Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +4 -0
- README.md +77 -0
- all_results.json +22 -0
- checkpoint-25000/config.json +59 -0
- checkpoint-25000/generation_config.json +12 -0
- checkpoint-25000/optimizer.pt +3 -0
- checkpoint-25000/pytorch_model.bin +3 -0
- checkpoint-25000/rng_state_0.pth +3 -0
- checkpoint-25000/rng_state_1.pth +3 -0
- checkpoint-25000/rng_state_2.pth +3 -0
- checkpoint-25000/rng_state_3.pth +3 -0
- checkpoint-25000/scheduler.pt +3 -0
- checkpoint-25000/sentencepiece.bpe.model +3 -0
- checkpoint-25000/special_tokens_map.json +69 -0
- checkpoint-25000/tokenizer.json +3 -0
- checkpoint-25000/tokenizer_config.json +528 -0
- checkpoint-25000/trainer_state.json +369 -0
- checkpoint-25000/training_args.bin +3 -0
- checkpoint-40000/config.json +59 -0
- checkpoint-40000/generation_config.json +12 -0
- checkpoint-40000/optimizer.pt +3 -0
- checkpoint-40000/pytorch_model.bin +3 -0
- checkpoint-40000/rng_state_0.pth +3 -0
- checkpoint-40000/rng_state_1.pth +3 -0
- checkpoint-40000/rng_state_2.pth +3 -0
- checkpoint-40000/rng_state_3.pth +3 -0
- checkpoint-40000/scheduler.pt +3 -0
- checkpoint-40000/sentencepiece.bpe.model +3 -0
- checkpoint-40000/special_tokens_map.json +69 -0
- checkpoint-40000/tokenizer.json +3 -0
- checkpoint-40000/tokenizer_config.json +528 -0
- checkpoint-40000/trainer_state.json +579 -0
- checkpoint-40000/training_args.bin +3 -0
- checkpoint-45000/config.json +59 -0
- checkpoint-45000/generation_config.json +12 -0
- checkpoint-45000/optimizer.pt +3 -0
- checkpoint-45000/pytorch_model.bin +3 -0
- checkpoint-45000/rng_state_0.pth +3 -0
- checkpoint-45000/rng_state_1.pth +3 -0
- checkpoint-45000/rng_state_2.pth +3 -0
- checkpoint-45000/rng_state_3.pth +3 -0
- checkpoint-45000/scheduler.pt +3 -0
- checkpoint-45000/sentencepiece.bpe.model +3 -0
- checkpoint-45000/special_tokens_map.json +69 -0
- checkpoint-45000/tokenizer.json +3 -0
- checkpoint-45000/tokenizer_config.json +528 -0
- checkpoint-45000/trainer_state.json +649 -0
- checkpoint-45000/training_args.bin +3 -0
- config.json +59 -0
- eval_results.json +10 -0
.gitattributes
CHANGED
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
checkpoint-25000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
37 |
+
checkpoint-40000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
38 |
+
checkpoint-45000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
39 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
- ko
|
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: enko_mbartLarge_36p_exp1
|
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 |
+
# enko_mbartLarge_36p_exp1
|
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: 1.2181
|
23 |
+
- Bleu: 15.4063
|
24 |
+
- Gen Len: 14.7808
|
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: 500
|
55 |
+
- num_epochs: 15
|
56 |
+
|
57 |
+
### Training results
|
58 |
+
|
59 |
+
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|
60 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
|
61 |
+
| 1.4235 | 0.46 | 5000 | 1.3893 | 12.3168 | 14.6634 |
|
62 |
+
| 1.3281 | 0.93 | 10000 | 1.2917 | 14.3522 | 14.9186 |
|
63 |
+
| 1.2506 | 1.39 | 15000 | 1.2669 | 14.3525 | 14.9494 |
|
64 |
+
| 1.1603 | 1.86 | 20000 | 1.2283 | 15.248 | 15.0062 |
|
65 |
+
| 1.0765 | 2.32 | 25000 | 1.2181 | 15.4063 | 14.7808 |
|
66 |
+
| 1.1019 | 2.79 | 30000 | 1.2753 | 14.3608 | 14.9014 |
|
67 |
+
| 1.0504 | 3.25 | 35000 | 1.2334 | 15.3253 | 14.7948 |
|
68 |
+
| 0.9431 | 3.72 | 40000 | 1.2512 | 15.2534 | 14.7293 |
|
69 |
+
| 0.8394 | 4.18 | 45000 | 1.2971 | 14.9999 | 14.7993 |
|
70 |
+
|
71 |
+
|
72 |
+
### Framework versions
|
73 |
+
|
74 |
+
- Transformers 4.34.1
|
75 |
+
- Pytorch 2.1.0+cu121
|
76 |
+
- Datasets 2.14.6
|
77 |
+
- Tokenizers 0.14.1
|
all_results.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.18,
|
3 |
+
"eval_bleu": 15.4063,
|
4 |
+
"eval_gen_len": 14.7808,
|
5 |
+
"eval_loss": 1.2180595397949219,
|
6 |
+
"eval_runtime": 1209.5725,
|
7 |
+
"eval_samples": 21605,
|
8 |
+
"eval_samples_per_second": 17.862,
|
9 |
+
"eval_steps_per_second": 1.117,
|
10 |
+
"predict_bleu": 16.163,
|
11 |
+
"predict_gen_len": 14.8185,
|
12 |
+
"predict_loss": 1.214425802230835,
|
13 |
+
"predict_runtime": 1210.3365,
|
14 |
+
"predict_samples": 21606,
|
15 |
+
"predict_samples_per_second": 17.851,
|
16 |
+
"predict_steps_per_second": 1.116,
|
17 |
+
"train_loss": 1.1734986402723524,
|
18 |
+
"train_runtime": 45995.7579,
|
19 |
+
"train_samples": 172202,
|
20 |
+
"train_samples_per_second": 56.158,
|
21 |
+
"train_steps_per_second": 3.51
|
22 |
+
}
|
checkpoint-25000/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": 250014,
|
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": "4.34.1",
|
57 |
+
"use_cache": true,
|
58 |
+
"vocab_size": 250054
|
59 |
+
}
|
checkpoint-25000/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": 250014,
|
7 |
+
"forced_eos_token_id": 2,
|
8 |
+
"max_length": 200,
|
9 |
+
"num_beams": 5,
|
10 |
+
"pad_token_id": 1,
|
11 |
+
"transformers_version": "4.34.1"
|
12 |
+
}
|
