Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- README.md +87 -0
- added_tokens.json +59 -0
- all_results.json +22 -0
- checkpoint-60000/added_tokens.json +59 -0
- checkpoint-60000/config.json +59 -0
- checkpoint-60000/generation_config.json +12 -0
- checkpoint-60000/optimizer.pt +3 -0
- checkpoint-60000/pytorch_model.bin +3 -0
- checkpoint-60000/rng_state_0.pth +3 -0
- checkpoint-60000/rng_state_1.pth +3 -0
- checkpoint-60000/rng_state_2.pth +3 -0
- checkpoint-60000/rng_state_3.pth +3 -0
- checkpoint-60000/scheduler.pt +3 -0
- checkpoint-60000/sentencepiece.bpe.model +3 -0
- checkpoint-60000/special_tokens_map.json +63 -0
- checkpoint-60000/tokenizer.json +3 -0
- checkpoint-60000/tokenizer_config.json +528 -0
- checkpoint-60000/trainer_state.json +889 -0
- checkpoint-60000/training_args.bin +3 -0
- config.json +59 -0
- eval_results.json +10 -0
- generated_predictions.txt +0 -0
- generation_config.json +12 -0
- predict_results.json +9 -0
- pytorch_model.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +63 -0
- tokenizer.json +3 -0
- tokenizer_config.json +528 -0
- train_results.json +8 -0
- trainer_state.json +1130 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ 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 |
+
checkpoint-60000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- ko
|
| 4 |
+
- en
|
| 5 |
+
base_model: facebook/mbart-large-50-many-to-many-mmt
|
| 6 |
+
tags:
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
metrics:
|
| 9 |
+
- bleu
|
| 10 |
+
model-index:
|
| 11 |
+
- name: ko-en_mbartLarge_exp20p_linear
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# ko-en_mbartLarge_exp20p_linear
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 1.1514
|
| 23 |
+
- Bleu: 29.2703
|
| 24 |
+
- Gen Len: 18.512
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 5e-05
|
| 44 |
+
- train_batch_size: 4
|
| 45 |
+
- eval_batch_size: 4
|
| 46 |
+
- seed: 42
|
| 47 |
+
- distributed_type: multi-GPU
|
| 48 |
+
- num_devices: 4
|
| 49 |
+
- gradient_accumulation_steps: 2
|
| 50 |
+
- total_train_batch_size: 32
|
| 51 |
+
- total_eval_batch_size: 16
|
| 52 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
+
- lr_scheduler_type: linear
|
| 54 |
+
- lr_scheduler_warmup_steps: 2000
|
| 55 |
+
- num_epochs: 40
|
| 56 |
+
|
| 57 |
+
### Training results
|
| 58 |
+
|
| 59 |
+
| Training Loss | Epoch | Step | Bleu | Gen Len | Validation Loss |
|
| 60 |
+
|:-------------:|:-----:|:-----:|:-------:|:-------:|:---------------:|
|
| 61 |
+
| 1.3977 | 0.46 | 4000 | 22.7153 | 18.7135 | 1.3720 |
|
| 62 |
+
| 1.2824 | 0.93 | 8000 | 24.8579 | 18.7821 | 1.2633 |
|
| 63 |
+
| 1.1989 | 1.39 | 12000 | 26.2533 | 18.7975 | 1.2069 |
|
| 64 |
+
| 1.1534 | 1.86 | 16000 | 26.1503 | 19.2075 | 1.1907 |
|
| 65 |
+
| 1.0245 | 2.32 | 20000 | 27.8764 | 18.6046 | 1.1464 |
|
| 66 |
+
| 1.0186 | 2.78 | 24000 | 28.4585 | 18.6731 | 1.1286 |
|
| 67 |
+
| 0.9245 | 3.25 | 28000 | 1.1264 | 28.4834 | 18.5428 |
|
| 68 |
+
| 0.9343 | 3.71 | 32000 | 1.1182 | 28.8235 | 18.7833 |
|
| 69 |
+
| 0.8215 | 4.18 | 36000 | 1.1331 | 28.6134 | 18.5656 |
|
| 70 |
+
| 0.8456 | 4.64 | 40000 | 1.1203 | 28.7324 | 18.459 |
|
| 71 |
+
| 0.7437 | 5.11 | 44000 | 1.1458 | 28.7297 | 18.7835 |
|
| 72 |
+
| 0.7829 | 5.57 | 48000 | 1.1367 | 28.8328 | 18.6052 |
|
| 73 |
+
| 0.7434 | 6.03 | 52000 | 1.1697 | 28.2106 | 18.4871 |
|
| 74 |
+
| 0.7153 | 6.5 | 56000 | 1.1771 | 28.1455 | 18.7413 |
|
| 75 |
+
| 0.6996 | 6.96 | 60000 | 1.1514 | 29.2694 | 18.5162 |
|
| 76 |
+
| 0.6336 | 7.43 | 64000 | 1.2213 | 28.1465 | 18.5439 |
|
| 77 |
+
| 0.7218 | 7.89 | 68000 | 1.1835 | 28.2245 | 18.5246 |
|
| 78 |
+
| 0.5934 | 8.35 | 72000 | 1.2387 | 28.3836 | 18.6717 |
|
| 79 |
+
| 0.5723 | 8.82 | 76000 | 1.2323 | 28.5925 | 18.5566 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- Transformers 4.34.0
|
| 85 |
+
- Pytorch 2.1.0+cu121
|
| 86 |
+
- Datasets 2.14.5
|
| 87 |
+
- Tokenizers 0.14.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<mask>": 250053,
|
| 4 |
+
"<pad>": 1,
|
| 5 |
+
"<s>": 0,
|
| 6 |
+
"<unk>": 3,
|
| 7 |
+
"af_ZA": 250026,
|
| 8 |
+
"ar_AR": 250001,
|
| 9 |
+
"az_AZ": 250027,
|
| 10 |
+
"bn_IN": 250028,
|
| 11 |
+
"cs_CZ": 250002,
|
| 12 |
+
"de_DE": 250003,
|
| 13 |
+
"en_XX": 250004,
|
| 14 |
+
"es_XX": 250005,
|
| 15 |
+
"et_EE": 250006,
|
| 16 |
+
"fa_IR": 250029,
|
| 17 |
+
"fi_FI": 250007,
|
| 18 |
+
"fr_XX": 250008,
|
| 19 |
+
"gl_ES": 250051,
|
| 20 |
+
"gu_IN": 250009,
|
| 21 |
+
"he_IL": 250030,
|
| 22 |
+
"hi_IN": 250010,
|
| 23 |
+
"hr_HR": 250031,
|
| 24 |
+
"id_ID": 250032,
|
| 25 |
+
"it_IT": 250011,
|
| 26 |
+
"ja_XX": 250012,
|
| 27 |
+
"ka_GE": 250033,
|
| 28 |
+
"kk_KZ": 250013,
|
| 29 |
+
"km_KH": 250034,
|
| 30 |
+
"ko_KR": 250014,
|
| 31 |
+
"lt_LT": 250015,
|
| 32 |
+
"lv_LV": 250016,
|
| 33 |
+
"mk_MK": 250035,
|
| 34 |
+
"ml_IN": 250036,
|
| 35 |
+
"mn_MN": 250037,
|
| 36 |
+
"mr_IN": 250038,
|
| 37 |
+
"my_MM": 250017,
|
| 38 |
+
"ne_NP": 250018,
|
| 39 |
+
"nl_XX": 250019,
|
| 40 |
+
"pl_PL": 250039,
|
| 41 |
+
"ps_AF": 250040,
|
| 42 |
+
"pt_XX": 250041,
|
| 43 |
+
"ro_RO": 250020,
|
| 44 |
+
"ru_RU": 250021,
|
| 45 |
+
"si_LK": 250022,
|
| 46 |
+
"sl_SI": 250052,
|
| 47 |
+
"sv_SE": 250042,
|
| 48 |
+
"sw_KE": 250043,
|
| 49 |
+
"ta_IN": 250044,
|
| 50 |
+
"te_IN": 250045,
|
| 51 |
+
"th_TH": 250046,
|
| 52 |
+
"tl_XX": 250047,
|
| 53 |
+
"tr_TR": 250023,
|
| 54 |
+
"uk_UA": 250048,
|
| 55 |
+
"ur_PK": 250049,
|
| 56 |
+
"vi_VN": 250024,
|
| 57 |
+
"xh_ZA": 250050,
|
| 58 |
+
"zh_CN": 250025
|
| 59 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.82,
|
| 3 |
+
"eval_bleu": 29.2703,
|
| 4 |
+
"eval_gen_len": 18.512,
|
| 5 |
+
"eval_loss": 1.1514174938201904,
|
| 6 |
+
"eval_runtime": 1139.1943,
|
| 7 |
+
"eval_samples": 17237,
|
| 8 |
+
"eval_samples_per_second": 15.131,
|
| 9 |
+
"eval_steps_per_second": 0.946,
|
| 10 |
+
"predict_bleu": 29.4157,
|
| 11 |
+
"predict_gen_len": 18.4677,
|
| 12 |
+
"predict_loss": 1.1349819898605347,
|
| 13 |
+
"predict_runtime": 1142.9928,
|
| 14 |
+
"predict_samples": 17237,
|
| 15 |
+
"predict_samples_per_second": 15.081,
|
| 16 |
+
"predict_steps_per_second": 0.943,
|
| 17 |
+
"train_loss": 0.5245111798738178,
|
| 18 |
+
"train_runtime": 53209.1351,
|
| 19 |
+
"train_samples": 137900,
|
| 20 |
+
"train_samples_per_second": 103.666,
|
| 21 |
+
"train_steps_per_second": 6.479
|
| 22 |
+
}
|
checkpoint-60000/added_tokens.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<mask>": 250053,
|
| 4 |
+
"<pad>": 1,
|
| 5 |
+
"<s>": 0,
|
| 6 |
+
"<unk>": 3,
|
| 7 |
+
"af_ZA": 250026,
|
| 8 |
+
"ar_AR": 250001,
|
| 9 |
+
"az_AZ": 250027,
|
| 10 |
+
"bn_IN": 250028,
|
| 11 |
+
"cs_CZ": 250002,
|
| 12 |
+
"de_DE": 250003,
|
| 13 |
+
"en_XX": 250004,
|
| 14 |
+
"es_XX": 250005,
|
| 15 |
+
"et_EE": 250006,
|
| 16 |
+
"fa_IR": 250029,
|
| 17 |
+
"fi_FI": 250007,
|
| 18 |
+
"fr_XX": 250008,
|
| 19 |
+
"gl_ES": 250051,
|
| 20 |
+
"gu_IN": 250009,
|
| 21 |
+
"he_IL": 250030,
|
| 22 |
+
"hi_IN": 250010,
|
| 23 |
+
"hr_HR": 250031,
|
| 24 |
+
"id_ID": 250032,
|
| 25 |
+
"it_IT": 250011,
|
| 26 |
+
"ja_XX": 250012,
|
| 27 |
+
"ka_GE": 250033,
|
| 28 |
+
"kk_KZ": 250013,
|
| 29 |
+
"km_KH": 250034,
|
| 30 |
+
"ko_KR": 250014,
|
| 31 |
+
"lt_LT": 250015,
|
| 32 |
+
"lv_LV": 250016,
|
| 33 |
+
"mk_MK": 250035,
|
| 34 |
+
"ml_IN": 250036,
|
| 35 |
+
"mn_MN": 250037,
|
| 36 |
+
"mr_IN": 250038,
|
| 37 |
+
"my_MM": 250017,
|
| 38 |
+
"ne_NP": 250018,
|
| 39 |
+
"nl_XX": 250019,
|
| 40 |
+
"pl_PL": 250039,
|
| 41 |
+
"ps_AF": 250040,
|
| 42 |
+
"pt_XX": 250041,
|
| 43 |
+
"ro_RO": 250020,
|
| 44 |
+
"ru_RU": 250021,
|
| 45 |
+
"si_LK": 250022,
|
| 46 |
+
"sl_SI": 250052,
|
| 47 |
+
"sv_SE": 250042,
|
| 48 |
+
"sw_KE": 250043,
|
| 49 |
+
"ta_IN": 250044,
|
| 50 |
+
"te_IN": 250045,
|
| 51 |
+
"th_TH": 250046,
|
| 52 |
+
"tl_XX": 250047,
|
| 53 |
+
"tr_TR": 250023,
|
| 54 |
+
"uk_UA": 250048,
|
| 55 |
+
"ur_PK": 250049,
|
| 56 |
+
"vi_VN": 250024,
|
| 57 |
+
"xh_ZA": 250050,
|
| 58 |
+
"zh_CN": 250025
|
| 59 |
+
}
|
checkpoint-60000/config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MBartForConditionalGeneration"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 12,
|
| 20 |
+
"decoder_start_token_id": 2,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"early_stopping": true,
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 4096,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 2,
|
| 28 |
+
"forced_bos_token_id": 250004,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"max_length": 200,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"model_type": "mbart",
|
| 46 |
+
"normalize_before": true,
|
| 47 |
+
"normalize_embedding": true,
|
| 48 |
+
"num_beams": 5,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"scale_embedding": true,
|
| 53 |
+
"static_position_embeddings": false,
|
| 54 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.34.0",
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 250054
|
| 59 |
+
}
|
checkpoint-60000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 250004,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"max_length": 200,
|
| 9 |
+
"num_beams": 5,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.34.0"
|
| 12 |
+
}
|
checkpoint-60000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a8cb3634881579edb3865155cb9669df457983cd6b07992d58a90fae1b49e9f
|
| 3 |
+
size 4887481135
|
checkpoint-60000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc97fc5ff225ece2754b9bbe091afbfb7dbf47e2b48680709028050fa6f07692
|
| 3 |
+
size 2444694490
|
checkpoint-60000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddd94a2f98f6527dd829619b426e00b09b72105a0a50d515ad8d341fab3fceb5
|
| 3 |
+
size 15024
|
checkpoint-60000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe09369bb01e96cdf251292ed1c8b4dade1f5afafdb2b7628f0901e9f095492d
|
| 3 |
+
size 15024
|
checkpoint-60000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7b4a869926519f11b2aea1ca9405011f6166d6011d8d0e07552017032a4a9ba
|
| 3 |
+
size 15024
|
checkpoint-60000/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9fc25d38f561ccfa43435d6ef5519c390bc343538761ad6217b5f954a0db7f01
|
| 3 |
+
size 15024
|
checkpoint-60000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0e218ab0c3ffaf664b7fa5d08be2d1727b59f25d39c849e0d781428f397e560
|
| 3 |
+
size 1064
|
checkpoint-60000/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-60000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"af_ZA",
|
| 29 |
+
"az_AZ",
|
| 30 |
+
"bn_IN",
|
| 31 |
+
"fa_IR",
|
| 32 |
+
"he_IL",
|
| 33 |
+
"hr_HR",
|
| 34 |
+
"id_ID",
|
| 35 |
+
"ka_GE",
|
| 36 |
+
"km_KH",
|
| 37 |
+
"mk_MK",
|
| 38 |
+
"ml_IN",
|
| 39 |
+
"mn_MN",
|
| 40 |
+
"mr_IN",
|
| 41 |
+
"pl_PL",
|
| 42 |
+
"ps_AF",
|
| 43 |
+
"pt_XX",
|
| 44 |
+
"sv_SE",
|
| 45 |
+
"sw_KE",
|
| 46 |
+
"ta_IN",
|
| 47 |
+
"te_IN",
|
| 48 |
+
"th_TH",
|
| 49 |
+
"tl_XX",
|
| 50 |
+
"uk_UA",
|
| 51 |
+
"ur_PK",
|
| 52 |
+
"xh_ZA",
|
| 53 |
+
"gl_ES",
|
| 54 |
+
"sl_SI"
|
| 55 |
+
],
|
| 56 |
+
"bos_token": "<s>",
|
| 57 |
+
"cls_token": "<s>",
|
| 58 |
+
"eos_token": "</s>",
|
| 59 |
+
"mask_token": "<mask>",
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"sep_token": "</s>",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
checkpoint-60000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
|
| 3 |
+
size 17109939
|
