter_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"architecture": "union",
|
| 4 |
+
"configs": [
|
| 5 |
+
{
|
| 6 |
+
"architecture": "prefix_tuning",
|
| 7 |
+
"bottleneck_size": 800,
|
| 8 |
+
"cross_prefix": true,
|
| 9 |
+
"dropout": 0.0,
|
| 10 |
+
"encoder_prefix": true,
|
| 11 |
+
"flat": false,
|
| 12 |
+
"leave_out": [],
|
| 13 |
+
"non_linearity": "tanh",
|
| 14 |
+
"prefix_length": 12,
|
| 15 |
+
"shared_gating": true,
|
| 16 |
+
"use_gating": false
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"adapter_residual_before_ln": false,
|
| 20 |
+
"cross_adapter": false,
|
| 21 |
+
"factorized_phm_W": true,
|
| 22 |
+
"factorized_phm_rule": false,
|
| 23 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 24 |
+
"init_weights": "mam_adapter",
|
| 25 |
+
"inv_adapter": null,
|
| 26 |
+
"inv_adapter_reduction_factor": null,
|
| 27 |
+
"is_parallel": true,
|
| 28 |
+
"learn_phm": true,
|
| 29 |
+
"leave_out": [],
|
| 30 |
+
"ln_after": false,
|
| 31 |
+
"ln_before": false,
|
| 32 |
+
"mh_adapter": false,
|
| 33 |
+
"non_linearity": "relu",
|
| 34 |
+
"original_ln_after": true,
|
| 35 |
+
"original_ln_before": false,
|
| 36 |
+
"output_adapter": true,
|
| 37 |
+
"phm_bias": true,
|
| 38 |
+
"phm_c_init": "normal",
|
| 39 |
+
"phm_dim": 4,
|
| 40 |
+
"phm_init_range": 0.0001,
|
| 41 |
+
"phm_layer": false,
|
| 42 |
+
"phm_rank": 1,
|
| 43 |
+
"reduction_factor": 2,
|
| 44 |
+
"residual_before_ln": true,
|
| 45 |
+
"scaling": 4.0,
|
| 46 |
+
"shared_W_phm": false,
|
| 47 |
+
"shared_phm_rule": true,
|
| 48 |
+
"use_gating": false
|
| 49 |
+
}
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"config_id": "4524cf66398dd5c5",
|
| 53 |
+
"hidden_size": 1024,
|
| 54 |
+
"model_class": "MBartForConditionalGeneration",
|
| 55 |
+
"model_name": "facebook/mbart-large-50",
|
| 56 |
+
"model_type": "mbart",
|
| 57 |
+
"name": "prefix_tuning_par_bn_zwj",
|
| 58 |
+
"version": "0.1.1"
|
| 59 |
+
}
|
prefix_tuning_par_bn_zwj/head_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1,
|
| 7 |
+
"LABEL_2": 2
|
| 8 |
+
},
|
| 9 |
+
"model_class": "MBartForConditionalGeneration",
|
| 10 |
+
"model_name": "facebook/mbart-large-50",
|
| 11 |
+
"model_type": "mbart",
|
| 12 |
+
"name": null,
|
| 13 |
+
"num_labels": 3,
|
| 14 |
+
"version": "0.1.1"
|
| 15 |
+
}
|
prefix_tuning_par_bn_zwj/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b746996b734c34ae450adcd4f59926714f2dc27d7d80c79417c3a1f59d13f6c6
|
| 3 |
+
size 347183084
|
prefix_tuning_par_bn_zwj/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:adcc43a196d39e07adb0d2ce5a89621565caf0187b28c81222c9bb08b8fa37f7
|
| 3 |
+
size 1025227034
|
prefix_tuning_par_bn_zwj_hiru/adapter_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"architecture": "union",
|
| 4 |
+
"configs": [
|
| 5 |
+
{
|
| 6 |
+
"architecture": "prefix_tuning",
|
| 7 |
+
"bottleneck_size": 800,
|
| 8 |
+
"cross_prefix": true,
|
| 9 |
+
"dropout": 0.0,
|
| 10 |
+
"encoder_prefix": true,
|
| 11 |
+
"flat": false,
|
| 12 |
+
"leave_out": [],
|
| 13 |
+
"non_linearity": "tanh",
|
| 14 |
+
"prefix_length": 9,
|
| 15 |
+
"shared_gating": true,
|
| 16 |
+
"use_gating": false
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"adapter_residual_before_ln": false,
|
| 20 |
+
"cross_adapter": false,