checkpoint-25000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eab12bdd09775b2fb75b11dc86debe61be53b4d0b675dd9952cc4b866b2c75ac
|
3 |
+
size 4887473903
|
checkpoint-25000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:157f1afc3a280a074a2f75ed004ebd020f535d14c7fb525bb62aa4c06e1cfbf5
|
3 |
+
size 2444694490
|
checkpoint-25000/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da595bf6989c5e0d05b21a9b2b4d56b0c953195d5e11cb0cb3a537b6e6f9d5fe
|
3 |
+
size 15024
|
checkpoint-25000/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2af25949f29ba20e191f1a0dcc3bdf0b1771fbd959d5e089909092d02f50dda7
|
3 |
+
size 15024
|
checkpoint-25000/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fecf21137b178a5796990ba366fe5e51c9fe02110c48bed8ca72240a60f9746d
|
3 |
+
size 15024
|
checkpoint-25000/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ceddac6bd5b441e3b774f1f8b0a5e8af4d7447bd8ecc6e161e01d03ddcd7aaf9
|
3 |
+
size 15024
|
checkpoint-25000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4170c4c41f64b0234023cba49cf610b9e09d8dd8305a8f01091e4cac8aa962c8
|
3 |
+
size 1064
|
checkpoint-25000/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-25000/special_tokens_map.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
checkpoint-25000/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:55c9c06dcc44c6b9872f61d601f88b88ac893ad5a3a77c035c1e1f2a69acdc27
|
3 |
+
size 17110043
|
checkpoint-25000/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": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
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": "en_XX",
|
525 |
+
"tgt_lang": "ko_KR",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
checkpoint-25000/trainer_state.json
ADDED
@@ -0,0 +1,369 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.2180595397949219,
|
3 |
+
"best_model_checkpoint": "./enko_mbartLarge_36p_exp1/checkpoint-25000",
|
4 |
+
"epoch": 2.3227724612097,
|
5 |
+
"eval_steps": 5000,
|
6 |
+
"global_step": 25000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"learning_rate": 5e-05,
|
14 |
+
"loss": 2.3583,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.09,
|
19 |
+
"learning_rate": 4.9844667432974e-05,
|
20 |
+
"loss": 1.7529,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.14,
|
25 |
+
"learning_rate": 4.9689334865948e-05,
|
26 |
+
"loss": 1.5715,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.19,
|
31 |
+
"learning_rate": 4.9534002298921995e-05,
|
32 |
+
"loss": 1.535,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.23,
|
37 |
+
"learning_rate": 4.937866973189599e-05,
|
38 |
+
"loss": 1.5009,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.28,
|
43 |
+
"learning_rate": 4.9223337164869984e-05,
|
44 |
+
"loss": 1.4783,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.33,
|
49 |
+
"learning_rate": 4.906800459784399e-05,
|
50 |
+
"loss": 1.4542,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.37,
|
55 |
+
"learning_rate": 4.8912672030817986e-05,
|
56 |
+
"loss": 1.441,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.42,
|
61 |
+
"learning_rate": 4.875733946379198e-05,
|
62 |
+
"loss": 1.425,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.46,
|
67 |
+
"learning_rate": 4.8602006896765975e-05,
|
68 |
+
"loss": 1.4235,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46,
|
73 |
+
"eval_bleu": 12.3168,
|
74 |
+
"eval_gen_len": 14.6634,
|
75 |
+
"eval_loss": 1.3893214464187622,
|
76 |
+
"eval_runtime": 1225.507,
|
77 |
+
"eval_samples_per_second": 17.629,
|
78 |
+
"eval_steps_per_second": 1.102,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.51,
|
83 |
+
"learning_rate": 4.844667432973998e-05,
|
84 |
+
"loss": 1.3969,
|
85 |
+
"step": 5500
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 0.56,
|
89 |
+
"learning_rate": 4.829134176271397e-05,
|
90 |
+
"loss": 1.3901,
|
91 |
+
"step": 6000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.6,
|
95 |
+
"learning_rate": 4.813600919568797e-05,
|
96 |
+
"loss": 1.3983,
|
97 |
+
"step": 6500
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 0.65,
|
101 |
+
"learning_rate": 4.798067662866197e-05,
|
102 |
+
"loss": 1.3744,
|
103 |
+
"step": 7000
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 0.7,
|
107 |
+
"learning_rate": 4.782534406163596e-05,
|
108 |
+
"loss": 1.3712,
|
109 |
+
"step": 7500
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.74,
|
113 |
+
"learning_rate": 4.767001149460996e-05,
|
114 |
+
"loss": 1.3604,
|
115 |
+
"step": 8000
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.79,
|
119 |
+
"learning_rate": 4.751467892758396e-05,
|
120 |
+
"loss": 1.3653,
|
121 |
+
"step": 8500
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.84,
|
125 |
+
"learning_rate": 4.7359346360557956e-05,
|
126 |
+
"loss": 1.3603,
|
127 |
+
"step": 9000
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 0.88,
|
131 |
+
"learning_rate": 4.7204013793531954e-05,
|
132 |
+
"loss": 1.3188,
|
133 |
+
"step": 9500
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 0.93,
|
137 |
+