checkpoint-60000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": true,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": true,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": true,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": true,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": true,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": true,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": true,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": true,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": true,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": true,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": true,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": true,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": true,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": true,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": true,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": true,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": true,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": true,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": true,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": true,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": true,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": true,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": true,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": true,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": true,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": true,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": true,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": true,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": true,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": true,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": true,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": true,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": true,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": true,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": true,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": true,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": true,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": true,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": true,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": true,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": true,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": true,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": true,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "af_ZA",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": true,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "az_AZ",
|
| 245 |
+
"lstrip": true,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": true,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"250028": {
|
| 252 |
+
"content": "bn_IN",
|
| 253 |
+
"lstrip": true,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": true,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"250029": {
|
| 260 |
+
"content": "fa_IR",
|
| 261 |
+
"lstrip": true,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": true,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"250030": {
|
| 268 |
+
"content": "he_IL",
|
| 269 |
+
"lstrip": true,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": true,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"250031": {
|
| 276 |
+
"content": "hr_HR",
|
| 277 |
+
"lstrip": true,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": true,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"250032": {
|
| 284 |
+
"content": "id_ID",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": true,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"250033": {
|
| 292 |
+
"content": "ka_GE",
|
| 293 |
+
"lstrip": true,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": true,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"250034": {
|
| 300 |
+
"content": "km_KH",
|
| 301 |
+
"lstrip": true,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": true,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"250035": {
|
| 308 |
+
"content": "mk_MK",
|
| 309 |
+
"lstrip": true,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": true,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"250036": {
|
| 316 |
+
"content": "ml_IN",
|
| 317 |
+
"lstrip": true,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": true,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"250037": {
|
| 324 |
+
"content": "mn_MN",
|
| 325 |
+
"lstrip": true,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": true,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"250038": {
|
| 332 |
+
"content": "mr_IN",
|
| 333 |
+
"lstrip": true,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": true,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"250039": {
|
| 340 |
+
"content": "pl_PL",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": true,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"250040": {
|
| 348 |
+
"content": "ps_AF",
|
| 349 |
+
"lstrip": true,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": true,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"250041": {
|
| 356 |
+
"content": "pt_XX",
|
| 357 |
+
"lstrip": true,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": true,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"250042": {
|
| 364 |
+
"content": "sv_SE",
|
| 365 |
+
"lstrip": true,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": true,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"250043": {
|
| 372 |
+
"content": "sw_KE",
|
| 373 |
+
"lstrip": true,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": true,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"250044": {
|
| 380 |
+
"content": "ta_IN",
|
| 381 |
+
"lstrip": true,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": true,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"250045": {
|
| 388 |
+
"content": "te_IN",
|
| 389 |
+
"lstrip": true,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": true,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"250046": {
|
| 396 |
+
"content": "th_TH",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": true,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"250047": {
|
| 404 |
+
"content": "tl_XX",
|
| 405 |
+
"lstrip": true,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": true,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"250048": {
|
| 412 |
+
"content": "uk_UA",
|
| 413 |
+
"lstrip": true,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": true,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"250049": {
|
| 420 |
+
"content": "ur_PK",
|
| 421 |
+
"lstrip": true,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": true,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"250050": {
|
| 428 |
+
"content": "xh_ZA",
|
| 429 |
+
"lstrip": true,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": true,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"250051": {
|
| 436 |
+
"content": "gl_ES",
|
| 437 |
+
"lstrip": true,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": true,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"250052": {
|
| 444 |
+
"content": "sl_SI",
|
| 445 |
+
"lstrip": true,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": true,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"250053": {
|
| 452 |
+
"content": "<mask>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
}
|
| 459 |
+
},
|
| 460 |
+
"additional_special_tokens": [
|
| 461 |
+
"ar_AR",
|
| 462 |
+
"cs_CZ",
|
| 463 |
+
"de_DE",
|
| 464 |
+
"en_XX",
|
| 465 |
+
"es_XX",
|
| 466 |
+
"et_EE",
|
| 467 |
+
"fi_FI",
|
| 468 |
+
"fr_XX",
|
| 469 |
+
"gu_IN",
|
| 470 |
+
"hi_IN",
|
| 471 |
+
"it_IT",
|
| 472 |
+
"ja_XX",
|
| 473 |
+
"kk_KZ",
|
| 474 |
+
"ko_KR",
|
| 475 |
+
"lt_LT",
|
| 476 |
+
"lv_LV",
|
| 477 |
+
"my_MM",
|
| 478 |
+
"ne_NP",
|
| 479 |
+
"nl_XX",
|
| 480 |
+
"ro_RO",
|
| 481 |
+
"ru_RU",
|
| 482 |
+
"si_LK",
|
| 483 |
+
"tr_TR",
|
| 484 |
+
"vi_VN",
|
| 485 |
+
"zh_CN",
|
| 486 |
+
"af_ZA",
|
| 487 |
+
"az_AZ",
|
| 488 |
+
"bn_IN",
|
| 489 |
+
"fa_IR",
|
| 490 |
+
"he_IL",
|
| 491 |
+
"hr_HR",
|
| 492 |
+
"id_ID",
|
| 493 |
+
"ka_GE",
|
| 494 |
+
"km_KH",
|
| 495 |
+
"mk_MK",
|
| 496 |
+
"ml_IN",
|
| 497 |
+
"mn_MN",
|
| 498 |
+
"mr_IN",
|
| 499 |
+
"pl_PL",
|
| 500 |
+
"ps_AF",
|
| 501 |
+
"pt_XX",
|
| 502 |
+
"sv_SE",
|
| 503 |
+
"sw_KE",
|
| 504 |
+
"ta_IN",
|
| 505 |
+
"te_IN",
|
| 506 |
+
"th_TH",
|
| 507 |
+
"tl_XX",
|
| 508 |
+
"uk_UA",
|
| 509 |
+
"ur_PK",
|
| 510 |
+
"xh_ZA",
|
| 511 |
+
"gl_ES",
|
| 512 |
+
"sl_SI"
|
| 513 |
+
],
|
| 514 |
+
"bos_token": "<s>",
|
| 515 |
+
"clean_up_tokenization_spaces": true,
|
| 516 |
+
"cls_token": "<s>",
|
| 517 |
+
"eos_token": "</s>",
|
| 518 |
+
"language_codes": "ML50",
|
| 519 |
+
"mask_token": "<mask>",
|
| 520 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 521 |
+
"pad_token": "<pad>",
|
| 522 |
+
"sep_token": "</s>",
|
| 523 |
+
"sp_model_kwargs": {},
|
| 524 |
+
"src_lang": "ko_KR",
|
| 525 |
+
"tgt_lang": "en_XX",
|
| 526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 527 |
+
"unk_token": "<unk>"
|
| 528 |
+
}
|
checkpoint-60000/trainer_state.json
ADDED
|
@@ -0,0 +1,889 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 29.2694,
|
| 3 |
+
"best_model_checkpoint": "./ko-en_mbartLarge_exp20p_linear/checkpoint-60000",
|
| 4 |
+
"epoch": 6.9613644274277755,
|
| 5 |
+
"eval_steps": 4000,
|
| 6 |
+
"global_step": 60000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 1.25e-05,
|
| 14 |
+
"loss": 1.9079,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.12,
|
| 19 |
+
"learning_rate": 2.5e-05,
|
| 20 |
+
"loss": 1.7987,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.17,
|
| 25 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 26 |
+
"loss": 1.648,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.23,
|
| 31 |
+
"learning_rate": 5e-05,
|
| 32 |
+
"loss": 1.5348,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 4.9927062667755864e-05,
|
| 38 |
+
"loss": 1.4769,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.35,
|
| 43 |
+
"learning_rate": 4.985412533551173e-05,
|
| 44 |
+
"loss": 1.4511,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.41,
|
| 49 |
+
"learning_rate": 4.978118800326759e-05,
|
| 50 |
+
"loss": 1.4201,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.46,
|
| 55 |
+
"learning_rate": 4.970825067102346e-05,
|
| 56 |
+
"loss": 1.3977,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.46,
|
| 61 |
+
"eval_bleu": 22.7153,
|
| 62 |
+
"eval_gen_len": 18.7135,
|
| 63 |
+
"eval_loss": 1.3720011711120605,
|
| 64 |
+
"eval_runtime": 1153.7,
|
| 65 |
+
"eval_samples_per_second": 14.941,
|
| 66 |
+
"eval_steps_per_second": 0.934,
|
| 67 |
+
"step": 4000
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.52,
|
| 71 |
+
"learning_rate": 4.963531333877932e-05,
|
| 72 |
+
"loss": 1.3792,
|
| 73 |
+
"step": 4500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.58,
|
| 77 |
+
"learning_rate": 4.956237600653519e-05,
|
| 78 |
+
"loss": 1.3552,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.64,
|
| 83 |
+
"learning_rate": 4.948943867429105e-05,
|
| 84 |
+
"loss": 1.3468,
|
| 85 |
+
"step": 5500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.7,
|
| 89 |
+
"learning_rate": 4.941650134204692e-05,
|
| 90 |
+
"loss": 1.3468,
|
| 91 |
+
"step": 6000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.75,
|
| 95 |
+
"learning_rate": 4.934356400980278e-05,
|
| 96 |
+
"loss": 1.3236,
|
| 97 |
+
"step": 6500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.81,
|
| 101 |
+
"learning_rate": 4.927062667755864e-05,
|
| 102 |
+
"loss": 1.2924,
|
| 103 |
+
"step": 7000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.87,
|
| 107 |
+
"learning_rate": 4.9197689345314504e-05,
|
| 108 |
+
"loss": 1.3005,
|
| 109 |
+
"step": 7500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.93,
|
| 113 |
+
"learning_rate": 4.912475201307037e-05,
|
| 114 |