|
| 21 |
+
"factorized_phm_W": true,
|
| 22 |
+
"factorized_phm_rule": false,
|
| 23 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 24 |
+
"init_weights": "mam_adapter",
|
| 25 |
+
"inv_adapter": null,
|
| 26 |
+
"inv_adapter_reduction_factor": null,
|
| 27 |
+
"is_parallel": true,
|
| 28 |
+
"learn_phm": true,
|
| 29 |
+
"leave_out": [],
|
| 30 |
+
"ln_after": false,
|
| 31 |
+
"ln_before": false,
|
| 32 |
+
"mh_adapter": false,
|
| 33 |
+
"non_linearity": "relu",
|
| 34 |
+
"original_ln_after": true,
|
| 35 |
+
"original_ln_before": false,
|
| 36 |
+
"output_adapter": true,
|
| 37 |
+
"phm_bias": true,
|
| 38 |
+
"phm_c_init": "normal",
|
| 39 |
+
"phm_dim": 4,
|
| 40 |
+
"phm_init_range": 0.0001,
|
| 41 |
+
"phm_layer": false,
|
| 42 |
+
"phm_rank": 1,
|
| 43 |
+
"reduction_factor": 32,
|
| 44 |
+
"residual_before_ln": true,
|
| 45 |
+
"scaling": 4.0,
|
| 46 |
+
"shared_W_phm": false,
|
| 47 |
+
"shared_phm_rule": true,
|
| 48 |
+
"use_gating": false
|
| 49 |
+
}
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"config_id": "aae04a7675ab140e",
|
| 53 |
+
"hidden_size": 1024,
|
| 54 |
+
"model_class": "MBartForConditionalGeneration",
|
| 55 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 56 |
+
"model_type": "mbart",
|
| 57 |
+
"name": "prefix_tuning_par_bn_zwj_hiru",
|
| 58 |
+
"version": "0.1.1"
|
| 59 |
+
}
|
prefix_tuning_par_bn_zwj_hiru/head_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1,
|
| 7 |
+
"LABEL_2": 2
|
| 8 |
+
},
|
| 9 |
+
"model_class": "MBartForConditionalGeneration",
|
| 10 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 11 |
+
"model_type": "mbart",
|
| 12 |
+
"name": null,
|
| 13 |
+
"num_labels": 3,
|
| 14 |
+
"version": "0.1.1"
|
| 15 |
+
}
|
prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce61b4ba598b6aba57abe9937edb372e2c3e7b04f36d1a1a8e3d4dd67dfe6c3b
|
| 3 |
+
size 252731308
|
prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11d847a7c53a8cf4cbf1435e76e4bd7db89bde724cc5aaa97a7111e858f6d02a
|
| 3 |
+
size 1025116314
|
run-0/checkpoint-144/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"": 250027
|
| 3 |
+
}
|
run-0/checkpoint-144/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2242a0ff4ab6aac6c10058cd759e3bf67b772714ac061fd734b35979e5426f73
|
| 3 |
+
size 505741754
|
run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/adapter_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"architecture": "union",
|
| 4 |
+
"configs": [
|
| 5 |
+
{
|
| 6 |
+
"architecture": "prefix_tuning",
|
| 7 |
+
"bottleneck_size": 800,
|
| 8 |
+
"cross_prefix": true,
|
| 9 |
+
"dropout": 0.0,
|
| 10 |
+
"encoder_prefix": true,
|
| 11 |
+
"flat": false,
|
| 12 |
+
"leave_out": [],
|
| 13 |
+
"non_linearity": "tanh",
|
| 14 |
+
"prefix_length": 30,
|
| 15 |
+
"shared_gating": true,
|
| 16 |
+
"use_gating": false
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"adapter_residual_before_ln": false,
|
| 20 |
+
"cross_adapter": false,
|
| 21 |
+
"factorized_phm_W": true,
|
| 22 |
+
"factorized_phm_rule": false,
|
| 23 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 24 |
+
"init_weights": "mam_adapter",
|
| 25 |
+
"inv_adapter": null,
|
| 26 |
+
"inv_adapter_reduction_factor": null,
|
| 27 |
+
"is_parallel": true,
|
| 28 |
+
"learn_phm": true,