"learning_rate": 4.704868122650595e-05,
|
138 |
+
"loss": 1.3281,
|
139 |
+
"step": 10000
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 0.93,
|
143 |
+
"eval_bleu": 14.3522,
|
144 |
+
"eval_gen_len": 14.9186,
|
145 |
+
"eval_loss": 1.2917265892028809,
|
146 |
+
"eval_runtime": 1232.5734,
|
147 |
+
"eval_samples_per_second": 17.528,
|
148 |
+
"eval_steps_per_second": 1.096,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.98,
|
153 |
+
"learning_rate": 4.689334865947995e-05,
|
154 |
+
"loss": 1.3329,
|
155 |
+
"step": 10500
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.02,
|
159 |
+
"learning_rate": 4.673801609245395e-05,
|
160 |
+
"loss": 1.3462,
|
161 |
+
"step": 11000
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.07,
|
165 |
+
"learning_rate": 4.6582683525427945e-05,
|
166 |
+
"loss": 1.3051,
|
167 |
+
"step": 11500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.11,
|
171 |
+
"learning_rate": 4.6427350958401935e-05,
|
172 |
+
"loss": 1.2672,
|
173 |
+
"step": 12000
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.16,
|
177 |
+
"learning_rate": 4.627201839137594e-05,
|
178 |
+
"loss": 1.2606,
|
179 |
+
"step": 12500
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.21,
|
183 |
+
"learning_rate": 4.611668582434994e-05,
|
184 |
+
"loss": 1.268,
|
185 |
+
"step": 13000
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.25,
|
189 |
+
"learning_rate": 4.596135325732393e-05,
|
190 |
+
"loss": 1.2733,
|
191 |
+
"step": 13500
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.3,
|
195 |
+
"learning_rate": 4.5806020690297926e-05,
|
196 |
+
"loss": 1.2652,
|
197 |
+
"step": 14000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.35,
|
201 |
+
"learning_rate": 4.565068812327193e-05,
|
202 |
+
"loss": 1.2389,
|
203 |
+
"step": 14500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.39,
|
207 |
+
"learning_rate": 4.549535555624592e-05,
|
208 |
+
"loss": 1.2506,
|
209 |
+
"step": 15000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.39,
|
213 |
+
"eval_bleu": 14.3525,
|
214 |
+
"eval_gen_len": 14.9494,
|
215 |
+
"eval_loss": 1.2668566703796387,
|
216 |
+
"eval_runtime": 1245.6055,
|
217 |
+
"eval_samples_per_second": 17.345,
|
218 |
+
"eval_steps_per_second": 1.085,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.44,
|
223 |
+
"learning_rate": 4.534002298921992e-05,
|
224 |
+
"loss": 1.2423,
|
225 |
+
"step": 15500
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.49,
|
229 |
+
"learning_rate": 4.5184690422193924e-05,
|
230 |
+
"loss": 1.2214,
|
231 |
+
"step": 16000
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.53,
|
235 |
+
"learning_rate": 4.5029357855167915e-05,
|
236 |
+
"loss": 1.1984,
|
237 |
+
"step": 16500
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.58,
|
241 |
+
"learning_rate": 4.487402528814191e-05,
|
242 |
+
"loss": 1.1922,
|
243 |
+
"step": 17000
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.63,
|
247 |
+
"learning_rate": 4.471869272111591e-05,
|
248 |
+
"loss": 1.1952,
|
249 |
+
"step": 17500
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.67,
|
253 |
+
"learning_rate": 4.456336015408991e-05,
|
254 |
+
"loss": 1.1998,
|
255 |
+
"step": 18000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.72,
|
259 |
+
"learning_rate": 4.4408027587063905e-05,
|
260 |
+
"loss": 1.1676,
|
261 |
+
"step": 18500
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.77,
|
265 |
+
"learning_rate": 4.42526950200379e-05,
|
266 |
+
"loss": 1.175,
|
267 |
+
"step": 19000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 1.81,
|
271 |
+
"learning_rate": 4.40973624530119e-05,
|
272 |
+
"loss": 1.1674,
|
273 |
+
"step": 19500
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 1.86,
|
277 |
+
"learning_rate": 4.39420298859859e-05,
|
278 |
+
"loss": 1.1603,
|
279 |
+
"step": 20000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.86,
|
283 |
+
"eval_bleu": 15.248,
|
284 |
+
"eval_gen_len": 15.0062,
|
285 |
+
"eval_loss": 1.2283018827438354,
|
286 |
+
"eval_runtime": 1235.7777,
|
287 |
+
"eval_samples_per_second": 17.483,
|
288 |
+
"eval_steps_per_second": 1.093,
|
289 |
+
"step": 20000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 1.9,
|
293 |
+
"learning_rate": 4.3786697318959896e-05,
|
294 |
+
"loss": 1.1944,
|
295 |
+
"step": 20500
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 1.95,
|
299 |
+
"learning_rate": 4.3631364751933894e-05,
|
300 |
+
"loss": 1.1644,
|
301 |
+
"step": 21000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 2.0,
|
305 |
+
"learning_rate": 4.347603218490789e-05,
|
306 |
+
"loss": 1.137,
|
307 |
+
"step": 21500
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 2.04,
|
311 |
+
"learning_rate": 4.332069961788189e-05,
|
312 |
+
"loss": 1.1543,
|
313 |
+
"step": 22000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 2.09,
|
317 |
+
"learning_rate": 4.316536705085589e-05,
|
318 |
+
"loss": 1.1359,
|
319 |
+
"step": 22500
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 2.14,
|
323 |
+
"learning_rate": 4.301003448382988e-05,