+
"loss": 1.2824,
|
| 115 |
+
"step": 8000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.93,
|
| 119 |
+
"eval_bleu": 24.8579,
|
| 120 |
+
"eval_gen_len": 18.7821,
|
| 121 |
+
"eval_loss": 1.2633070945739746,
|
| 122 |
+
"eval_runtime": 1161.9505,
|
| 123 |
+
"eval_samples_per_second": 14.835,
|
| 124 |
+
"eval_steps_per_second": 0.928,
|
| 125 |
+
"step": 8000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.99,
|
| 129 |
+
"learning_rate": 4.905181468082623e-05,
|
| 130 |
+
"loss": 1.2758,
|
| 131 |
+
"step": 8500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.04,
|
| 135 |
+
"learning_rate": 4.89788773485821e-05,
|
| 136 |
+
"loss": 1.2295,
|
| 137 |
+
"step": 9000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.1,
|
| 141 |
+
"learning_rate": 4.890594001633796e-05,
|
| 142 |
+
"loss": 1.2196,
|
| 143 |
+
"step": 9500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.16,
|
| 147 |
+
"learning_rate": 4.883300268409383e-05,
|
| 148 |
+
"loss": 1.2043,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.22,
|
| 153 |
+
"learning_rate": 4.876006535184969e-05,
|
| 154 |
+
"loss": 1.2144,
|
| 155 |
+
"step": 10500
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.28,
|
| 159 |
+
"learning_rate": 4.868712801960555e-05,
|
| 160 |
+
"loss": 1.195,
|
| 161 |
+
"step": 11000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.33,
|
| 165 |
+
"learning_rate": 4.861419068736142e-05,
|
| 166 |
+
"loss": 1.184,
|
| 167 |
+
"step": 11500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.39,
|
| 171 |
+
"learning_rate": 4.854125335511728e-05,
|
| 172 |
+
"loss": 1.1989,
|
| 173 |
+
"step": 12000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.39,
|
| 177 |
+
"eval_bleu": 26.2533,
|
| 178 |
+
"eval_gen_len": 18.7975,
|
| 179 |
+
"eval_loss": 1.2069392204284668,
|
| 180 |
+
"eval_runtime": 1167.3322,
|
| 181 |
+
"eval_samples_per_second": 14.766,
|
| 182 |
+
"eval_steps_per_second": 0.923,
|
| 183 |
+
"step": 12000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.45,
|
| 187 |
+
"learning_rate": 4.846831602287315e-05,
|
| 188 |
+
"loss": 1.186,
|
| 189 |
+
"step": 12500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.51,
|
| 193 |
+
"learning_rate": 4.839537869062901e-05,
|
| 194 |
+
"loss": 1.1785,
|
| 195 |
+
"step": 13000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.57,
|
| 199 |
+
"learning_rate": 4.832244135838488e-05,
|
| 200 |
+
"loss": 1.1659,
|
| 201 |
+
"step": 13500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.62,
|
| 205 |
+
"learning_rate": 4.824950402614074e-05,
|
| 206 |
+
"loss": 1.1624,
|
| 207 |
+
"step": 14000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.68,
|
| 211 |
+
"learning_rate": 4.817656669389661e-05,
|
| 212 |
+
"loss": 1.1785,
|
| 213 |
+
"step": 14500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.74,
|
| 217 |
+
"learning_rate": 4.810362936165247e-05,
|
| 218 |
+
"loss": 1.1465,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.8,
|
| 223 |
+
"learning_rate": 4.803069202940834e-05,
|
| 224 |
+
"loss": 1.1617,
|
| 225 |
+
"step": 15500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.86,
|
| 229 |
+
"learning_rate": 4.79577546971642e-05,
|
| 230 |
+
"loss": 1.1534,
|
| 231 |
+
"step": 16000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.86,
|
| 235 |
+
"eval_bleu": 26.1503,
|
| 236 |
+
"eval_gen_len": 19.2075,
|
| 237 |
+
"eval_loss": 1.190694808959961,
|
| 238 |
+
"eval_runtime": 1177.3192,
|
| 239 |
+
"eval_samples_per_second": 14.641,
|
| 240 |
+
"eval_steps_per_second": 0.916,
|
| 241 |
+
"step": 16000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.91,
|
| 245 |
+
"learning_rate": 4.788481736492006e-05,
|
| 246 |
+
"loss": 1.1745,
|
| 247 |
+
"step": 16500
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.97,
|
| 251 |
+
"learning_rate": 4.781188003267593e-05,
|
| 252 |
+
"loss": 1.1663,
|
| 253 |
+
"step": 17000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.03,
|
| 257 |
+
"learning_rate": 4.773894270043179e-05,
|
| 258 |
+
"loss": 1.1029,
|
| 259 |
+
"step": 17500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.09,
|
| 263 |
+
"learning_rate": 4.766600536818765e-05,
|
| 264 |
+
"loss": 1.0464,
|
| 265 |
+
"step": 18000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.15,
|
| 269 |
+
"learning_rate": 4.759306803594352e-05,
|
| 270 |
+
"loss": 1.0348,
|
| 271 |
+
"step": 18500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.2,
|
| 275 |
+
"learning_rate": 4.752013070369938e-05,
|
| 276 |
+
"loss": 1.0228,
|
| 277 |
+
"step": 19000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.26,
|
| 281 |
+
"learning_rate": 4.744719337145524e-05,
|
| 282 |
+
"loss": 1.0503,
|
| 283 |
+
"step": 19500
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.32,
|
| 287 |
+
"learning_rate": 4.737425603921111e-05,
|
| 288 |
+
"loss": 1.0245,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.32,
|
| 293 |
+
"eval_bleu": 27.8764,
|
| 294 |
+
"eval_gen_len": 18.6046,
|
| 295 |
+
"eval_loss": 1.146370768547058,
|
| 296 |
+
"eval_runtime": 1146.7204,
|
| 297 |
+
"eval_samples_per_second": 15.032,
|
| 298 |
+
"eval_steps_per_second": 0.94,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.38,
|
| 303 |
+
"learning_rate": 4.730131870696697e-05,
|
| 304 |
+
"loss": 1.0493,
|
| 305 |
+
"step": 20500
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.44,
|
| 309 |
+
"learning_rate": 4.722838137472284e-05,
|
| 310 |
+
"loss": 1.0287,
|
| 311 |
+
"step": 21000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.49,
|
| 315 |
+
"learning_rate": 4.71554440424787e-05,
|
| 316 |
+
"loss": 1.03,
|
| 317 |
+
"step": 21500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.55,
|
| 321 |
+
"learning_rate": 4.708250671023457e-05,
|
| 322 |
+
"loss": 1.0354,
|
| 323 |
+
"step": 22000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.61,
|
| 327 |
+
"learning_rate": 4.700956937799043e-05,
|
| 328 |
+
"loss": 1.0268,
|
| 329 |
+
"step": 22500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.67,
|
| 333 |
+
"learning_rate": 4.69366320457463e-05,
|
| 334 |
+
"loss": 1.039,
|
| 335 |
+
"step": 23000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.73,
|
| 339 |
+
"learning_rate": 4.686369471350216e-05,
|
| 340 |
+
"loss": 1.0343,
|
| 341 |
+
"step": 23500
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 2.78,
|
| 345 |
+
"learning_rate": 4.679075738125803e-05,
|
| 346 |
+
"loss": 1.0186,
|
| 347 |
+
"step": 24000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.78,
|
| 351 |
+
"eval_bleu": 28.4585,
|
| 352 |
+
"eval_gen_len": 18.6731,
|
| 353 |
+
"eval_loss": 1.1286325454711914,
|
| 354 |
+
"eval_runtime": 1151.5333,
|
| 355 |
+
"eval_samples_per_second": 14.969,
|
| 356 |
+
"eval_steps_per_second": 0.936,
|
| 357 |
+
"step": 24000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.84,
|
| 361 |
+
"learning_rate": 4.671782004901389e-05,
|
| 362 |
+
"loss": 1.0383,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.9,
|
| 367 |
+
"learning_rate": 4.664488271676976e-05,
|
| 368 |
+
"loss": 1.0488,
|
| 369 |
+
"step": 25000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.96,
|
| 373 |
+
"learning_rate": 4.657194538452562e-05,
|
| 374 |
+
"loss": 1.0572,
|
| 375 |
+
"step": 25500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.02,
|
| 379 |
+
"learning_rate": 4.649900805228149e-05,
|
| 380 |
+
"loss": 1.0168,
|
| 381 |
+
"step": 26000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.07,
|
| 385 |
+
"learning_rate": 4.642607072003735e-05,
|
| 386 |
+
"loss": 0.9238,
|
| 387 |
+
"step": 26500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.13,
|
| 391 |
+
"learning_rate": 4.635313338779321e-05,
|
| 392 |
+
"loss": 0.9281,
|
| 393 |
+
"step": 27000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.19,
|
| 397 |
+
"learning_rate": 4.628019605554907e-05,
|
| 398 |
+
"loss": 0.9225,
|
| 399 |
+
"step": 27500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 3.25,
|
| 403 |
+
"learning_rate": 4.620725872330493e-05,
|
| 404 |
+
"loss": 0.9245,
|
| 405 |
+
"step": 28000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 3.25,
|
| 409 |
+
"eval_bleu": 28.4834,
|
| 410 |
+
"eval_gen_len": 18.5428,
|
| 411 |
+
"eval_loss": 1.1263633966445923,
|
| 412 |
+
"eval_runtime": 1147.1298,
|
| 413 |
+
"eval_samples_per_second": 15.026,
|
| 414 |
+
"eval_steps_per_second": 0.94,
|
| 415 |
+
"step": 28000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.31,
|
| 419 |
+
"learning_rate": 4.61343213910608e-05,
|
| 420 |
+
"loss": 0.9145,
|
| 421 |
+
"step": 28500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.36,
|
| 425 |
+
"learning_rate": 4.606138405881666e-05,
|
| 426 |
+
"loss": 0.9228,
|
| 427 |
+
"step": 29000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.42,
|
| 431 |
+
"learning_rate": 4.598844672657253e-05,
|
| 432 |
+
"loss": 0.9196,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 3.48,
|
| 437 |
+
"learning_rate": 4.591550939432839e-05,
|
| 438 |
+
"loss": 0.9138,
|
| 439 |
+
"step": 30000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 3.54,
|
| 443 |
+
"learning_rate": 4.584257206208426e-05,
|
| 444 |
+
"loss": 0.9274,
|
| 445 |
+
"step": 30500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 3.6,
|
| 449 |
+
"learning_rate": 4.576963472984012e-05,
|
| 450 |
+
"loss": 0.9077,
|
| 451 |
+
"step": 31000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.65,
|
| 455 |
+
"learning_rate": 4.569669739759599e-05,
|
| 456 |
+
"loss": 0.9312,
|
| 457 |
+
"step": 31500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.71,
|
| 461 |
+
"learning_rate": 4.562376006535185e-05,
|
| 462 |
+
"loss": 0.9343,
|
| 463 |
+
"step": 32000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.71,
|
| 467 |
+
"eval_bleu": 28.8235,
|
| 468 |
+
"eval_gen_len": 18.7833,
|
| 469 |
+
"eval_loss": 1.1182423830032349,
|
| 470 |
+
"eval_runtime": 1155.993,
|
| 471 |
+
"eval_samples_per_second": 14.911,
|
| 472 |
+
"eval_steps_per_second": 0.933,
|
| 473 |
+
"step": 32000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.77,
|
| 477 |
+
"learning_rate": 4.555082273310772e-05,
|
| 478 |
+
"loss": 0.9232,
|
| 479 |
+
"step": 32500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.83,
|
| 483 |
+
"learning_rate": 4.547788540086358e-05,
|
| 484 |
+
"loss": 0.9308,
|
| 485 |
+
"step": 33000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.89,
|
| 489 |
+
"learning_rate": 4.540494806861945e-05,
|
| 490 |
+
"loss": 0.9361,
|
| 491 |
+
"step": 33500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.94,
|
| 495 |
+
"learning_rate": 4.533201073637531e-05,
|
| 496 |
+
"loss": 0.9139,
|
| 497 |
+
"step": 34000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.0,
|
| 501 |
+
"learning_rate": 4.525907340413118e-05,
|
| 502 |
+
"loss": 0.919,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.06,
|
| 507 |
+
"learning_rate": 4.518613607188704e-05,
|
| 508 |
+
"loss": 0.8226,
|
| 509 |
+
"step": 35000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.12,
|
| 513 |
+
"learning_rate": 4.5113198739642907e-05,
|
| 514 |
+
"loss": 0.8242,
|
| 515 |
+
"step": 35500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.18,
|
| 519 |
+
"learning_rate": 4.504026140739877e-05,
|
| 520 |
+
"loss": 0.8215,
|
| 521 |
+
"step": 36000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.18,
|
| 525 |
+
"eval_bleu": 28.6134,
|
| 526 |
+
"eval_gen_len": 18.5656,
|
| 527 |
+
"eval_loss": 1.133116602897644,
|
| 528 |
+
"eval_runtime": 1144.5462,
|
| 529 |
+
"eval_samples_per_second": 15.06,
|
| 530 |
+
"eval_steps_per_second": 0.942,
|
| 531 |
+
"step": 36000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.23,
|
| 535 |
+
"learning_rate": 4.496732407515463e-05,
|
| 536 |