|
| 29 |
+
"leave_out": [],
|
| 30 |
+
"ln_after": false,
|
| 31 |
+
"ln_before": false,
|
| 32 |
+
"mh_adapter": false,
|
| 33 |
+
"non_linearity": "relu",
|
| 34 |
+
"original_ln_after": true,
|
| 35 |
+
"original_ln_before": false,
|
| 36 |
+
"output_adapter": true,
|
| 37 |
+
"phm_bias": true,
|
| 38 |
+
"phm_c_init": "normal",
|
| 39 |
+
"phm_dim": 4,
|
| 40 |
+
"phm_init_range": 0.0001,
|
| 41 |
+
"phm_layer": false,
|
| 42 |
+
"phm_rank": 1,
|
| 43 |
+
"reduction_factor": 32,
|
| 44 |
+
"residual_before_ln": true,
|
| 45 |
+
"scaling": 4.0,
|
| 46 |
+
"shared_W_phm": false,
|
| 47 |
+
"shared_phm_rule": true,
|
| 48 |
+
"use_gating": false
|
| 49 |
+
}
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"config_id": "19b1d26166c610bc",
|
| 53 |
+
"hidden_size": 1024,
|
| 54 |
+
"model_class": "MBartForConditionalGeneration",
|
| 55 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 56 |
+
"model_type": "mbart",
|
| 57 |
+
"name": "prefix_tuning_par_bn_zwj_hiru",
|
| 58 |
+
"version": "0.1.1"
|
| 59 |
+
}
|
run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/head_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1,
|
| 7 |
+
"LABEL_2": 2
|
| 8 |
+
},
|
| 9 |
+
"model_class": "MBartForConditionalGeneration",
|
| 10 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 11 |
+
"model_type": "mbart",
|
| 12 |
+
"name": null,
|
| 13 |
+
"num_labels": 3,
|
| 14 |
+
"version": "0.1.1"
|
| 15 |
+
}
|
run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be9b334a02eba51430328bf1cff1d8b15d20910389a5e385af57d5538b2ab1ac
|
| 3 |
+
size 252980268
|
run-0/checkpoint-144/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8cc995c049adbc360d7d9065243e7958f030aa58ac84e9b1b2b63c90a26e60d
|
| 3 |
+
size 1025116314
|
run-0/checkpoint-144/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57e944d28753d0a2cf694952ed56d0d053c9499de058df0e36f7d9ca216294c4
|
| 3 |
+
size 13990
|
run-0/checkpoint-144/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80903ef536160804a1bc9200a8738bd1b8818ddac913b6ae1dfcbc4ab5ae5f49
|
| 3 |
+
size 1064
|
run-0/checkpoint-144/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
run-0/checkpoint-144/special_tokens_map.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"ar_AR",
|
| 4 |
+
"cs_CZ",
|
| 5 |
+
"de_DE",
|
| 6 |
+
"en_XX",
|
| 7 |
+
"es_XX",
|
| 8 |
+
"et_EE",
|
| 9 |
+
"fi_FI",
|
| 10 |
+
"fr_XX",
|
| 11 |
+
"gu_IN",
|
| 12 |
+
"hi_IN",
|
| 13 |
+
"it_IT",
|
| 14 |
+
"ja_XX",
|
| 15 |
+
"kk_KZ",
|
| 16 |
+
"ko_KR",
|
| 17 |
+
"lt_LT",
|
| 18 |
+
"lv_LV",
|
| 19 |
+
"my_MM",
|
| 20 |
+
"ne_NP",
|
| 21 |
+
"nl_XX",
|
| 22 |
+
"ro_RO",
|
| 23 |
+
"ru_RU",
|
| 24 |
+
"si_LK",
|
| 25 |
+
"tr_TR",
|
| 26 |
+
"vi_VN",
|
| 27 |
+
"zh_CN",
|
| 28 |
+
""
|
| 29 |
+
],
|
| 30 |
+
"bos_token": "<s>",
|
| 31 |
+
"cls_token": "<s>",
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"mask_token": {
|
| 34 |
+
"content": "<mask>",
|
| 35 |
+
"lstrip": true,
|
| 36 |
+
"normalized": false,
|
| 37 |
+
"rstrip": false,
|
| 38 |
+
"single_word": false
|
| 39 |
+
},
|
| 40 |
+
"pad_token": "<pad>",
|
| 41 |
+
"sep_token": "</s>",
|
| 42 |
+
"unk_token": "<unk>"
|
| 43 |
+
}
|
run-0/checkpoint-144/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02c939cfc64478c07fb245694168d7d472437ccf9587b9d727958097e8c993c4