|
324 |
+
"loss": 1.1373,
|
325 |
+
"step": 23000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 2.18,
|
329 |
+
"learning_rate": 4.285470191680388e-05,
|
330 |
+
"loss": 1.1229,
|
331 |
+
"step": 23500
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 2.23,
|
335 |
+
"learning_rate": 4.269936934977788e-05,
|
336 |
+
"loss": 1.1076,
|
337 |
+
"step": 24000
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 2.28,
|
341 |
+
"learning_rate": 4.254403678275187e-05,
|
342 |
+
"loss": 1.0849,
|
343 |
+
"step": 24500
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 2.32,
|
347 |
+
"learning_rate": 4.2388704215725875e-05,
|
348 |
+
"loss": 1.0765,
|
349 |
+
"step": 25000
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 2.32,
|
353 |
+
"eval_bleu": 15.4063,
|
354 |
+
"eval_gen_len": 14.7808,
|
355 |
+
"eval_loss": 1.2180595397949219,
|
356 |
+
"eval_runtime": 1212.0146,
|
357 |
+
"eval_samples_per_second": 17.826,
|
358 |
+
"eval_steps_per_second": 1.115,
|
359 |
+
"step": 25000
|
360 |
+
}
|
361 |
+
],
|
362 |
+
"logging_steps": 500,
|
363 |
+
"max_steps": 161445,
|
364 |
+
"num_train_epochs": 15,
|
365 |
+
"save_steps": 5000,
|
366 |
+
"total_flos": 8.670599521689928e+17,
|
367 |
+
"trial_name": null,
|
368 |
+
"trial_params": null
|
369 |
+
}
|
checkpoint-25000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae0f38c3140ce223a1dea38a47eeea582e94c01a40ce282a2ae0ea647f40877c
|
3 |
+
size 4728
|
checkpoint-40000/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": 250014,
|
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": "4.34.1",
|
57 |
+
"use_cache": true,
|
58 |
+
"vocab_size": 250054
|
59 |
+
}
|
checkpoint-40000/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": 250014,
|
7 |
+
"forced_eos_token_id": 2,
|
8 |
+
"max_length": 200,
|
9 |
+
"num_beams": 5,
|
10 |
+
"pad_token_id": 1,
|
11 |
+
"transformers_version": "4.34.1"
|
12 |
+
}
|
checkpoint-40000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3a50ac1d054b984a44edbef518aa5adc4d48b9fc0c967552151f1d9b1e8591e2
|
3 |
+
size 4887473903
|
checkpoint-40000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7fd9b74d8bb2a2865097cb42206c9266506311582015d45108e609f11e7ad62e
|
3 |
+
size 2444694490
|
checkpoint-40000/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e1d723ebcd7d2eada2e62e723bf7df26261806552e53f440789abe1961dcf4f
|
3 |
+
size 15024
|
checkpoint-40000/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ffd69e3745c044eeb2f87cf741da5b0c97e5ccf61f64148fe7448b27a69f522b
|
3 |
+
size 15024
|
checkpoint-40000/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5bcfa9e18a635df9462a7e6db397627347be031869a2803c6ad5a8a18d11a384
|
3 |
+
size 15024
|
checkpoint-40000/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1746f952926fb4190ded4ca637a4874eda199bf38c020a2e9042e561faa68f3a
|
3 |
+
size 15024
|
checkpoint-40000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:94d4750d600da7e0ada2a051d0e8a6c149fea33f5f41d27bab1dba7be57973a7
|
3 |
+
size 1064
|
checkpoint-40000/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-40000/special_tokens_map.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
checkpoint-40000/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:55c9c06dcc44c6b9872f61d601f88b88ac893ad5a3a77c035c1e1f2a69acdc27
|
3 |
+
size 17110043
|
checkpoint-40000/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": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
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": "en_XX",
|
525 |
+
"tgt_lang": "ko_KR",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
checkpoint-40000/trainer_state.json
ADDED
@@ -0,0 +1,579 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.2180595397949219,
|
3 |
+
"best_model_checkpoint": "./enko_mbartLarge_36p_exp1/checkpoint-25000",
|
4 |
+
"epoch": 3.71643593793552,
|
5 |
+
"eval_steps": 5000,
|
6 |
+
"global_step": 40000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"learning_rate": 5e-05,
|
14 |
+
"loss": 2.3583,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.09,
|
19 |
+
"learning_rate": 4.9844667432974e-05,
|
20 |
+
"loss": 1.7529,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.14,
|
25 |
+
"learning_rate": 4.9689334865948e-05,
|
26 |
+
"loss": 1.5715,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.19,
|
31 |
+
"learning_rate": 4.9534002298921995e-05,
|
32 |
+
"loss": 1.535,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.23,
|
37 |
+
"learning_rate": 4.937866973189599e-05,
|
38 |
+
"loss": 1.5009,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.28,
|
43 |
+
"learning_rate": 4.9223337164869984e-05,
|
44 |
+
"loss": 1.4783,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.33,
|
49 |
+
"learning_rate": 4.906800459784399e-05,
|
50 |
+
"loss": 1.4542,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.37,
|
55 |
+
"learning_rate": 4.8912672030817986e-05,
|
56 |
+
"loss": 1.441,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.42,
|
61 |
+
"learning_rate": 4.875733946379198e-05,
|
62 |
+