+
"loss": 0.833,
|
| 537 |
+
"step": 36500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 4.29,
|
| 541 |
+
"learning_rate": 4.489438674291049e-05,
|
| 542 |
+
"loss": 0.8175,
|
| 543 |
+
"step": 37000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 4.35,
|
| 547 |
+
"learning_rate": 4.482144941066635e-05,
|
| 548 |
+
"loss": 0.8313,
|
| 549 |
+
"step": 37500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 4.41,
|
| 553 |
+
"learning_rate": 4.474851207842222e-05,
|
| 554 |
+
"loss": 0.832,
|
| 555 |
+
"step": 38000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 4.47,
|
| 559 |
+
"learning_rate": 4.467557474617808e-05,
|
| 560 |
+
"loss": 0.8397,
|
| 561 |
+
"step": 38500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 4.52,
|
| 565 |
+
"learning_rate": 4.460263741393395e-05,
|
| 566 |
+
"loss": 0.8324,
|
| 567 |
+
"step": 39000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 4.58,
|
| 571 |
+
"learning_rate": 4.452970008168981e-05,
|
| 572 |
+
"loss": 0.8379,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 4.64,
|
| 577 |
+
"learning_rate": 4.445676274944568e-05,
|
| 578 |
+
"loss": 0.8456,
|
| 579 |
+
"step": 40000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 4.64,
|
| 583 |
+
"eval_bleu": 28.7324,
|
| 584 |
+
"eval_gen_len": 18.459,
|
| 585 |
+
"eval_loss": 1.1203099489212036,
|
| 586 |
+
"eval_runtime": 1135.006,
|
| 587 |
+
"eval_samples_per_second": 15.187,
|
| 588 |
+
"eval_steps_per_second": 0.95,
|
| 589 |
+
"step": 40000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 4.7,
|
| 593 |
+
"learning_rate": 4.438382541720154e-05,
|
| 594 |
+
"loss": 0.8481,
|
| 595 |
+
"step": 40500
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.76,
|
| 599 |
+
"learning_rate": 4.431088808495741e-05,
|
| 600 |
+
"loss": 0.8376,
|
| 601 |
+
"step": 41000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 4.81,
|
| 605 |
+
"learning_rate": 4.423795075271327e-05,
|
| 606 |
+
"loss": 0.8416,
|
| 607 |
+
"step": 41500
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 4.87,
|
| 611 |
+
"learning_rate": 4.416501342046914e-05,
|
| 612 |
+
"loss": 0.8462,
|
| 613 |
+
"step": 42000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 4.93,
|
| 617 |
+
"learning_rate": 4.4092076088225e-05,
|
| 618 |
+
"loss": 0.846,
|
| 619 |
+
"step": 42500
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 4.99,
|
| 623 |
+
"learning_rate": 4.401913875598087e-05,
|
| 624 |
+
"loss": 0.8429,
|
| 625 |
+
"step": 43000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 5.05,
|
| 629 |
+
"learning_rate": 4.394620142373673e-05,
|
| 630 |
+
"loss": 0.7648,
|
| 631 |
+
"step": 43500
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 5.11,
|
| 635 |
+
"learning_rate": 4.3873264091492596e-05,
|
| 636 |
+
"loss": 0.7437,
|
| 637 |
+
"step": 44000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.11,
|
| 641 |
+
"eval_bleu": 28.7297,
|
| 642 |
+
"eval_gen_len": 18.7835,
|
| 643 |
+
"eval_loss": 1.1458019018173218,
|
| 644 |
+
"eval_runtime": 1166.3289,
|
| 645 |
+
"eval_samples_per_second": 14.779,
|
| 646 |
+
"eval_steps_per_second": 0.924,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 5.16,
|
| 651 |
+
"learning_rate": 4.380032675924846e-05,
|
| 652 |
+
"loss": 0.7562,
|
| 653 |
+
"step": 44500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 5.22,
|
| 657 |
+
"learning_rate": 4.3727389427004326e-05,
|
| 658 |
+
"loss": 0.752,
|
| 659 |
+
"step": 45000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 5.28,
|
| 663 |
+
"learning_rate": 4.365445209476019e-05,
|
| 664 |
+
"loss": 0.7526,
|
| 665 |
+
"step": 45500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.34,
|
| 669 |
+
"learning_rate": 4.358151476251605e-05,
|
| 670 |
+
"loss": 0.7779,
|
| 671 |
+
"step": 46000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 5.4,
|
| 675 |
+
"learning_rate": 4.350857743027191e-05,
|
| 676 |
+
"loss": 0.7942,
|
| 677 |
+
"step": 46500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.45,
|
| 681 |
+
"learning_rate": 4.343564009802778e-05,
|
| 682 |
+
"loss": 0.7954,
|
| 683 |
+
"step": 47000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.51,
|
| 687 |
+
"learning_rate": 4.336270276578364e-05,
|
| 688 |
+
"loss": 0.7874,
|
| 689 |
+
"step": 47500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.57,
|
| 693 |
+
"learning_rate": 4.32897654335395e-05,
|
| 694 |
+
"loss": 0.7829,
|
| 695 |
+
"step": 48000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.57,
|
| 699 |
+
"eval_bleu": 28.8328,
|
| 700 |
+
"eval_gen_len": 18.6052,
|
| 701 |
+
"eval_loss": 1.1367288827896118,
|
| 702 |
+
"eval_runtime": 1145.791,
|
| 703 |
+
"eval_samples_per_second": 15.044,
|
| 704 |
+
"eval_steps_per_second": 0.941,
|
| 705 |
+
"step": 48000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 5.63,
|
| 709 |
+
"learning_rate": 4.321682810129537e-05,
|
| 710 |
+
"loss": 0.7853,
|
| 711 |
+
"step": 48500
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 5.69,
|
| 715 |
+
"learning_rate": 4.314389076905123e-05,
|
| 716 |
+
"loss": 0.7916,
|
| 717 |
+
"step": 49000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 5.74,
|
| 721 |
+
"learning_rate": 4.30709534368071e-05,
|
| 722 |
+
"loss": 0.7914,
|
| 723 |
+
"step": 49500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 5.8,
|
| 727 |
+
"learning_rate": 4.299801610456296e-05,
|
| 728 |
+
"loss": 0.8096,
|
| 729 |
+
"step": 50000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 5.86,
|
| 733 |
+
"learning_rate": 4.292507877231883e-05,
|
| 734 |
+
"loss": 0.8128,
|
| 735 |
+
"step": 50500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 5.92,
|
| 739 |
+
"learning_rate": 4.285214144007469e-05,
|
| 740 |
+
"loss": 0.7987,
|
| 741 |
+
"step": 51000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 5.98,
|
| 745 |
+
"learning_rate": 4.2779204107830557e-05,
|
| 746 |
+
"loss": 0.8185,
|
| 747 |
+
"step": 51500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 6.03,
|
| 751 |
+
"learning_rate": 4.270626677558642e-05,
|
| 752 |
+
"loss": 0.7434,
|
| 753 |
+
"step": 52000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 6.03,
|
| 757 |
+
"eval_bleu": 28.2106,
|
| 758 |
+
"eval_gen_len": 18.4871,
|
| 759 |
+
"eval_loss": 1.1697230339050293,
|
| 760 |
+
"eval_runtime": 1139.5757,
|
| 761 |
+
"eval_samples_per_second": 15.126,
|
| 762 |
+
"eval_steps_per_second": 0.946,
|
| 763 |
+
"step": 52000
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 6.09,
|
| 767 |
+
"learning_rate": 4.2633329443342286e-05,
|
| 768 |
+
"loss": 0.6938,
|
| 769 |
+
"step": 52500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 6.15,
|
| 773 |
+
"learning_rate": 4.256039211109815e-05,
|
| 774 |
+
"loss": 0.695,
|
| 775 |
+
"step": 53000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 6.21,
|
| 779 |
+
"learning_rate": 4.2487454778854015e-05,
|
| 780 |
+
"loss": 0.7016,
|
| 781 |
+
"step": 53500
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 6.27,
|
| 785 |
+
"learning_rate": 4.241451744660988e-05,
|
| 786 |
+
"loss": 0.6836,
|
| 787 |
+
"step": 54000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 6.32,
|
| 791 |
+
"learning_rate": 4.234158011436574e-05,
|
| 792 |
+
"loss": 0.6807,
|
| 793 |
+
"step": 54500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 6.38,
|
| 797 |
+
"learning_rate": 4.2268642782121606e-05,
|
| 798 |
+
"loss": 0.674,
|
| 799 |
+
"step": 55000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 6.44,
|
| 803 |
+
"learning_rate": 4.219570544987747e-05,
|
| 804 |
+
"loss": 0.6897,
|
| 805 |
+
"step": 55500
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 6.5,
|
| 809 |
+
"learning_rate": 4.2122768117633335e-05,
|
| 810 |
+
"loss": 0.7153,
|
| 811 |
+
"step": 56000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 6.5,
|
| 815 |
+
"eval_bleu": 28.1455,
|
| 816 |
+
"eval_gen_len": 18.7413,
|
| 817 |
+
"eval_loss": 1.1770974397659302,
|
| 818 |
+
"eval_runtime": 1156.7175,
|
| 819 |
+
"eval_samples_per_second": 14.902,
|
| 820 |
+
"eval_steps_per_second": 0.932,
|
| 821 |
+
"step": 56000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 6.56,
|
| 825 |
+
"learning_rate": 4.20498307853892e-05,
|
| 826 |
+
"loss": 0.7177,
|
| 827 |
+
"step": 56500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 6.61,
|
| 831 |
+
"learning_rate": 4.197689345314506e-05,
|
| 832 |
+
"loss": 0.716,
|
| 833 |
+
"step": 57000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 6.67,
|
| 837 |
+
"learning_rate": 4.190395612090092e-05,
|
| 838 |
+
"loss": 0.721,
|
| 839 |
+
"step": 57500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.73,
|
| 843 |
+
"learning_rate": 4.183101878865679e-05,
|
| 844 |
+
"loss": 0.7152,
|
| 845 |
+
"step": 58000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 6.79,
|
| 849 |
+
"learning_rate": 4.175808145641265e-05,
|
| 850 |
+
"loss": 0.7012,
|
| 851 |
+
"step": 58500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 6.85,
|
| 855 |
+
"learning_rate": 4.168514412416852e-05,
|
| 856 |
+
"loss": 0.7069,
|
| 857 |
+
"step": 59000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 6.9,
|
| 861 |
+
"learning_rate": 4.161220679192438e-05,
|
| 862 |
+
"loss": 0.7088,
|
| 863 |
+
"step": 59500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 6.96,
|
| 867 |
+
"learning_rate": 4.1539269459680246e-05,
|
| 868 |
+
"loss": 0.6996,
|
| 869 |
+
"step": 60000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.96,
|
| 873 |
+
"eval_bleu": 29.2694,
|
| 874 |
+
"eval_gen_len": 18.5162,
|
| 875 |
+
"eval_loss": 1.1514174938201904,
|
| 876 |
+
"eval_runtime": 1140.5069,
|
| 877 |
+
"eval_samples_per_second": 15.113,
|
| 878 |
+
"eval_steps_per_second": 0.945,
|
| 879 |
+
"step": 60000
|
| 880 |
+
}
|
| 881 |
+
],
|
| 882 |
+
"logging_steps": 500,
|
| 883 |
+
"max_steps": 344760,
|
| 884 |
+
"num_train_epochs": 40,
|
| 885 |
+
"save_steps": 4000,
|
| 886 |
+
"total_flos": 2.080583273523708e+18,
|
| 887 |
+
"trial_name": null,
|
| 888 |
+
"trial_params": null
|
| 889 |
+
}
|
checkpoint-60000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e917f03bf6223d6917033778988c8bd45cc9111d9f4fed1584656468d3a4d435
|
| 3 |
+
size 4792
|
config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
|
| 3 |
+
"_num_labels": 3,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"add_bias_logits": false,
|
| 7 |
+
"add_final_layer_norm": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"MBartForConditionalGeneration"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"classif_dropout": 0.0,
|
| 14 |
+
"classifier_dropout": 0.0,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 12,
|
| 20 |
+
"decoder_start_token_id": 2,
|
| 21 |
+
"dropout": 0.1,
|
| 22 |
+
"early_stopping": true,
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 4096,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 2,
|
| 28 |
+
"forced_bos_token_id": 250004,
|
| 29 |
+
"forced_eos_token_id": 2,
|
| 30 |
+
"gradient_checkpointing": false,
|
| 31 |
+
"id2label": {
|
| 32 |
+
"0": "LABEL_0",
|
| 33 |
+
"1": "LABEL_1",
|
| 34 |
+
"2": "LABEL_2"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"LABEL_0": 0,
|
| 40 |
+
"LABEL_1": 1,
|
| 41 |
+
"LABEL_2": 2
|
| 42 |
+
},
|
| 43 |
+
"max_length": 200,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"model_type": "mbart",
|
| 46 |
+
"normalize_before": true,
|
| 47 |
+
"normalize_embedding": true,
|
| 48 |
+
"num_beams": 5,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"output_past": true,
|
| 51 |
+
"pad_token_id": 1,
|
| 52 |
+
"scale_embedding": true,
|
| 53 |
+
"static_position_embeddings": false,
|
| 54 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 55 |
+
"torch_dtype": "float32",
|
| 56 |
+
"transformers_version": "4.34.0",
|