|
| 3 |
+
size 17088451
|
run-0/checkpoint-144/tokenizer_config.json
ADDED
|
@@ -0,0 +1,292 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "ar_AR",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": false,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": false,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": false,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": false,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": false,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": false,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": false,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": false,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": false,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": false,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": false,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": false,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": false,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": false,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "<mask>",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": false,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "",
|
| 245 |
+
"lstrip": false,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": false,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
}
|
| 251 |
+
},
|
| 252 |
+
"additional_special_tokens": [
|
| 253 |
+
"ar_AR",
|
| 254 |
+
"cs_CZ",
|
| 255 |
+
"de_DE",
|
| 256 |
+
"en_XX",
|
| 257 |
+
"es_XX",
|
| 258 |
+
"et_EE",
|
| 259 |
+
"fi_FI",
|
| 260 |
+
"fr_XX",
|
| 261 |
+
"gu_IN",
|
| 262 |
+
"hi_IN",
|
| 263 |
+
"it_IT",
|
| 264 |
+
"ja_XX",
|
| 265 |
+
"kk_KZ",
|
| 266 |
+
"ko_KR",
|
| 267 |
+
"lt_LT",
|
| 268 |
+
"lv_LV",
|
| 269 |
+
"my_MM",
|
| 270 |
+
"ne_NP",
|
| 271 |
+
"nl_XX",
|
| 272 |
+
"ro_RO",
|
| 273 |
+
"ru_RU",
|
| 274 |
+
"si_LK",
|
| 275 |
+
"tr_TR",
|
| 276 |
+
"vi_VN",
|
| 277 |
+
"zh_CN",
|
| 278 |
+
""
|
| 279 |
+
],
|
| 280 |
+
"bos_token": "<s>",
|
| 281 |
+
"clean_up_tokenization_spaces": true,
|
| 282 |
+
"cls_token": "<s>",
|
| 283 |
+
"eos_token": "</s>",
|
| 284 |
+
"mask_token": "<mask>",
|
| 285 |
+
"model_max_length": 1024,
|
| 286 |
+
"pad_token": "<pad>",
|
| 287 |
+
"sep_token": "</s>",
|
| 288 |
+
"src_lang": "si_LK",
|
| 289 |
+
"tgt_lang": "si_LK",
|
| 290 |
+
"tokenizer_class": "MBartTokenizer",
|
| 291 |
+
"unk_token": "<unk>"
|
| 292 |
+
}
|
run-0/checkpoint-144/trainer_state.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 11.73315715789795,
|
| 3 |
+
"best_model_checkpoint": "PEFT/adapters-lib/output/mam/dataset-5100/test/run-0/checkpoint-144",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 144,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_gen_len": 50.0,
|
| 14 |
+
"eval_loss": 12.573781967163086,
|
| 15 |
+
"eval_rouge-1": 1.0377,
|
| 16 |
+
"eval_rouge-2": 0.2759,
|
| 17 |
+
"eval_rouge-l": 1.0377,
|
| 18 |
+
"eval_runtime": 78.5593,
|
| 19 |
+
"eval_samples_per_second": 0.636,
|
| 20 |
+
"eval_steps_per_second": 0.089,
|
| 21 |
+
"step": 72
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 2.0,
|
| 25 |
+
"eval_gen_len": 29.04,
|
| 26 |
+
"eval_loss": 11.73315715789795,
|
| 27 |
+
"eval_rouge-1": 0.0,
|
| 28 |
+
"eval_rouge-2": 0.0,