"loss": 1.425,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.46,
|
67 |
+
"learning_rate": 4.8602006896765975e-05,
|
68 |
+
"loss": 1.4235,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46,
|
73 |
+
"eval_bleu": 12.3168,
|
74 |
+
"eval_gen_len": 14.6634,
|
75 |
+
"eval_loss": 1.3893214464187622,
|
76 |
+
"eval_runtime": 1225.507,
|
77 |
+
"eval_samples_per_second": 17.629,
|
78 |
+
"eval_steps_per_second": 1.102,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.51,
|
83 |
+
"learning_rate": 4.844667432973998e-05,
|
84 |
+
"loss": 1.3969,
|
85 |
+
"step": 5500
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 0.56,
|
89 |
+
"learning_rate": 4.829134176271397e-05,
|
90 |
+
"loss": 1.3901,
|
91 |
+
"step": 6000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.6,
|
95 |
+
"learning_rate": 4.813600919568797e-05,
|
96 |
+
"loss": 1.3983,
|
97 |
+
"step": 6500
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 0.65,
|
101 |
+
"learning_rate": 4.798067662866197e-05,
|
102 |
+
"loss": 1.3744,
|
103 |
+
"step": 7000
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 0.7,
|
107 |
+
"learning_rate": 4.782534406163596e-05,
|
108 |
+
"loss": 1.3712,
|
109 |
+
"step": 7500
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.74,
|
113 |
+
"learning_rate": 4.767001149460996e-05,
|
114 |
+
"loss": 1.3604,
|
115 |
+
"step": 8000
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.79,
|
119 |
+
"learning_rate": 4.751467892758396e-05,
|
120 |
+
"loss": 1.3653,
|
121 |
+
"step": 8500
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.84,
|
125 |
+
"learning_rate": 4.7359346360557956e-05,
|
126 |
+
"loss": 1.3603,
|
127 |
+
"step": 9000
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 0.88,
|
131 |
+
"learning_rate": 4.7204013793531954e-05,
|
132 |
+
"loss": 1.3188,
|
133 |
+
"step": 9500
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 0.93,
|
137 |
+
"learning_rate": 4.704868122650595e-05,
|
138 |
+
"loss": 1.3281,
|
139 |
+
"step": 10000
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 0.93,
|
143 |
+
"eval_bleu": 14.3522,
|
144 |
+
"eval_gen_len": 14.9186,
|
145 |
+
"eval_loss": 1.2917265892028809,
|
146 |
+
"eval_runtime": 1232.5734,
|
147 |
+
"eval_samples_per_second": 17.528,
|
148 |
+
"eval_steps_per_second": 1.096,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.98,
|
153 |
+
"learning_rate": 4.689334865947995e-05,
|
154 |
+
"loss": 1.3329,
|
155 |
+
"step": 10500
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.02,
|
159 |
+
"learning_rate": 4.673801609245395e-05,
|
160 |
+
"loss": 1.3462,
|
161 |
+
"step": 11000
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.07,
|
165 |
+
"learning_rate": 4.6582683525427945e-05,
|
166 |
+
"loss": 1.3051,
|
167 |
+
"step": 11500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.11,
|
171 |
+
"learning_rate": 4.6427350958401935e-05,
|
172 |
+
"loss": 1.2672,
|
173 |
+
"step": 12000
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.16,
|
177 |
+
"learning_rate": 4.627201839137594e-05,
|
178 |
+
"loss": 1.2606,
|
179 |
+
"step": 12500
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.21,
|
183 |
+
"learning_rate": 4.611668582434994e-05,
|
184 |
+
"loss": 1.268,
|
185 |
+
"step": 13000
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.25,
|
189 |
+
"learning_rate": 4.596135325732393e-05,
|
190 |
+
"loss": 1.2733,
|
191 |
+
"step": 13500
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.3,
|
195 |
+
"learning_rate": 4.5806020690297926e-05,
|
196 |
+
"loss": 1.2652,
|
197 |
+
"step": 14000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.35,
|
201 |
+
"learning_rate": 4.565068812327193e-05,
|
202 |
+
"loss": 1.2389,
|
203 |
+
"step": 14500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.39,
|
207 |
+
"learning_rate": 4.549535555624592e-05,
|
208 |
+
"loss": 1.2506,
|
209 |
+
"step": 15000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.39,
|
213 |
+
"eval_bleu": 14.3525,
|
214 |
+
"eval_gen_len": 14.9494,
|
215 |
+
"eval_loss": 1.2668566703796387,
|
216 |
+
"eval_runtime": 1245.6055,
|
217 |
+
"eval_samples_per_second": 17.345,
|
218 |
+
"eval_steps_per_second": 1.085,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.44,
|
223 |
+
"learning_rate": 4.534002298921992e-05,
|
224 |
+
"loss": 1.2423,
|
225 |
+
"step": 15500
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.49,
|
229 |
+
"learning_rate": 4.5184690422193924e-05,
|
230 |
+
"loss": 1.2214,
|
231 |
+
"step": 16000
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.53,
|
235 |
+
"learning_rate": 4.5029357855167915e-05,
|
236 |
+
"loss": 1.1984,
|
237 |
+
"step": 16500
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.58,
|
241 |
+
"learning_rate": 4.487402528814191e-05,
|
242 |
+
"loss": 1.1922,
|
243 |
+
"step": 17000
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.63,
|
247 |
+
"learning_rate": 4.471869272111591e-05,
|
248 |
+
"loss": 1.1952,
|
249 |
+
"step": 17500
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.67,
|
253 |
+
"learning_rate": 4.456336015408991e-05,