| 57 |
+
"use_cache": true,
|
| 58 |
+
"vocab_size": 250054
|
| 59 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.82,
|
| 3 |
+
"eval_bleu": 29.2703,
|
| 4 |
+
"eval_gen_len": 18.512,
|
| 5 |
+
"eval_loss": 1.1514174938201904,
|
| 6 |
+
"eval_runtime": 1139.1943,
|
| 7 |
+
"eval_samples": 17237,
|
| 8 |
+
"eval_samples_per_second": 15.131,
|
| 9 |
+
"eval_steps_per_second": 0.946
|
| 10 |
+
}
|
generated_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"early_stopping": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"forced_bos_token_id": 250004,
|
| 7 |
+
"forced_eos_token_id": 2,
|
| 8 |
+
"max_length": 200,
|
| 9 |
+
"num_beams": 5,
|
| 10 |
+
"pad_token_id": 1,
|
| 11 |
+
"transformers_version": "4.34.0"
|
| 12 |
+
}
|
predict_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"predict_bleu": 29.4157,
|
| 3 |
+
"predict_gen_len": 18.4677,
|
| 4 |
+
"predict_loss": 1.1349819898605347,
|
| 5 |
+
"predict_runtime": 1142.9928,
|
| 6 |
+
"predict_samples": 17237,
|
| 7 |
+
"predict_samples_per_second": 15.081,
|
| 8 |
+
"predict_steps_per_second": 0.943
|
| 9 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc97fc5ff225ece2754b9bbe091afbfb7dbf47e2b48680709028050fa6f07692
|
| 3 |
+
size 2444694490
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"af_ZA",
|
| 29 |
+
"az_AZ",
|
| 30 |
+
"bn_IN",
|
| 31 |
+
"fa_IR",
|
| 32 |
+
"he_IL",
|
| 33 |
+
"hr_HR",
|
| 34 |
+
"id_ID",
|
| 35 |
+
"ka_GE",
|
| 36 |
+
"km_KH",
|
| 37 |
+
"mk_MK",
|
| 38 |
+
"ml_IN",
|
| 39 |
+
"mn_MN",
|
| 40 |
+
"mr_IN",
|
| 41 |
+
"pl_PL",
|
| 42 |
+
"ps_AF",
|
| 43 |
+
"pt_XX",
|
| 44 |
+
"sv_SE",
|
| 45 |
+
"sw_KE",
|
| 46 |
+
"ta_IN",
|
| 47 |
+
"te_IN",
|
| 48 |
+
"th_TH",
|
| 49 |
+
"tl_XX",
|
| 50 |
+
"uk_UA",
|
| 51 |
+
"ur_PK",
|
| 52 |
+
"xh_ZA",
|
| 53 |
+
"gl_ES",
|
| 54 |
+
"sl_SI"
|
| 55 |
+
],
|
| 56 |
+
"bos_token": "<s>",
|
| 57 |
+
"cls_token": "<s>",
|
| 58 |
+
"eos_token": "</s>",
|
| 59 |
+
"mask_token": "<mask>",
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"sep_token": "</s>",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
|
| 3 |
+
size 17109939
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": true,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"250002": {
|
| 44 |
+
"content": "cs_CZ",
|
| 45 |
+
"lstrip": true,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": true,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"250003": {
|
| 52 |
+
"content": "de_DE",
|
| 53 |
+
"lstrip": true,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": true,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"250004": {
|
| 60 |
+
"content": "en_XX",
|
| 61 |
+
"lstrip": true,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": true,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"250005": {
|
| 68 |
+
"content": "es_XX",
|
| 69 |
+
"lstrip": true,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": true,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"250006": {
|
| 76 |
+
"content": "et_EE",
|
| 77 |
+
"lstrip": true,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": true,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"250007": {
|
| 84 |
+
"content": "fi_FI",
|
| 85 |
+
"lstrip": true,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": true,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"250008": {
|
| 92 |
+
"content": "fr_XX",
|
| 93 |
+
"lstrip": true,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": true,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"250009": {
|
| 100 |
+
"content": "gu_IN",
|
| 101 |
+
"lstrip": true,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": true,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"250010": {
|
| 108 |
+
"content": "hi_IN",
|
| 109 |
+
"lstrip": true,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": true,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"250011": {
|
| 116 |
+
"content": "it_IT",
|
| 117 |
+
"lstrip": true,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": true,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"250012": {
|
| 124 |
+
"content": "ja_XX",
|
| 125 |
+
"lstrip": true,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": true,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"250013": {
|
| 132 |
+
"content": "kk_KZ",
|
| 133 |
+
"lstrip": true,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": true,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"250014": {
|
| 140 |
+
"content": "ko_KR",
|
| 141 |
+
"lstrip": true,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": true,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"250015": {
|
| 148 |
+
"content": "lt_LT",
|
| 149 |
+
"lstrip": true,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": true,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"250016": {
|
| 156 |
+
"content": "lv_LV",
|
| 157 |
+
"lstrip": true,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": true,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"250017": {
|
| 164 |
+
"content": "my_MM",
|
| 165 |
+
"lstrip": true,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": true,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"250018": {
|
| 172 |
+
"content": "ne_NP",
|
| 173 |
+
"lstrip": true,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": true,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
},
|
| 179 |
+
"250019": {
|
| 180 |
+
"content": "nl_XX",
|
| 181 |
+
"lstrip": true,
|
| 182 |
+
"normalized": false,
|
| 183 |
+
"rstrip": true,
|
| 184 |
+
"single_word": false,
|
| 185 |
+
"special": true
|
| 186 |
+
},
|
| 187 |
+
"250020": {
|
| 188 |
+
"content": "ro_RO",
|
| 189 |
+
"lstrip": true,
|
| 190 |
+
"normalized": false,
|
| 191 |
+
"rstrip": true,
|
| 192 |
+
"single_word": false,
|
| 193 |
+
"special": true
|
| 194 |
+
},
|
| 195 |
+
"250021": {
|
| 196 |
+
"content": "ru_RU",
|
| 197 |
+
"lstrip": true,
|
| 198 |
+
"normalized": false,
|
| 199 |
+
"rstrip": true,
|
| 200 |
+
"single_word": false,
|
| 201 |
+
"special": true
|
| 202 |
+
},
|
| 203 |
+
"250022": {
|
| 204 |
+
"content": "si_LK",
|
| 205 |
+
"lstrip": true,
|
| 206 |
+
"normalized": false,
|
| 207 |
+
"rstrip": true,
|
| 208 |
+
"single_word": false,
|
| 209 |
+
"special": true
|
| 210 |
+
},
|
| 211 |
+
"250023": {
|
| 212 |
+
"content": "tr_TR",
|
| 213 |
+
"lstrip": true,
|
| 214 |
+
"normalized": false,
|
| 215 |
+
"rstrip": true,
|
| 216 |
+
"single_word": false,
|
| 217 |
+
"special": true
|
| 218 |
+
},
|
| 219 |
+
"250024": {
|
| 220 |
+
"content": "vi_VN",
|
| 221 |
+
"lstrip": true,
|
| 222 |
+
"normalized": false,
|
| 223 |
+
"rstrip": true,
|
| 224 |
+
"single_word": false,
|
| 225 |
+
"special": true
|
| 226 |
+
},
|
| 227 |
+
"250025": {
|
| 228 |
+
"content": "zh_CN",
|
| 229 |
+
"lstrip": true,
|
| 230 |
+
"normalized": false,
|
| 231 |
+
"rstrip": true,
|
| 232 |
+
"single_word": false,
|
| 233 |
+
"special": true
|
| 234 |
+
},
|
| 235 |
+
"250026": {
|
| 236 |
+
"content": "af_ZA",
|
| 237 |
+
"lstrip": true,
|
| 238 |
+
"normalized": false,
|
| 239 |
+
"rstrip": true,
|
| 240 |
+
"single_word": false,
|
| 241 |
+
"special": true
|
| 242 |
+
},
|
| 243 |
+
"250027": {
|
| 244 |
+
"content": "az_AZ",
|
| 245 |
+
"lstrip": true,
|
| 246 |
+
"normalized": false,
|
| 247 |
+
"rstrip": true,
|
| 248 |
+
"single_word": false,
|
| 249 |
+
"special": true
|
| 250 |
+
},
|
| 251 |
+
"250028": {
|
| 252 |
+
"content": "bn_IN",
|
| 253 |
+
"lstrip": true,
|
| 254 |
+
"normalized": false,
|
| 255 |
+
"rstrip": true,
|
| 256 |
+
"single_word": false,
|
| 257 |
+
"special": true
|
| 258 |
+
},
|
| 259 |
+
"250029": {
|
| 260 |
+
"content": "fa_IR",
|
| 261 |
+
"lstrip": true,
|
| 262 |
+
"normalized": false,
|
| 263 |
+
"rstrip": true,
|
| 264 |
+
"single_word": false,
|
| 265 |
+
"special": true
|
| 266 |
+
},
|
| 267 |
+
"250030": {
|
| 268 |
+
"content": "he_IL",
|
| 269 |
+
"lstrip": true,
|
| 270 |
+
"normalized": false,
|
| 271 |
+
"rstrip": true,
|
| 272 |
+
"single_word": false,
|
| 273 |
+
"special": true
|
| 274 |
+
},
|
| 275 |
+
"250031": {
|
| 276 |
+
"content": "hr_HR",
|
| 277 |
+
"lstrip": true,
|
| 278 |
+
"normalized": false,
|
| 279 |
+
"rstrip": true,
|
| 280 |
+
"single_word": false,
|
| 281 |
+
"special": true
|
| 282 |
+
},
|
| 283 |
+
"250032": {
|
| 284 |
+
"content": "id_ID",
|
| 285 |
+
"lstrip": true,
|
| 286 |
+
"normalized": false,
|
| 287 |
+
"rstrip": true,
|
| 288 |
+
"single_word": false,
|
| 289 |
+
"special": true
|
| 290 |
+
},
|
| 291 |
+
"250033": {
|
| 292 |
+
"content": "ka_GE",
|
| 293 |
+
"lstrip": true,
|
| 294 |
+
"normalized": false,
|
| 295 |
+
"rstrip": true,
|
| 296 |
+
"single_word": false,
|
| 297 |
+
"special": true
|
| 298 |
+
},
|
| 299 |
+
"250034": {
|
| 300 |
+
"content": "km_KH",
|
| 301 |
+
"lstrip": true,
|
| 302 |
+
"normalized": false,
|
| 303 |
+
"rstrip": true,
|
| 304 |
+
"single_word": false,
|
| 305 |
+
"special": true
|
| 306 |
+
},
|
| 307 |
+
"250035": {
|
| 308 |
+
"content": "mk_MK",
|
| 309 |
+
"lstrip": true,
|
| 310 |
+
"normalized": false,
|
| 311 |
+
"rstrip": true,
|
| 312 |
+
"single_word": false,
|
| 313 |
+
"special": true
|
| 314 |
+
},
|
| 315 |
+
"250036": {
|
| 316 |
+
"content": "ml_IN",
|
| 317 |
+
"lstrip": true,
|
| 318 |
+
"normalized": false,
|
| 319 |
+
"rstrip": true,
|
| 320 |
+
"single_word": false,
|
| 321 |
+
"special": true
|
| 322 |
+
},
|
| 323 |
+
"250037": {
|
| 324 |
+
"content": "mn_MN",
|
| 325 |
+
"lstrip": true,
|
| 326 |
+
"normalized": false,
|
| 327 |
+
"rstrip": true,
|
| 328 |
+
"single_word": false,
|
| 329 |
+
"special": true
|
| 330 |
+
},
|
| 331 |
+
"250038": {
|
| 332 |
+
"content": "mr_IN",
|
| 333 |
+
"lstrip": true,
|
| 334 |
+
"normalized": false,
|
| 335 |
+
"rstrip": true,
|
| 336 |
+
"single_word": false,
|
| 337 |
+
"special": true
|
| 338 |
+
},
|
| 339 |
+
"250039": {
|
| 340 |
+
"content": "pl_PL",
|
| 341 |
+
"lstrip": true,
|
| 342 |
+
"normalized": false,
|
| 343 |
+
"rstrip": true,
|
| 344 |
+
"single_word": false,
|
| 345 |
+
"special": true
|
| 346 |
+
},
|
| 347 |
+
"250040": {
|
| 348 |
+
"content": "ps_AF",
|
| 349 |
+
"lstrip": true,
|
| 350 |
+
"normalized": false,
|
| 351 |
+
"rstrip": true,
|
| 352 |
+
"single_word": false,
|
| 353 |
+
"special": true
|
| 354 |
+
},
|
| 355 |
+
"250041": {
|
| 356 |
+
"content": "pt_XX",
|
| 357 |
+
"lstrip": true,
|
| 358 |
+
"normalized": false,
|
| 359 |
+
"rstrip": true,
|
| 360 |
+
"single_word": false,
|
| 361 |
+
"special": true
|
| 362 |
+
},
|
| 363 |
+
"250042": {
|
| 364 |
+
"content": "sv_SE",
|
| 365 |
+
"lstrip": true,
|
| 366 |
+
"normalized": false,
|
| 367 |
+
"rstrip": true,
|
| 368 |
+
"single_word": false,
|
| 369 |
+
"special": true
|
| 370 |
+
},
|
| 371 |
+
"250043": {
|
| 372 |
+
"content": "sw_KE",
|
| 373 |
+
"lstrip": true,
|
| 374 |
+
"normalized": false,
|
| 375 |
+
"rstrip": true,
|
| 376 |
+
"single_word": false,
|
| 377 |
+
"special": true
|
| 378 |
+
},
|
| 379 |
+
"250044": {
|
| 380 |
+
"content": "ta_IN",
|
| 381 |
+
"lstrip": true,
|
| 382 |
+
"normalized": false,
|
| 383 |
+
"rstrip": true,
|
| 384 |
+
"single_word": false,
|
| 385 |
+
"special": true
|
| 386 |
+
},
|
| 387 |
+
"250045": {
|
| 388 |
+
"content": "te_IN",
|
| 389 |
+
"lstrip": true,
|
| 390 |
+
"normalized": false,
|
| 391 |
+
"rstrip": true,
|
| 392 |
+
"single_word": false,
|
| 393 |
+
"special": true
|
| 394 |
+
},
|
| 395 |
+
"250046": {
|
| 396 |
+
"content": "th_TH",
|
| 397 |
+
"lstrip": true,
|
| 398 |
+
"normalized": false,
|
| 399 |
+
"rstrip": true,
|
| 400 |
+
"single_word": false,
|
| 401 |
+
"special": true
|
| 402 |
+
},
|
| 403 |
+
"250047": {
|
| 404 |
+
"content": "tl_XX",
|
| 405 |
+
"lstrip": true,
|
| 406 |
+
"normalized": false,
|
| 407 |
+
"rstrip": true,
|
| 408 |