|
| 29 |
+
"eval_rouge-l": 0.0,
|
| 30 |
+
"eval_runtime": 80.7187,
|
| 31 |
+
"eval_samples_per_second": 0.619,
|
| 32 |
+
"eval_steps_per_second": 0.087,
|
| 33 |
+
"step": 144
|
| 34 |
+
}
|
| 35 |
+
],
|
| 36 |
+
"logging_steps": 5000,
|
| 37 |
+
"max_steps": 648,
|
| 38 |
+
"num_train_epochs": 9,
|
| 39 |
+
"save_steps": 500,
|
| 40 |
+
"total_flos": 0,
|
| 41 |
+
"trial_name": null,
|
| 42 |
+
"trial_params": {
|
| 43 |
+
"learning_rate": 4.225170346079199e-05,
|
| 44 |
+
"num_train_epochs": 9,
|
| 45 |
+
"per_device_train_batch_size": 8
|
| 46 |
+
}
|
| 47 |
+
}
|
run-0/checkpoint-144/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bacf3eb9deeb9dd69f8bea35f04a260e1ce7c13d1950bd18cf5555081360a5f4
|
| 3 |
+
size 4792
|
run-0/checkpoint-216/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"": 250027
|
| 3 |
+
}
|
run-0/checkpoint-216/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e05d1c277025322a1da7f25525e4060c26cae573685e0ae5fa289f52fa71c33
|
| 3 |
+
size 505741754
|
run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/adapter_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"architecture": "union",
|
| 4 |
+
"configs": [
|
| 5 |
+
{
|
| 6 |
+
"architecture": "prefix_tuning",
|
| 7 |
+
"bottleneck_size": 800,
|
| 8 |
+
"cross_prefix": true,
|
| 9 |
+
"dropout": 0.0,
|
| 10 |
+
"encoder_prefix": true,
|
| 11 |
+
"flat": false,
|
| 12 |
+
"leave_out": [],
|
| 13 |
+
"non_linearity": "tanh",
|
| 14 |
+
"prefix_length": 30,
|
| 15 |
+
"shared_gating": true,
|
| 16 |
+
"use_gating": false
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"adapter_residual_before_ln": false,
|
| 20 |
+
"cross_adapter": false,
|
| 21 |
+
"factorized_phm_W": true,
|
| 22 |
+
"factorized_phm_rule": false,
|
| 23 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 24 |
+
"init_weights": "mam_adapter",
|
| 25 |
+
"inv_adapter": null,
|
| 26 |
+
"inv_adapter_reduction_factor": null,
|
| 27 |
+
"is_parallel": true,
|
| 28 |
+
"learn_phm": true,
|
| 29 |
+
"leave_out": [],
|
| 30 |
+
"ln_after": false,
|
| 31 |
+
"ln_before": false,
|
| 32 |
+
"mh_adapter": false,
|
| 33 |
+
"non_linearity": "relu",
|
| 34 |
+
"original_ln_after": true,
|
| 35 |
+
"original_ln_before": false,
|
| 36 |
+
"output_adapter": true,
|
| 37 |
+
"phm_bias": true,
|
| 38 |
+
"phm_c_init": "normal",
|
| 39 |
+
"phm_dim": 4,
|
| 40 |
+
"phm_init_range": 0.0001,
|
| 41 |
+
"phm_layer": false,
|
| 42 |
+
"phm_rank": 1,
|
| 43 |
+
"reduction_factor": 32,
|
| 44 |
+
"residual_before_ln": true,
|
| 45 |
+
"scaling": 4.0,
|
| 46 |
+
"shared_W_phm": false,
|
| 47 |
+
"shared_phm_rule": true,
|
| 48 |
+
"use_gating": false
|
| 49 |
+
}
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"config_id": "19b1d26166c610bc",
|
| 53 |
+
"hidden_size": 1024,
|
| 54 |
+
"model_class": "MBartForConditionalGeneration",
|
| 55 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 56 |
+
"model_type": "mbart",
|
| 57 |
+
"name": "prefix_tuning_par_bn_zwj_hiru",
|
| 58 |
+
"version": "0.1.1"
|
| 59 |
+
}
|
run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/head_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1,
|
| 7 |
+
"LABEL_2": 2
|
| 8 |
+
},
|
| 9 |
+
"model_class": "MBartForConditionalGeneration",
|
| 10 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 11 |