|
254 |
+
"loss": 1.1998,
|
255 |
+
"step": 18000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.72,
|
259 |
+
"learning_rate": 4.4408027587063905e-05,
|
260 |
+
"loss": 1.1676,
|
261 |
+
"step": 18500
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.77,
|
265 |
+
"learning_rate": 4.42526950200379e-05,
|
266 |
+
"loss": 1.175,
|
267 |
+
"step": 19000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 1.81,
|
271 |
+
"learning_rate": 4.40973624530119e-05,
|
272 |
+
"loss": 1.1674,
|
273 |
+
"step": 19500
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 1.86,
|
277 |
+
"learning_rate": 4.39420298859859e-05,
|
278 |
+
"loss": 1.1603,
|
279 |
+
"step": 20000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.86,
|
283 |
+
"eval_bleu": 15.248,
|
284 |
+
"eval_gen_len": 15.0062,
|
285 |
+
"eval_loss": 1.2283018827438354,
|
286 |
+
"eval_runtime": 1235.7777,
|
287 |
+
"eval_samples_per_second": 17.483,
|
288 |
+
"eval_steps_per_second": 1.093,
|
289 |
+
"step": 20000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 1.9,
|
293 |
+
"learning_rate": 4.3786697318959896e-05,
|
294 |
+
"loss": 1.1944,
|
295 |
+
"step": 20500
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 1.95,
|
299 |
+
"learning_rate": 4.3631364751933894e-05,
|
300 |
+
"loss": 1.1644,
|
301 |
+
"step": 21000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 2.0,
|
305 |
+
"learning_rate": 4.347603218490789e-05,
|
306 |
+
"loss": 1.137,
|
307 |
+
"step": 21500
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 2.04,
|
311 |
+
"learning_rate": 4.332069961788189e-05,
|
312 |
+
"loss": 1.1543,
|
313 |
+
"step": 22000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 2.09,
|
317 |
+
"learning_rate": 4.316536705085589e-05,
|
318 |
+
"loss": 1.1359,
|
319 |
+
"step": 22500
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 2.14,
|
323 |
+
"learning_rate": 4.301003448382988e-05,
|
324 |
+
"loss": 1.1373,
|
325 |
+
"step": 23000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 2.18,
|
329 |
+
"learning_rate": 4.285470191680388e-05,
|
330 |
+
"loss": 1.1229,
|
331 |
+
"step": 23500
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 2.23,
|
335 |
+
"learning_rate": 4.269936934977788e-05,
|
336 |
+
"loss": 1.1076,
|
337 |
+
"step": 24000
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 2.28,
|
341 |
+
"learning_rate": 4.254403678275187e-05,
|
342 |
+
"loss": 1.0849,
|
343 |
+
"step": 24500
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 2.32,
|
347 |
+
"learning_rate": 4.2388704215725875e-05,
|
348 |
+
"loss": 1.0765,
|
349 |
+
"step": 25000
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 2.32,
|
353 |
+
"eval_bleu": 15.4063,
|
354 |
+
"eval_gen_len": 14.7808,
|
355 |
+
"eval_loss": 1.2180595397949219,
|
356 |
+
"eval_runtime": 1212.0146,
|
357 |
+
"eval_samples_per_second": 17.826,
|
358 |
+
"eval_steps_per_second": 1.115,
|
359 |
+
"step": 25000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 2.37,
|
363 |
+
"learning_rate": 4.223337164869987e-05,
|
364 |
+
"loss": 1.107,
|
365 |
+
"step": 25500
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 2.42,
|
369 |
+
"learning_rate": 4.2078039081673864e-05,
|
370 |
+
"loss": 1.1183,
|
371 |
+
"step": 26000
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 2.46,
|
375 |
+
"learning_rate": 4.192270651464786e-05,
|
376 |
+
"loss": 1.1004,
|
377 |
+
"step": 26500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 2.51,
|
381 |
+
"learning_rate": 4.1767373947621866e-05,
|
382 |
+
"loss": 1.0793,
|
383 |
+
"step": 27000
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 2.56,
|
387 |
+
"learning_rate": 4.161204138059586e-05,
|
388 |
+
"loss": 1.0901,
|
389 |
+
"step": 27500
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 2.6,
|
393 |
+
"learning_rate": 4.1456708813569854e-05,
|
394 |
+
"loss": 1.1028,
|
395 |
+
"step": 28000
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.65,
|
399 |
+
"learning_rate": 4.130137624654385e-05,
|
400 |
+
"loss": 1.1398,
|
401 |
+
"step": 28500
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 2.69,
|
405 |
+
"learning_rate": 4.114604367951785e-05,
|
406 |
+
"loss": 1.088,
|
407 |
+
"step": 29000
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 2.74,
|
411 |
+
"learning_rate": 4.099071111249185e-05,
|
412 |
+
"loss": 1.0459,
|
413 |
+
"step": 29500
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 2.79,
|
417 |
+
"learning_rate": 4.0835378545465845e-05,
|
418 |
+
"loss": 1.1019,
|
419 |
+
"step": 30000
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 2.79,
|
423 |
+
"eval_bleu": 14.3608,
|
424 |
+
"eval_gen_len": 14.9014,
|
425 |
+
"eval_loss": 1.2753331661224365,
|
426 |
+
"eval_runtime": 1243.9853,
|
427 |
+
"eval_samples_per_second": 17.368,
|
428 |
+
"eval_steps_per_second": 1.086,
|
429 |
+
"step": 30000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 2.83,
|
433 |
+
"learning_rate": 4.068004597843984e-05,
|
434 |
+
"loss": 1.0986,
|
435 |
+
"step": 30500
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 2.88,