+
"single_word": false,
|
| 409 |
+
"special": true
|
| 410 |
+
},
|
| 411 |
+
"250048": {
|
| 412 |
+
"content": "uk_UA",
|
| 413 |
+
"lstrip": true,
|
| 414 |
+
"normalized": false,
|
| 415 |
+
"rstrip": true,
|
| 416 |
+
"single_word": false,
|
| 417 |
+
"special": true
|
| 418 |
+
},
|
| 419 |
+
"250049": {
|
| 420 |
+
"content": "ur_PK",
|
| 421 |
+
"lstrip": true,
|
| 422 |
+
"normalized": false,
|
| 423 |
+
"rstrip": true,
|
| 424 |
+
"single_word": false,
|
| 425 |
+
"special": true
|
| 426 |
+
},
|
| 427 |
+
"250050": {
|
| 428 |
+
"content": "xh_ZA",
|
| 429 |
+
"lstrip": true,
|
| 430 |
+
"normalized": false,
|
| 431 |
+
"rstrip": true,
|
| 432 |
+
"single_word": false,
|
| 433 |
+
"special": true
|
| 434 |
+
},
|
| 435 |
+
"250051": {
|
| 436 |
+
"content": "gl_ES",
|
| 437 |
+
"lstrip": true,
|
| 438 |
+
"normalized": false,
|
| 439 |
+
"rstrip": true,
|
| 440 |
+
"single_word": false,
|
| 441 |
+
"special": true
|
| 442 |
+
},
|
| 443 |
+
"250052": {
|
| 444 |
+
"content": "sl_SI",
|
| 445 |
+
"lstrip": true,
|
| 446 |
+
"normalized": false,
|
| 447 |
+
"rstrip": true,
|
| 448 |
+
"single_word": false,
|
| 449 |
+
"special": true
|
| 450 |
+
},
|
| 451 |
+
"250053": {
|
| 452 |
+
"content": "<mask>",
|
| 453 |
+
"lstrip": true,
|
| 454 |
+
"normalized": true,
|
| 455 |
+
"rstrip": false,
|
| 456 |
+
"single_word": false,
|
| 457 |
+
"special": true
|
| 458 |
+
}
|
| 459 |
+
},
|
| 460 |
+
"additional_special_tokens": [
|
| 461 |
+
"ar_AR",
|
| 462 |
+
"cs_CZ",
|
| 463 |
+
"de_DE",
|
| 464 |
+
"en_XX",
|
| 465 |
+
"es_XX",
|
| 466 |
+
"et_EE",
|
| 467 |
+
"fi_FI",
|
| 468 |
+
"fr_XX",
|
| 469 |
+
"gu_IN",
|
| 470 |
+
"hi_IN",
|
| 471 |
+
"it_IT",
|
| 472 |
+
"ja_XX",
|
| 473 |
+
"kk_KZ",
|
| 474 |
+
"ko_KR",
|
| 475 |
+
"lt_LT",
|
| 476 |
+
"lv_LV",
|
| 477 |
+
"my_MM",
|
| 478 |
+
"ne_NP",
|
| 479 |
+
"nl_XX",
|
| 480 |
+
"ro_RO",
|
| 481 |
+
"ru_RU",
|
| 482 |
+
"si_LK",
|
| 483 |
+
"tr_TR",
|
| 484 |
+
"vi_VN",
|
| 485 |
+
"zh_CN",
|
| 486 |
+
"af_ZA",
|
| 487 |
+
"az_AZ",
|
| 488 |
+
"bn_IN",
|
| 489 |
+
"fa_IR",
|
| 490 |
+
"he_IL",
|
| 491 |
+
"hr_HR",
|
| 492 |
+
"id_ID",
|
| 493 |
+
"ka_GE",
|
| 494 |
+
"km_KH",
|
| 495 |
+
"mk_MK",
|
| 496 |
+
"ml_IN",
|
| 497 |
+
"mn_MN",
|
| 498 |
+
"mr_IN",
|
| 499 |
+
"pl_PL",
|
| 500 |
+
"ps_AF",
|
| 501 |
+
"pt_XX",
|
| 502 |
+
"sv_SE",
|
| 503 |
+
"sw_KE",
|
| 504 |
+
"ta_IN",
|
| 505 |
+
"te_IN",
|
| 506 |
+
"th_TH",
|
| 507 |
+
"tl_XX",
|
| 508 |
+
"uk_UA",
|
| 509 |
+
"ur_PK",
|
| 510 |
+
"xh_ZA",
|
| 511 |
+
"gl_ES",
|
| 512 |
+
"sl_SI"
|
| 513 |
+
],
|
| 514 |
+
"bos_token": "<s>",
|
| 515 |
+
"clean_up_tokenization_spaces": true,
|
| 516 |
+
"cls_token": "<s>",
|
| 517 |
+
"eos_token": "</s>",
|
| 518 |
+
"language_codes": "ML50",
|
| 519 |
+
"mask_token": "<mask>",
|
| 520 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 521 |
+
"pad_token": "<pad>",
|
| 522 |
+
"sep_token": "</s>",
|
| 523 |
+
"sp_model_kwargs": {},
|
| 524 |
+
"src_lang": "ko_KR",
|
| 525 |
+
"tgt_lang": "en_XX",
|
| 526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
| 527 |
+
"unk_token": "<unk>"
|
| 528 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.82,
|
| 3 |
+
"train_loss": 0.5245111798738178,
|
| 4 |
+
"train_runtime": 53209.1351,
|
| 5 |
+
"train_samples": 137900,
|
| 6 |
+
"train_samples_per_second": 103.666,
|
| 7 |
+
"train_steps_per_second": 6.479
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 29.2694,
|
| 3 |
+
"best_model_checkpoint": "./ko-en_mbartLarge_exp20p_linear/checkpoint-60000",
|
| 4 |
+
"epoch": 8.81772827474185,
|
| 5 |
+
"eval_steps": 4000,
|
| 6 |
+
"global_step": 76000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 1.25e-05,
|
| 14 |
+
"loss": 1.9079,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.12,
|
| 19 |
+
"learning_rate": 2.5e-05,
|
| 20 |
+
"loss": 1.7987,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.17,
|
| 25 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 26 |
+
"loss": 1.648,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.23,
|
| 31 |
+
"learning_rate": 5e-05,
|
| 32 |
+
"loss": 1.5348,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 4.9927062667755864e-05,
|
| 38 |
+
"loss": 1.4769,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.35,
|
| 43 |
+
"learning_rate": 4.985412533551173e-05,
|
| 44 |
+
"loss": 1.4511,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.41,
|
| 49 |
+
"learning_rate": 4.978118800326759e-05,
|
| 50 |
+
"loss": 1.4201,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.46,
|
| 55 |
+
"learning_rate": 4.970825067102346e-05,
|
| 56 |
+
"loss": 1.3977,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.46,
|
| 61 |
+
"eval_bleu": 22.7153,
|
| 62 |
+
"eval_gen_len": 18.7135,
|
| 63 |
+
"eval_loss": 1.3720011711120605,
|
| 64 |
+
"eval_runtime": 1153.7,
|
| 65 |
+
"eval_samples_per_second": 14.941,
|
| 66 |
+
"eval_steps_per_second": 0.934,
|
| 67 |
+
"step": 4000
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.52,
|
| 71 |
+
"learning_rate": 4.963531333877932e-05,
|
| 72 |
+
"loss": 1.3792,
|
| 73 |
+
"step": 4500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.58,
|
| 77 |
+
"learning_rate": 4.956237600653519e-05,
|
| 78 |
+
"loss": 1.3552,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.64,
|
| 83 |
+
"learning_rate": 4.948943867429105e-05,
|
| 84 |
+
"loss": 1.3468,
|
| 85 |
+
"step": 5500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.7,
|
| 89 |
+
"learning_rate": 4.941650134204692e-05,
|
| 90 |
+
"loss": 1.3468,
|
| 91 |
+
"step": 6000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.75,
|
| 95 |
+
"learning_rate": 4.934356400980278e-05,
|
| 96 |
+
"loss": 1.3236,
|
| 97 |
+
"step": 6500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.81,
|
| 101 |
+
"learning_rate": 4.927062667755864e-05,
|
| 102 |
+
"loss": 1.2924,
|
| 103 |
+
"step": 7000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.87,
|
| 107 |
+
"learning_rate": 4.9197689345314504e-05,
|
| 108 |
+
"loss": 1.3005,
|
| 109 |
+
"step": 7500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.93,
|
| 113 |
+
"learning_rate": 4.912475201307037e-05,
|
| 114 |
+
"loss": 1.2824,
|
| 115 |
+
"step": 8000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.93,
|
| 119 |
+
"eval_bleu": 24.8579,
|
| 120 |
+
"eval_gen_len": 18.7821,
|
| 121 |
+
"eval_loss": 1.2633070945739746,
|
| 122 |
+
"eval_runtime": 1161.9505,
|
| 123 |
+
"eval_samples_per_second": 14.835,
|
| 124 |
+
"eval_steps_per_second": 0.928,
|
| 125 |
+
"step": 8000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.99,
|
| 129 |
+
"learning_rate": 4.905181468082623e-05,
|
| 130 |
+
"loss": 1.2758,
|
| 131 |
+
"step": 8500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.04,
|
| 135 |
+
"learning_rate": 4.89788773485821e-05,
|
| 136 |
+
"loss": 1.2295,
|
| 137 |
+
"step": 9000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.1,
|
| 141 |
+
"learning_rate": 4.890594001633796e-05,
|
| 142 |
+
"loss": 1.2196,
|
| 143 |
+
"step": 9500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.16,
|
| 147 |
+
"learning_rate": 4.883300268409383e-05,
|
| 148 |
+
"loss": 1.2043,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.22,
|
| 153 |
+
"learning_rate": 4.876006535184969e-05,
|
| 154 |
+
"loss": 1.2144,
|
| 155 |
+
"step": 10500
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.28,
|
| 159 |
+
"learning_rate": 4.868712801960555e-05,
|
| 160 |
+
"loss": 1.195,
|
| 161 |
+
"step": 11000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.33,
|
| 165 |
+
"learning_rate": 4.861419068736142e-05,
|
| 166 |
+
"loss": 1.184,
|
| 167 |
+
"step": 11500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.39,
|
| 171 |
+
"learning_rate": 4.854125335511728e-05,
|
| 172 |
+
"loss": 1.1989,
|
| 173 |
+
"step": 12000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.39,
|
| 177 |
+
"eval_bleu": 26.2533,
|
| 178 |
+
"eval_gen_len": 18.7975,
|
| 179 |
+
"eval_loss": 1.2069392204284668,
|
| 180 |
+
"eval_runtime": 1167.3322,
|
| 181 |
+
"eval_samples_per_second": 14.766,
|
| 182 |
+
"eval_steps_per_second": 0.923,
|
| 183 |
+
"step": 12000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.45,
|
| 187 |
+
"learning_rate": 4.846831602287315e-05,
|
| 188 |
+
"loss": 1.186,
|
| 189 |
+
"step": 12500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.51,
|
| 193 |
+
"learning_rate": 4.839537869062901e-05,
|
| 194 |
+
"loss": 1.1785,
|
| 195 |
+
"step": 13000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.57,
|
| 199 |
+
"learning_rate": 4.832244135838488e-05,
|
| 200 |
+
"loss": 1.1659,
|
| 201 |
+
"step": 13500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.62,
|
| 205 |
+
"learning_rate": 4.824950402614074e-05,
|
| 206 |
+
"loss": 1.1624,
|
| 207 |
+
"step": 14000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.68,
|
| 211 |
+
"learning_rate": 4.817656669389661e-05,
|
| 212 |
+
"loss": 1.1785,
|
| 213 |
+
"step": 14500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.74,
|
| 217 |
+
"learning_rate": 4.810362936165247e-05,
|
| 218 |
+
"loss": 1.1465,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.8,
|
| 223 |
+
"learning_rate": 4.803069202940834e-05,
|
| 224 |
+
"loss": 1.1617,
|
| 225 |
+
"step": 15500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.86,
|
| 229 |
+
"learning_rate": 4.79577546971642e-05,
|
| 230 |
+
"loss": 1.1534,
|
| 231 |
+
"step": 16000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.86,
|
| 235 |
+
"eval_bleu": 26.1503,
|
| 236 |
+
"eval_gen_len": 19.2075,
|
| 237 |
+
"eval_loss": 1.190694808959961,
|
| 238 |
+
"eval_runtime": 1177.3192,
|
| 239 |
+
"eval_samples_per_second": 14.641,
|
| 240 |
+
"eval_steps_per_second": 0.916,
|
| 241 |
+
"step": 16000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.91,
|
| 245 |
+
"learning_rate": 4.788481736492006e-05,
|
| 246 |
+
"loss": 1.1745,
|
| 247 |
+
"step": 16500
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.97,
|
| 251 |
+
"learning_rate": 4.781188003267593e-05,
|
| 252 |
+
"loss": 1.1663,
|
| 253 |
+
"step": 17000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.03,
|
| 257 |
+
"learning_rate": 4.773894270043179e-05,
|
| 258 |
+
"loss": 1.1029,
|
| 259 |
+
"step": 17500
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.09,
|
| 263 |
+
"learning_rate": 4.766600536818765e-05,
|
| 264 |
+
"loss": 1.0464,
|
| 265 |
+
"step": 18000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.15,
|
| 269 |
+
"learning_rate": 4.759306803594352e-05,
|
| 270 |
+
"loss": 1.0348,
|
| 271 |
+
"step": 18500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.2,
|
| 275 |
+
"learning_rate": 4.752013070369938e-05,
|
| 276 |
+
"loss": 1.0228,
|
| 277 |
+
"step": 19000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.26,
|
| 281 |
+
"learning_rate": 4.744719337145524e-05,
|
| 282 |
+
"loss": 1.0503,
|
| 283 |
+
"step": 19500
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.32,
|
| 287 |
+
"learning_rate": 4.737425603921111e-05,
|
| 288 |
+
"loss": 1.0245,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.32,
|
| 293 |
+
"eval_bleu": 27.8764,
|
| 294 |
+
"eval_gen_len": 18.6046,
|
| 295 |
+
"eval_loss": 1.146370768547058,
|
| 296 |
+
"eval_runtime": 1146.7204,
|
| 297 |
+
"eval_samples_per_second": 15.032,
|
| 298 |
+
"eval_steps_per_second": 0.94,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.38,
|
| 303 |
+
"learning_rate": 4.730131870696697e-05,
|
| 304 |
+
"loss": 1.0493,
|
| 305 |
+
"step": 20500
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.44,
|
| 309 |
+
"learning_rate": 4.722838137472284e-05,
|
| 310 |
+
"loss": 1.0287,
|
| 311 |
+
"step": 21000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.49,
|
| 315 |
+
"learning_rate": 4.71554440424787e-05,
|