+
"model_type": "mbart",
|
| 12 |
+
"name": null,
|
| 13 |
+
"num_labels": 3,
|
| 14 |
+
"version": "0.1.1"
|
| 15 |
+
}
|
run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4edfc5b8a3a88dbf49a4cd06f5ae75488c0def85121910bb7586033db3d8abd4
|
| 3 |
+
size 252980268
|
run-0/checkpoint-216/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8cc995c049adbc360d7d9065243e7958f030aa58ac84e9b1b2b63c90a26e60d
|
| 3 |
+
size 1025116314
|
run-0/checkpoint-216/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75cf16e042ee4c688aac926458234c2fadd7857d3cca82d4497845f642abf1ef
|
| 3 |
+
size 13990
|
run-0/checkpoint-216/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7523984b48606d4670dadcc4330233a22965af511db15c00b4cb4bda0f3bd90d
|
| 3 |
+
size 1064
|
run-0/checkpoint-216/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
run-0/checkpoint-216/special_tokens_map.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"ar_AR",
|
| 4 |
+
"cs_CZ",
|
| 5 |
+
"de_DE",
|
| 6 |
+
"en_XX",
|
| 7 |
+
"es_XX",
|
| 8 |
+
"et_EE",
|
| 9 |
+
"fi_FI",
|
| 10 |
+
"fr_XX",
|
| 11 |
+
"gu_IN",
|
| 12 |
+
"hi_IN",
|
| 13 |
+
"it_IT",
|
| 14 |
+
"ja_XX",
|
| 15 |
+
"kk_KZ",
|
| 16 |
+
"ko_KR",
|
| 17 |
+
"lt_LT",
|
| 18 |
+
"lv_LV",
|
| 19 |
+
"my_MM",
|
| 20 |
+
"ne_NP",
|
| 21 |
+
"nl_XX",
|
| 22 |
+
"ro_RO",
|
| 23 |
+
"ru_RU",
|
| 24 |
+
"si_LK",
|
| 25 |
+
"tr_TR",
|
| 26 |
+
"vi_VN",
|
| 27 |
+
"zh_CN",
|
| 28 |
+
""
|
| 29 |
+
],
|
| 30 |
+
"bos_token": "<s>",
|
| 31 |
+
"cls_token": "<s>",
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"mask_token": {
|
| 34 |
+
"content": "<mask>",
|
| 35 |
+
"lstrip": true,
|
| 36 |
+
"normalized": false,
|
| 37 |
+
"rstrip": false,
|
| 38 |
+
"single_word": false
|
| 39 |
+
},
|
| 40 |
+
"pad_token": "<pad>",
|
| 41 |
+
"sep_token": "</s>",
|
| 42 |
+
"unk_token": "<unk>"
|
| 43 |
+
}
|
run-0/checkpoint-216/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02c939cfc64478c07fb245694168d7d472437ccf9587b9d727958097e8c993c4
|
| 3 |
+
size 17088451
|
run-0/checkpoint-216/tokenizer_config.json
ADDED
|
@@ -0,0 +1,292 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "ar_AR",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": false,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": false,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": false,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": false,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": false,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": false,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": false,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": false,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": false,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": false,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": false,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": false,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": false,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": false,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "<mask>",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": false,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "",
|
| 245 |
+