|
439 |
+
"learning_rate": 4.052471341141384e-05,
|
440 |
+
"loss": 1.107,
|
441 |
+
"step": 31000
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 2.93,
|
445 |
+
"learning_rate": 4.036938084438784e-05,
|
446 |
+
"loss": 1.1234,
|
447 |
+
"step": 31500
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 2.97,
|
451 |
+
"learning_rate": 4.021404827736183e-05,
|
452 |
+
"loss": 1.138,
|
453 |
+
"step": 32000
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 3.02,
|
457 |
+
"learning_rate": 4.0058715710335834e-05,
|
458 |
+
"loss": 1.1106,
|
459 |
+
"step": 32500
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 3.07,
|
463 |
+
"learning_rate": 3.990338314330983e-05,
|
464 |
+
"loss": 1.0363,
|
465 |
+
"step": 33000
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 3.11,
|
469 |
+
"learning_rate": 3.974805057628382e-05,
|
470 |
+
"loss": 1.01,
|
471 |
+
"step": 33500
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 3.16,
|
475 |
+
"learning_rate": 3.959271800925782e-05,
|
476 |
+
"loss": 1.0107,
|
477 |
+
"step": 34000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 3.21,
|
481 |
+
"learning_rate": 3.9437385442231824e-05,
|
482 |
+
"loss": 1.0122,
|
483 |
+
"step": 34500
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 3.25,
|
487 |
+
"learning_rate": 3.9282052875205815e-05,
|
488 |
+
"loss": 1.0504,
|
489 |
+
"step": 35000
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 3.25,
|
493 |
+
"eval_bleu": 15.3253,
|
494 |
+
"eval_gen_len": 14.7948,
|
495 |
+
"eval_loss": 1.2334309816360474,
|
496 |
+
"eval_runtime": 1248.5109,
|
497 |
+
"eval_samples_per_second": 17.305,
|
498 |
+
"eval_steps_per_second": 1.082,
|
499 |
+
"step": 35000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 3.3,
|
503 |
+
"learning_rate": 3.912672030817981e-05,
|
504 |
+
"loss": 1.0407,
|
505 |
+
"step": 35500
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 3.34,
|
509 |
+
"learning_rate": 3.897138774115382e-05,
|
510 |
+
"loss": 1.0186,
|
511 |
+
"step": 36000
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 3.39,
|
515 |
+
"learning_rate": 3.881605517412781e-05,
|
516 |
+
"loss": 0.9753,
|
517 |
+
"step": 36500
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 3.44,
|
521 |
+
"learning_rate": 3.8660722607101806e-05,
|
522 |
+
"loss": 0.9714,
|
523 |
+
"step": 37000
|
524 |
+
},
|
525 |
+
{
|
526 |
+
"epoch": 3.48,
|
527 |
+
"learning_rate": 3.8505390040075804e-05,
|
528 |
+
"loss": 0.9734,
|
529 |
+
"step": 37500
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 3.53,
|
533 |
+
"learning_rate": 3.83500574730498e-05,
|
534 |
+
"loss": 0.963,
|
535 |
+
"step": 38000
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 3.58,
|
539 |
+
"learning_rate": 3.81947249060238e-05,
|
540 |
+
"loss": 1.0455,
|
541 |
+
"step": 38500
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 3.62,
|
545 |
+
"learning_rate": 3.80393923389978e-05,
|
546 |
+
"loss": 1.064,
|
547 |
+
"step": 39000
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 3.67,
|
551 |
+
"learning_rate": 3.7884059771971794e-05,
|
552 |
+
"loss": 1.005,
|
553 |
+
"step": 39500
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 3.72,
|
557 |
+
"learning_rate": 3.772872720494579e-05,
|
558 |
+
"loss": 0.9431,
|
559 |
+
"step": 40000
|
560 |
+
},
|
561 |
+
{
|
562 |
+
"epoch": 3.72,
|
563 |
+
"eval_bleu": 15.2534,
|
564 |
+
"eval_gen_len": 14.7293,
|
565 |
+
"eval_loss": 1.2512198686599731,
|
566 |
+
"eval_runtime": 1212.804,
|
567 |
+
"eval_samples_per_second": 17.814,
|
568 |
+
"eval_steps_per_second": 1.114,
|
569 |
+
"step": 40000
|
570 |
+
}
|
571 |
+
],
|
572 |
+
"logging_steps": 500,
|
573 |
+
"max_steps": 161445,
|
574 |
+
"num_train_epochs": 15,
|
575 |
+
"save_steps": 5000,
|
576 |
+
"total_flos": 1.3871710964736328e+18,
|
577 |
+
"trial_name": null,
|
578 |
+
"trial_params": null
|
579 |
+
}
|
checkpoint-40000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae0f38c3140ce223a1dea38a47eeea582e94c01a40ce282a2ae0ea647f40877c
|
3 |
+
size 4728
|
checkpoint-45000/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": 250014,
|
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": "4.34.1",
|
57 |
+
"use_cache": true,
|
58 |
+
"vocab_size": 250054
|
59 |
+
}
|
checkpoint-45000/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": 250014,
|
7 |
+
"forced_eos_token_id": 2,
|
8 |
+
"max_length": 200,
|
9 |
+
"num_beams": 5,
|
10 |
+
"pad_token_id": 1,
|
11 |
+
"transformers_version": "4.34.1"
|
12 |
+
}
|
checkpoint-45000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f4895cd3b2ed949e9346ab53272104105ae514e6c552a94842a06f754edc4e15
|
3 |
+
size 4887473903
|
checkpoint-45000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:37708d8d93f879518585139cca2dacdaa6e64c1867bdeb8397784bffd54717cf
|
3 |
+
size 2444694490
|
checkpoint-45000/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d05c86be9fd47bb9f2537de2153ccfb65cac5fef7acd85c2b6ee1383290cfed9
|
3 |
+
size 15024
|