| 316 |
+
"loss": 1.03,
|
| 317 |
+
"step": 21500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.55,
|
| 321 |
+
"learning_rate": 4.708250671023457e-05,
|
| 322 |
+
"loss": 1.0354,
|
| 323 |
+
"step": 22000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.61,
|
| 327 |
+
"learning_rate": 4.700956937799043e-05,
|
| 328 |
+
"loss": 1.0268,
|
| 329 |
+
"step": 22500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.67,
|
| 333 |
+
"learning_rate": 4.69366320457463e-05,
|
| 334 |
+
"loss": 1.039,
|
| 335 |
+
"step": 23000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 2.73,
|
| 339 |
+
"learning_rate": 4.686369471350216e-05,
|
| 340 |
+
"loss": 1.0343,
|
| 341 |
+
"step": 23500
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 2.78,
|
| 345 |
+
"learning_rate": 4.679075738125803e-05,
|
| 346 |
+
"loss": 1.0186,
|
| 347 |
+
"step": 24000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.78,
|
| 351 |
+
"eval_bleu": 28.4585,
|
| 352 |
+
"eval_gen_len": 18.6731,
|
| 353 |
+
"eval_loss": 1.1286325454711914,
|
| 354 |
+
"eval_runtime": 1151.5333,
|
| 355 |
+
"eval_samples_per_second": 14.969,
|
| 356 |
+
"eval_steps_per_second": 0.936,
|
| 357 |
+
"step": 24000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.84,
|
| 361 |
+
"learning_rate": 4.671782004901389e-05,
|
| 362 |
+
"loss": 1.0383,
|
| 363 |
+
"step": 24500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.9,
|
| 367 |
+
"learning_rate": 4.664488271676976e-05,
|
| 368 |
+
"loss": 1.0488,
|
| 369 |
+
"step": 25000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.96,
|
| 373 |
+
"learning_rate": 4.657194538452562e-05,
|
| 374 |
+
"loss": 1.0572,
|
| 375 |
+
"step": 25500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.02,
|
| 379 |
+
"learning_rate": 4.649900805228149e-05,
|
| 380 |
+
"loss": 1.0168,
|
| 381 |
+
"step": 26000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.07,
|
| 385 |
+
"learning_rate": 4.642607072003735e-05,
|
| 386 |
+
"loss": 0.9238,
|
| 387 |
+
"step": 26500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.13,
|
| 391 |
+
"learning_rate": 4.635313338779321e-05,
|
| 392 |
+
"loss": 0.9281,
|
| 393 |
+
"step": 27000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.19,
|
| 397 |
+
"learning_rate": 4.628019605554907e-05,
|
| 398 |
+
"loss": 0.9225,
|
| 399 |
+
"step": 27500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 3.25,
|
| 403 |
+
"learning_rate": 4.620725872330493e-05,
|
| 404 |
+
"loss": 0.9245,
|
| 405 |
+
"step": 28000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 3.25,
|
| 409 |
+
"eval_bleu": 28.4834,
|
| 410 |
+
"eval_gen_len": 18.5428,
|
| 411 |
+
"eval_loss": 1.1263633966445923,
|
| 412 |
+
"eval_runtime": 1147.1298,
|
| 413 |
+
"eval_samples_per_second": 15.026,
|
| 414 |
+
"eval_steps_per_second": 0.94,
|
| 415 |
+
"step": 28000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 3.31,
|
| 419 |
+
"learning_rate": 4.61343213910608e-05,
|
| 420 |
+
"loss": 0.9145,
|
| 421 |
+
"step": 28500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 3.36,
|
| 425 |
+
"learning_rate": 4.606138405881666e-05,
|
| 426 |
+
"loss": 0.9228,
|
| 427 |
+
"step": 29000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.42,
|
| 431 |
+
"learning_rate": 4.598844672657253e-05,
|
| 432 |
+
"loss": 0.9196,
|
| 433 |
+
"step": 29500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 3.48,
|
| 437 |
+
"learning_rate": 4.591550939432839e-05,
|
| 438 |
+
"loss": 0.9138,
|
| 439 |
+
"step": 30000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 3.54,
|
| 443 |
+
"learning_rate": 4.584257206208426e-05,
|
| 444 |
+
"loss": 0.9274,
|
| 445 |
+
"step": 30500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 3.6,
|
| 449 |
+
"learning_rate": 4.576963472984012e-05,
|
| 450 |
+
"loss": 0.9077,
|
| 451 |
+
"step": 31000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.65,
|
| 455 |
+
"learning_rate": 4.569669739759599e-05,
|
| 456 |
+
"loss": 0.9312,
|
| 457 |
+
"step": 31500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.71,
|
| 461 |
+
"learning_rate": 4.562376006535185e-05,
|
| 462 |
+
"loss": 0.9343,
|
| 463 |
+
"step": 32000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 3.71,
|
| 467 |
+
"eval_bleu": 28.8235,
|
| 468 |
+
"eval_gen_len": 18.7833,
|
| 469 |
+
"eval_loss": 1.1182423830032349,
|
| 470 |
+
"eval_runtime": 1155.993,
|
| 471 |
+
"eval_samples_per_second": 14.911,
|
| 472 |
+
"eval_steps_per_second": 0.933,
|
| 473 |
+
"step": 32000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.77,
|
| 477 |
+
"learning_rate": 4.555082273310772e-05,
|
| 478 |
+
"loss": 0.9232,
|
| 479 |
+
"step": 32500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.83,
|
| 483 |
+
"learning_rate": 4.547788540086358e-05,
|
| 484 |
+
"loss": 0.9308,
|
| 485 |
+
"step": 33000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.89,
|
| 489 |
+
"learning_rate": 4.540494806861945e-05,
|
| 490 |
+
"loss": 0.9361,
|
| 491 |
+
"step": 33500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.94,
|
| 495 |
+
"learning_rate": 4.533201073637531e-05,
|
| 496 |
+
"loss": 0.9139,
|
| 497 |
+
"step": 34000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.0,
|
| 501 |
+
"learning_rate": 4.525907340413118e-05,
|
| 502 |
+
"loss": 0.919,
|
| 503 |
+
"step": 34500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.06,
|
| 507 |
+
"learning_rate": 4.518613607188704e-05,
|
| 508 |
+
"loss": 0.8226,
|
| 509 |
+
"step": 35000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.12,
|
| 513 |
+
"learning_rate": 4.5113198739642907e-05,
|
| 514 |
+
"loss": 0.8242,
|
| 515 |
+
"step": 35500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 4.18,
|
| 519 |
+
"learning_rate": 4.504026140739877e-05,
|
| 520 |
+
"loss": 0.8215,
|
| 521 |
+
"step": 36000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 4.18,
|
| 525 |
+
"eval_bleu": 28.6134,
|
| 526 |
+
"eval_gen_len": 18.5656,
|
| 527 |
+
"eval_loss": 1.133116602897644,
|
| 528 |
+
"eval_runtime": 1144.5462,
|
| 529 |
+
"eval_samples_per_second": 15.06,
|
| 530 |
+
"eval_steps_per_second": 0.942,
|
| 531 |
+
"step": 36000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.23,
|
| 535 |
+
"learning_rate": 4.496732407515463e-05,
|
| 536 |
+
"loss": 0.833,
|
| 537 |
+
"step": 36500
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 4.29,
|
| 541 |
+
"learning_rate": 4.489438674291049e-05,
|
| 542 |
+
"loss": 0.8175,
|
| 543 |
+
"step": 37000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 4.35,
|
| 547 |
+
"learning_rate": 4.482144941066635e-05,
|
| 548 |
+
"loss": 0.8313,
|
| 549 |
+
"step": 37500
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 4.41,
|
| 553 |
+
"learning_rate": 4.474851207842222e-05,
|
| 554 |
+
"loss": 0.832,
|
| 555 |
+
"step": 38000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 4.47,
|
| 559 |
+
"learning_rate": 4.467557474617808e-05,
|
| 560 |
+
"loss": 0.8397,
|
| 561 |
+
"step": 38500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 4.52,
|
| 565 |
+
"learning_rate": 4.460263741393395e-05,
|
| 566 |
+
"loss": 0.8324,
|
| 567 |
+
"step": 39000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 4.58,
|
| 571 |
+
"learning_rate": 4.452970008168981e-05,
|
| 572 |
+
"loss": 0.8379,
|
| 573 |
+
"step": 39500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 4.64,
|
| 577 |
+
"learning_rate": 4.445676274944568e-05,
|
| 578 |
+
"loss": 0.8456,
|
| 579 |
+
"step": 40000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 4.64,
|
| 583 |
+
"eval_bleu": 28.7324,
|
| 584 |
+
"eval_gen_len": 18.459,
|
| 585 |
+
"eval_loss": 1.1203099489212036,
|
| 586 |
+
"eval_runtime": 1135.006,
|
| 587 |
+
"eval_samples_per_second": 15.187,
|
| 588 |
+
"eval_steps_per_second": 0.95,
|
| 589 |
+
"step": 40000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 4.7,
|
| 593 |
+
"learning_rate": 4.438382541720154e-05,
|
| 594 |
+
"loss": 0.8481,
|
| 595 |
+
"step": 40500
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 4.76,
|
| 599 |
+
"learning_rate": 4.431088808495741e-05,
|
| 600 |
+
"loss": 0.8376,
|
| 601 |
+
"step": 41000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 4.81,
|
| 605 |
+
"learning_rate": 4.423795075271327e-05,
|
| 606 |
+
"loss": 0.8416,
|
| 607 |
+
"step": 41500
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 4.87,
|
| 611 |
+
"learning_rate": 4.416501342046914e-05,
|
| 612 |
+
"loss": 0.8462,
|
| 613 |
+
"step": 42000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 4.93,
|
| 617 |
+
"learning_rate": 4.4092076088225e-05,
|
| 618 |
+
"loss": 0.846,
|
| 619 |
+
"step": 42500
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 4.99,
|
| 623 |
+
"learning_rate": 4.401913875598087e-05,
|
| 624 |
+
"loss": 0.8429,
|
| 625 |
+
"step": 43000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 5.05,
|
| 629 |
+
"learning_rate": 4.394620142373673e-05,
|
| 630 |
+
"loss": 0.7648,
|
| 631 |
+
"step": 43500
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 5.11,
|
| 635 |
+
"learning_rate": 4.3873264091492596e-05,
|
| 636 |
+
"loss": 0.7437,
|
| 637 |
+
"step": 44000
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.11,
|
| 641 |
+
"eval_bleu": 28.7297,
|
| 642 |
+
"eval_gen_len": 18.7835,
|
| 643 |
+
"eval_loss": 1.1458019018173218,
|
| 644 |
+
"eval_runtime": 1166.3289,
|
| 645 |
+
"eval_samples_per_second": 14.779,
|
| 646 |
+
"eval_steps_per_second": 0.924,
|
| 647 |
+
"step": 44000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 5.16,
|
| 651 |
+
"learning_rate": 4.380032675924846e-05,
|
| 652 |
+
"loss": 0.7562,
|
| 653 |
+
"step": 44500
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 5.22,
|
| 657 |
+
"learning_rate": 4.3727389427004326e-05,
|
| 658 |
+
"loss": 0.752,
|
| 659 |
+
"step": 45000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 5.28,
|
| 663 |
+
"learning_rate": 4.365445209476019e-05,
|
| 664 |
+
"loss": 0.7526,
|
| 665 |
+
"step": 45500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.34,
|
| 669 |
+
"learning_rate": 4.358151476251605e-05,
|
| 670 |
+
"loss": 0.7779,
|
| 671 |
+
"step": 46000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 5.4,
|
| 675 |
+
"learning_rate": 4.350857743027191e-05,
|
| 676 |
+
"loss": 0.7942,
|
| 677 |
+
"step": 46500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 5.45,
|
| 681 |
+
"learning_rate": 4.343564009802778e-05,
|
| 682 |
+
"loss": 0.7954,
|
| 683 |
+
"step": 47000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 5.51,
|
| 687 |
+
"learning_rate": 4.336270276578364e-05,
|
| 688 |
+
"loss": 0.7874,
|
| 689 |
+
"step": 47500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 5.57,
|
| 693 |
+
"learning_rate": 4.32897654335395e-05,
|
| 694 |
+
"loss": 0.7829,
|
| 695 |
+
"step": 48000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.57,
|
| 699 |
+
"eval_bleu": 28.8328,
|
| 700 |
+
"eval_gen_len": 18.6052,
|
| 701 |
+
"eval_loss": 1.1367288827896118,
|
| 702 |
+
"eval_runtime": 1145.791,
|
| 703 |
+
"eval_samples_per_second": 15.044,
|
| 704 |
+
"eval_steps_per_second": 0.941,
|
| 705 |
+
"step": 48000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 5.63,
|
| 709 |
+
"learning_rate": 4.321682810129537e-05,
|
| 710 |
+
"loss": 0.7853,
|
| 711 |
+
"step": 48500
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 5.69,
|
| 715 |
+
"learning_rate": 4.314389076905123e-05,
|
| 716 |
+
"loss": 0.7916,
|
| 717 |
+
"step": 49000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 5.74,
|
| 721 |
+
"learning_rate": 4.30709534368071e-05,
|
| 722 |
+
"loss": 0.7914,
|
| 723 |
+
"step": 49500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 5.8,
|
| 727 |
+
"learning_rate": 4.299801610456296e-05,
|
| 728 |
+
"loss": 0.8096,
|
| 729 |
+