"lstrip": false,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": false,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
}
|
| 251 |
+
},
|
| 252 |
+
"additional_special_tokens": [
|
| 253 |
+
"ar_AR",
|
| 254 |
+
"cs_CZ",
|
| 255 |
+
"de_DE",
|
| 256 |
+
"en_XX",
|
| 257 |
+
"es_XX",
|
| 258 |
+
"et_EE",
|
| 259 |
+
"fi_FI",
|
| 260 |
+
"fr_XX",
|
| 261 |
+
"gu_IN",
|
| 262 |
+
"hi_IN",
|
| 263 |
+
"it_IT",
|
| 264 |
+
"ja_XX",
|
| 265 |
+
"kk_KZ",
|
| 266 |
+
"ko_KR",
|
| 267 |
+
"lt_LT",
|
| 268 |
+
"lv_LV",
|
| 269 |
+
"my_MM",
|
| 270 |
+
"ne_NP",
|
| 271 |
+
"nl_XX",
|
| 272 |
+
"ro_RO",
|
| 273 |
+
"ru_RU",
|
| 274 |
+
"si_LK",
|
| 275 |
+
"tr_TR",
|
| 276 |
+
"vi_VN",
|
| 277 |
+
"zh_CN",
|
| 278 |
+
""
|
| 279 |
+
],
|
| 280 |
+
"bos_token": "<s>",
|
| 281 |
+
"clean_up_tokenization_spaces": true,
|
| 282 |
+
"cls_token": "<s>",
|
| 283 |
+
"eos_token": "</s>",
|
| 284 |
+
"mask_token": "<mask>",
|
| 285 |
+
"model_max_length": 1024,
|
| 286 |
+
"pad_token": "<pad>",
|
| 287 |
+
"sep_token": "</s>",
|
| 288 |
+
"src_lang": "si_LK",
|
| 289 |
+
"tgt_lang": "si_LK",
|
| 290 |
+
"tokenizer_class": "MBartTokenizer",
|
| 291 |
+
"unk_token": "<unk>"
|
| 292 |
+
}
|
run-0/checkpoint-216/trainer_state.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 10.864657402038574,
|
| 3 |
+
"best_model_checkpoint": "PEFT/adapters-lib/output/mam/dataset-5100/test/run-0/checkpoint-216",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 216,
|
| 7 |
+
"is_hyper_param_search": true,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_gen_len": 50.0,
|
| 14 |
+
"eval_loss": 12.573781967163086,
|
| 15 |
+
"eval_rouge-1": 1.0377,
|
| 16 |
+
"eval_rouge-2": 0.2759,
|
| 17 |
+
"eval_rouge-l": 1.0377,
|
| 18 |
+
"eval_runtime": 78.5593,
|
| 19 |
+
"eval_samples_per_second": 0.636,
|
| 20 |
+
"eval_steps_per_second": 0.089,
|
| 21 |
+
"step": 72
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 2.0,
|
| 25 |
+
"eval_gen_len": 29.04,
|
| 26 |
+
"eval_loss": 11.73315715789795,
|
| 27 |
+
"eval_rouge-1": 0.0,
|
| 28 |
+
"eval_rouge-2": 0.0,
|
| 29 |
+
"eval_rouge-l": 0.0,
|
| 30 |
+
"eval_runtime": 80.7187,
|
| 31 |
+
"eval_samples_per_second": 0.619,
|
| 32 |
+
"eval_steps_per_second": 0.087,
|
| 33 |
+
"step": 144
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 3.0,
|
| 37 |
+
"eval_gen_len": 25.06,
|
| 38 |
+
"eval_loss": 10.864657402038574,
|
| 39 |
+
"eval_rouge-1": 0.0,
|
| 40 |
+
"eval_rouge-2": 0.0,
|
| 41 |
+
"eval_rouge-l": 0.0,
|
| 42 |
+
"eval_runtime": 249.5788,
|
| 43 |
+
"eval_samples_per_second": 0.2,
|
| 44 |
+
"eval_steps_per_second": 0.028,
|
| 45 |
+
"step": 216
|
| 46 |
+
}
|
| 47 |
+
],
|
| 48 |
+
"logging_steps": 5000,
|
| 49 |
+
"max_steps": 648,
|
| 50 |
+
"num_train_epochs": 9,
|
| 51 |
+
"save_steps": 500,
|
| 52 |
+
"total_flos": 0,
|
| 53 |
+
"trial_name": null,
|
| 54 |
+
"trial_params": {
|
| 55 |
+
"learning_rate": 4.225170346079199e-05,
|
| 56 |
+
"num_train_epochs": 9,
|
| 57 |
+
"per_device_train_batch_size": 8
|
| 58 |
+
}
|
| 59 |
+
}
|
run-0/checkpoint-216/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bacf3eb9deeb9dd69f8bea35f04a260e1ce7c13d1950bd18cf5555081360a5f4
|
| 3 |
+
size 4792
|
run-0/checkpoint-72/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"": 250027
|
| 3 |
+
}
|
run-0/checkpoint-72/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d84eb29e08642b3958cea1f6a3348c464484ab32fed5a4f84713db82adb8c696
|
| 3 |
+
size 505741754
|
run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/adapter_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": {
|
| 3 |
+
"architecture": "union",
|
| 4 |
+
"configs": [
|
| 5 |
+
{
|
| 6 |
+
"architecture": "prefix_tuning",
|
| 7 |
+
"bottleneck_size": 800,
|
| 8 |
+
"cross_prefix": true,
|
| 9 |
+
"dropout": 0.0,
|
| 10 |
+
"encoder_prefix": true,
|
| 11 |
+
"flat": false,
|
| 12 |
+
"leave_out": [],
|
| 13 |
+
"non_linearity": "tanh",
|
| 14 |
+
"prefix_length": 30,
|
| 15 |
+
"shared_gating": true,
|
| 16 |
+
"use_gating": false
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"adapter_residual_before_ln": false,
|
| 20 |
+
"cross_adapter": false,
|
| 21 |
+
"factorized_phm_W": true,
|
| 22 |
+
"factorized_phm_rule": false,
|
| 23 |
+
"hypercomplex_nonlinearity": "glorot-uniform",
|
| 24 |
+
"init_weights": "mam_adapter",
|
| 25 |
+
"inv_adapter": null,
|
| 26 |
+
"inv_adapter_reduction_factor": null,
|
| 27 |
+
"is_parallel": true,
|
| 28 |
+
"learn_phm": true,
|
| 29 |
+
"leave_out": [],
|
| 30 |
+
"ln_after": false,
|
| 31 |
+
"ln_before": false,
|
| 32 |
+
"mh_adapter": false,
|
| 33 |
+
"non_linearity": "relu",
|
| 34 |
+
"original_ln_after": true,
|
| 35 |
+
"original_ln_before": false,
|
| 36 |
+
"output_adapter": true,
|
| 37 |
+
"phm_bias": true,
|
| 38 |
+
"phm_c_init": "normal",
|
| 39 |
+
"phm_dim": 4,
|
| 40 |
+
"phm_init_range": 0.0001,
|
| 41 |
+
"phm_layer": false,
|
| 42 |
+
"phm_rank": 1,
|
| 43 |
+
"reduction_factor": 32,
|
| 44 |
+
"residual_before_ln": true,
|
| 45 |
+
"scaling": 4.0,
|
| 46 |
+
"shared_W_phm": false,
|
| 47 |
+
"shared_phm_rule": true,
|
| 48 |
+
"use_gating": false
|
| 49 |
+
}
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"config_id": "19b1d26166c610bc",
|
| 53 |
+
"hidden_size": 1024,
|
| 54 |
+
"model_class": "MBartForConditionalGeneration",
|
| 55 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 56 |
+
"model_type": "mbart",
|
| 57 |
+
"name": "prefix_tuning_par_bn_zwj_hiru",
|
| 58 |
+
"version": "0.1.1"
|
| 59 |
+
}
|
run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/head_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"config": null,
|
| 3 |
+
"hidden_size": 1024,
|
| 4 |
+
"label2id": {
|
| 5 |
+
"LABEL_0": 0,
|
| 6 |
+
"LABEL_1": 1,
|
| 7 |
+
"LABEL_2": 2
|
| 8 |
+
},
|
| 9 |
+
"model_class": "MBartForConditionalGeneration",
|
| 10 |
+
"model_name": "facebook/mbart-large-cc25",
|
| 11 |
+
"model_type": "mbart",
|
| 12 |
+
"name": null,
|
| 13 |
+
"num_labels": 3,
|
| 14 |
+
"version": "0.1.1"
|
| 15 |
+
}
|
run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/pytorch_adapter.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81a6807b690c494c69ad00898c6b0f0c6473e02982b79647de2f870c438dd0c1
|
| 3 |
+
size 252980268
|
run-0/checkpoint-72/prefix_tuning_par_bn_zwj_hiru/pytorch_model_head.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8cc995c049adbc360d7d9065243e7958f030aa58ac84e9b1b2b63c90a26e60d
|
| 3 |
+
size 1025116314
|
run-0/checkpoint-72/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32af8313bd2c49e427fbd3fcf3926e25d03ea3174b1f4c67659d3708737943f1
|
| 3 |
+
size 13990
|