"step": 50000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 5.86,
|
| 733 |
+
"learning_rate": 4.292507877231883e-05,
|
| 734 |
+
"loss": 0.8128,
|
| 735 |
+
"step": 50500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 5.92,
|
| 739 |
+
"learning_rate": 4.285214144007469e-05,
|
| 740 |
+
"loss": 0.7987,
|
| 741 |
+
"step": 51000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 5.98,
|
| 745 |
+
"learning_rate": 4.2779204107830557e-05,
|
| 746 |
+
"loss": 0.8185,
|
| 747 |
+
"step": 51500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 6.03,
|
| 751 |
+
"learning_rate": 4.270626677558642e-05,
|
| 752 |
+
"loss": 0.7434,
|
| 753 |
+
"step": 52000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 6.03,
|
| 757 |
+
"eval_bleu": 28.2106,
|
| 758 |
+
"eval_gen_len": 18.4871,
|
| 759 |
+
"eval_loss": 1.1697230339050293,
|
| 760 |
+
"eval_runtime": 1139.5757,
|
| 761 |
+
"eval_samples_per_second": 15.126,
|
| 762 |
+
"eval_steps_per_second": 0.946,
|
| 763 |
+
"step": 52000
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 6.09,
|
| 767 |
+
"learning_rate": 4.2633329443342286e-05,
|
| 768 |
+
"loss": 0.6938,
|
| 769 |
+
"step": 52500
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 6.15,
|
| 773 |
+
"learning_rate": 4.256039211109815e-05,
|
| 774 |
+
"loss": 0.695,
|
| 775 |
+
"step": 53000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 6.21,
|
| 779 |
+
"learning_rate": 4.2487454778854015e-05,
|
| 780 |
+
"loss": 0.7016,
|
| 781 |
+
"step": 53500
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 6.27,
|
| 785 |
+
"learning_rate": 4.241451744660988e-05,
|
| 786 |
+
"loss": 0.6836,
|
| 787 |
+
"step": 54000
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 6.32,
|
| 791 |
+
"learning_rate": 4.234158011436574e-05,
|
| 792 |
+
"loss": 0.6807,
|
| 793 |
+
"step": 54500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 6.38,
|
| 797 |
+
"learning_rate": 4.2268642782121606e-05,
|
| 798 |
+
"loss": 0.674,
|
| 799 |
+
"step": 55000
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 6.44,
|
| 803 |
+
"learning_rate": 4.219570544987747e-05,
|
| 804 |
+
"loss": 0.6897,
|
| 805 |
+
"step": 55500
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 6.5,
|
| 809 |
+
"learning_rate": 4.2122768117633335e-05,
|
| 810 |
+
"loss": 0.7153,
|
| 811 |
+
"step": 56000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 6.5,
|
| 815 |
+
"eval_bleu": 28.1455,
|
| 816 |
+
"eval_gen_len": 18.7413,
|
| 817 |
+
"eval_loss": 1.1770974397659302,
|
| 818 |
+
"eval_runtime": 1156.7175,
|
| 819 |
+
"eval_samples_per_second": 14.902,
|
| 820 |
+
"eval_steps_per_second": 0.932,
|
| 821 |
+
"step": 56000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 6.56,
|
| 825 |
+
"learning_rate": 4.20498307853892e-05,
|
| 826 |
+
"loss": 0.7177,
|
| 827 |
+
"step": 56500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 6.61,
|
| 831 |
+
"learning_rate": 4.197689345314506e-05,
|
| 832 |
+
"loss": 0.716,
|
| 833 |
+
"step": 57000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 6.67,
|
| 837 |
+
"learning_rate": 4.190395612090092e-05,
|
| 838 |
+
"loss": 0.721,
|
| 839 |
+
"step": 57500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.73,
|
| 843 |
+
"learning_rate": 4.183101878865679e-05,
|
| 844 |
+
"loss": 0.7152,
|
| 845 |
+
"step": 58000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 6.79,
|
| 849 |
+
"learning_rate": 4.175808145641265e-05,
|
| 850 |
+
"loss": 0.7012,
|
| 851 |
+
"step": 58500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 6.85,
|
| 855 |
+
"learning_rate": 4.168514412416852e-05,
|
| 856 |
+
"loss": 0.7069,
|
| 857 |
+
"step": 59000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 6.9,
|
| 861 |
+
"learning_rate": 4.161220679192438e-05,
|
| 862 |
+
"loss": 0.7088,
|
| 863 |
+
"step": 59500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 6.96,
|
| 867 |
+
"learning_rate": 4.1539269459680246e-05,
|
| 868 |
+
"loss": 0.6996,
|
| 869 |
+
"step": 60000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.96,
|
| 873 |
+
"eval_bleu": 29.2694,
|
| 874 |
+
"eval_gen_len": 18.5162,
|
| 875 |
+
"eval_loss": 1.1514174938201904,
|
| 876 |
+
"eval_runtime": 1140.5069,
|
| 877 |
+
"eval_samples_per_second": 15.113,
|
| 878 |
+
"eval_steps_per_second": 0.945,
|
| 879 |
+
"step": 60000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 7.02,
|
| 883 |
+
"learning_rate": 4.146633212743611e-05,
|
| 884 |
+
"loss": 0.6796,
|
| 885 |
+
"step": 60500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 7.08,
|
| 889 |
+
"learning_rate": 4.1393394795191976e-05,
|
| 890 |
+
"loss": 0.6009,
|
| 891 |
+
"step": 61000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 7.14,
|
| 895 |
+
"learning_rate": 4.132045746294784e-05,
|
| 896 |
+
"loss": 0.5985,
|
| 897 |
+
"step": 61500
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 7.19,
|
| 901 |
+
"learning_rate": 4.1247520130703705e-05,
|
| 902 |
+
"loss": 0.5966,
|
| 903 |
+
"step": 62000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 7.25,
|
| 907 |
+
"learning_rate": 4.1174582798459566e-05,
|
| 908 |
+
"loss": 0.6132,
|
| 909 |
+
"step": 62500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 7.31,
|
| 913 |
+
"learning_rate": 4.110164546621543e-05,
|
| 914 |
+
"loss": 0.6048,
|
| 915 |
+
"step": 63000
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 7.37,
|
| 919 |
+
"learning_rate": 4.1028708133971296e-05,
|
| 920 |
+
"loss": 0.6038,
|
| 921 |
+
"step": 63500
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 7.43,
|
| 925 |
+
"learning_rate": 4.095577080172716e-05,
|
| 926 |
+
"loss": 0.6336,
|
| 927 |
+
"step": 64000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 7.43,
|
| 931 |
+
"eval_bleu": 28.1465,
|
| 932 |
+
"eval_gen_len": 18.5439,
|
| 933 |
+
"eval_loss": 1.2213271856307983,
|
| 934 |
+
"eval_runtime": 1140.4863,
|
| 935 |
+
"eval_samples_per_second": 15.114,
|
| 936 |
+
"eval_steps_per_second": 0.945,
|
| 937 |
+
"step": 64000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 7.48,
|
| 941 |
+
"learning_rate": 4.0882833469483025e-05,
|
| 942 |
+
"loss": 0.6943,
|
| 943 |
+
"step": 64500
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 7.54,
|
| 947 |
+
"learning_rate": 4.0809896137238886e-05,
|
| 948 |
+
"loss": 0.7508,
|
| 949 |
+
"step": 65000
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 7.6,
|
| 953 |
+
"learning_rate": 4.0736958804994754e-05,
|
| 954 |
+
"loss": 0.7606,
|
| 955 |
+
"step": 65500
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 7.66,
|
| 959 |
+
"learning_rate": 4.0664021472750616e-05,
|
| 960 |
+
"loss": 0.7578,
|
| 961 |
+
"step": 66000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 7.72,
|
| 965 |
+
"learning_rate": 4.059108414050648e-05,
|
| 966 |
+
"loss": 0.751,
|
| 967 |
+
"step": 66500
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 7.77,
|
| 971 |
+
"learning_rate": 4.051814680826234e-05,
|
| 972 |
+
"loss": 0.738,
|
| 973 |
+
"step": 67000
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 7.83,
|
| 977 |
+
"learning_rate": 4.0445209476018206e-05,
|
| 978 |
+
"loss": 0.7185,
|
| 979 |
+
"step": 67500
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 7.89,
|
| 983 |
+
"learning_rate": 4.037227214377407e-05,
|
| 984 |
+
"loss": 0.7218,
|
| 985 |
+
"step": 68000
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 7.89,
|
| 989 |
+
"eval_bleu": 28.2245,
|
| 990 |
+
"eval_gen_len": 18.5246,
|
| 991 |
+
"eval_loss": 1.1834964752197266,
|
| 992 |
+
"eval_runtime": 1143.8445,
|
| 993 |
+
"eval_samples_per_second": 15.069,
|
| 994 |
+
"eval_steps_per_second": 0.942,
|
| 995 |
+
"step": 68000
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 7.95,
|
| 999 |
+
"learning_rate": 4.0299334811529936e-05,
|
| 1000 |
+
"loss": 0.723,
|
| 1001 |
+
"step": 68500
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 8.01,
|
| 1005 |
+
"learning_rate": 4.02263974792858e-05,
|
| 1006 |
+
"loss": 0.7128,
|
| 1007 |
+
"step": 69000
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 8.06,
|
| 1011 |
+
"learning_rate": 4.0153460147041665e-05,
|
| 1012 |
+
"loss": 0.5863,
|
| 1013 |
+
"step": 69500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 8.12,
|
| 1017 |
+
"learning_rate": 4.0080522814797527e-05,
|
| 1018 |
+
"loss": 0.5939,
|
| 1019 |
+
"step": 70000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 8.18,
|
| 1023 |
+
"learning_rate": 4.0007585482553395e-05,
|
| 1024 |
+
"loss": 0.6018,
|
| 1025 |
+
"step": 70500
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 8.24,
|
| 1029 |
+
"learning_rate": 3.9934648150309256e-05,
|
| 1030 |
+
"loss": 0.588,
|
| 1031 |
+
"step": 71000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 8.3,
|
| 1035 |
+
"learning_rate": 3.986171081806512e-05,
|
| 1036 |
+
"loss": 0.5816,
|
| 1037 |
+
"step": 71500
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 8.35,
|
| 1041 |
+
"learning_rate": 3.9788773485820985e-05,
|
| 1042 |
+
"loss": 0.5934,
|
| 1043 |
+
"step": 72000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 8.35,
|
| 1047 |
+
"eval_bleu": 28.3836,
|
| 1048 |
+
"eval_gen_len": 18.6717,
|
| 1049 |
+
"eval_loss": 1.2387008666992188,
|
| 1050 |
+
"eval_runtime": 1156.4248,
|
| 1051 |
+
"eval_samples_per_second": 14.905,
|
| 1052 |
+
"eval_steps_per_second": 0.932,
|
| 1053 |
+
"step": 72000
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 8.41,
|
| 1057 |
+
"learning_rate": 3.971583615357685e-05,
|
| 1058 |
+
"loss": 0.5772,
|
| 1059 |
+
"step": 72500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 8.47,
|
| 1063 |
+
"learning_rate": 3.9642898821332715e-05,
|
| 1064 |
+
"loss": 0.576,
|
| 1065 |
+
"step": 73000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 8.53,
|
| 1069 |
+
"learning_rate": 3.9569961489088576e-05,
|
| 1070 |
+
"loss": 0.5707,
|
| 1071 |
+
"step": 73500
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 8.59,
|
| 1075 |
+
"learning_rate": 3.9497024156844444e-05,
|
| 1076 |
+
"loss": 0.5771,
|
| 1077 |
+
"step": 74000
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 8.64,
|
| 1081 |
+
"learning_rate": 3.9424086824600305e-05,
|
| 1082 |
+
"loss": 0.5699,
|
| 1083 |
+
"step": 74500
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 8.7,
|
| 1087 |
+
"learning_rate": 3.9351149492356174e-05,
|
| 1088 |
+
"loss": 0.5673,
|
| 1089 |
+
"step": 75000
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 8.76,
|
| 1093 |
+
"learning_rate": 3.9278212160112035e-05,
|
| 1094 |
+
"loss": 0.5779,
|
| 1095 |
+
"step": 75500
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 8.82,
|
| 1099 |
+
"learning_rate": 3.9205274827867896e-05,
|
| 1100 |
+
"loss": 0.5723,
|
| 1101 |
+
"step": 76000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 8.82,
|
| 1105 |
+
"eval_bleu": 28.5925,
|
| 1106 |
+
"eval_gen_len": 18.5566,
|
| 1107 |
+
"eval_loss": 1.232297658920288,
|
| 1108 |
+
"eval_runtime": 1148.9097,
|
| 1109 |
+
"eval_samples_per_second": 15.003,
|
| 1110 |
+
"eval_steps_per_second": 0.938,
|
| 1111 |
+
"step": 76000
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 8.82,
|
| 1115 |
+
"step": 76000,
|
| 1116 |
+
"total_flos": 2.635368494115324e+18,
|
| 1117 |
+
"train_loss": 0.5245111798738178,
|
| 1118 |
+
"train_runtime": 53209.1351,
|
| 1119 |
+
"train_samples_per_second": 103.666,
|
| 1120 |
+
"train_steps_per_second": 6.479
|
| 1121 |
+
}
|
| 1122 |
+
],
|
| 1123 |
+
"logging_steps": 500,
|
| 1124 |
+
"max_steps": 344760,
|
| 1125 |
+
"num_train_epochs": 40,
|
| 1126 |
+
"save_steps": 4000,
|
| 1127 |
+
"total_flos": 2.635368494115324e+18,
|
| 1128 |
+
"trial_name": null,
|
| 1129 |
+
"trial_params": null
|
| 1130 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e917f03bf6223d6917033778988c8bd45cc9111d9f4fed1584656468d3a4d435
|
| 3 |
+
size 4792
|