checkpoint-45000/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e3cfdc512092693523b8806af3cd70ab9bc73eb4709ab1b51e366f9a79b3f10a
|
3 |
+
size 15024
|
checkpoint-45000/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e391c57d288be58a91a5c614c0e31befa1a7da34f51f67842437981c1597432
|
3 |
+
size 15024
|
checkpoint-45000/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4348689859b16fbb5ea9ed7122758f21bf879ac705718128e655045cdf8d59d4
|
3 |
+
size 15024
|
checkpoint-45000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dba8961d19d9a72707910216643a0333b3c9308e3c0849df754067322e9cc294
|
3 |
+
size 1064
|
checkpoint-45000/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-45000/special_tokens_map.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
checkpoint-45000/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:55c9c06dcc44c6b9872f61d601f88b88ac893ad5a3a77c035c1e1f2a69acdc27
|
3 |
+
size 17110043
|
checkpoint-45000/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": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
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": "en_XX",
|
525 |
+
"tgt_lang": "ko_KR",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
checkpoint-45000/trainer_state.json
ADDED
@@ -0,0 +1,649 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.2180595397949219,
|
3 |
+
"best_model_checkpoint": "./enko_mbartLarge_36p_exp1/checkpoint-25000",
|
4 |
+
"epoch": 4.180990430177459,
|
5 |
+
"eval_steps": 5000,
|
6 |
+
"global_step": 45000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.05,
|
13 |
+
"learning_rate": 5e-05,
|
14 |
+
"loss": 2.3583,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.09,
|
19 |
+
"learning_rate": 4.9844667432974e-05,
|
20 |
+
"loss": 1.7529,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.14,
|
25 |
+
"learning_rate": 4.9689334865948e-05,
|
26 |
+
"loss": 1.5715,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.19,
|
31 |
+
"learning_rate": 4.9534002298921995e-05,
|
32 |
+
"loss": 1.535,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.23,
|
37 |
+
"learning_rate": 4.937866973189599e-05,
|
38 |
+
"loss": 1.5009,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.28,
|
43 |
+
"learning_rate": 4.9223337164869984e-05,
|
44 |
+
"loss": 1.4783,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.33,
|
49 |
+
"learning_rate": 4.906800459784399e-05,
|
50 |
+
"loss": 1.4542,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.37,
|
55 |
+
"learning_rate": 4.8912672030817986e-05,
|
56 |
+
"loss": 1.441,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.42,
|
61 |
+
"learning_rate": 4.875733946379198e-05,
|
62 |
+
"loss": 1.425,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.46,
|
67 |
+
"learning_rate": 4.8602006896765975e-05,
|
68 |
+
"loss": 1.4235,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46,
|
73 |
+
"eval_bleu": 12.3168,
|
74 |
+
"eval_gen_len": 14.6634,
|
75 |
+
"eval_loss": 1.3893214464187622,
|
76 |
+
"eval_runtime": 1225.507,
|
77 |
+
"eval_samples_per_second": 17.629,
|
78 |
+
"eval_steps_per_second": 1.102,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.51,
|
83 |
+
"learning_rate": 4.844667432973998e-05,
|
84 |
+
"loss": 1.3969,
|
85 |
+
"step": 5500
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 0.56,
|
89 |
+
"learning_rate": 4.829134176271397e-05,
|
90 |
+
"loss": 1.3901,
|
91 |
+
"step": 6000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.6,
|
95 |
+
"learning_rate": 4.813600919568797e-05,
|
96 |
+
"loss": 1.3983,
|
97 |
+
"step": 6500
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 0.65,
|
101 |
+
"learning_rate": 4.798067662866197e-05,
|
102 |
+
"loss": 1.3744,
|
103 |
+
"step": 7000
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 0.7,
|
107 |
+
"learning_rate": 4.782534406163596e-05,
|
108 |
+
"loss": 1.3712,
|
109 |
+
"step": 7500
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 0.74,
|
113 |
+
"learning_rate": 4.767001149460996e-05,
|
114 |
+
"loss": 1.3604,
|
115 |
+
"step": 8000
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.79,
|
119 |
+
"learning_rate": 4.751467892758396e-05,
|
120 |
+
"loss": 1.3653,
|
121 |
+
"step": 8500
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.84,
|
125 |
+
"learning_rate": 4.7359346360557956e-05,
|
126 |
+
"loss": 1.3603,
|
127 |
+
"step": 9000
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 0.88,
|
131 |
+
"learning_rate": 4.7204013793531954e-05,
|
132 |
+
"loss": 1.3188,
|
133 |
+
"step": 9500
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 0.93,
|
137 |
+
"learning_rate": 4.704868122650595e-05,
|
138 |
+
"loss": 1.3281,
|
139 |
+
"step": 10000
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 0.93,
|
143 |
+
"eval_bleu": 14.3522,
|
144 |
+
"eval_gen_len": 14.9186,
|
145 |
+
"eval_loss": 1.2917265892028809,
|
146 |
+
"eval_runtime": 1232.5734,
|
147 |
+
"eval_samples_per_second": 17.528,
|
148 |
+
"eval_steps_per_second": 1.096,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.98,
|
153 |
+
"learning_rate": 4.689334865947995e-05,
|
154 |
+
"loss": 1.3329,
|
155 |
+
"step": 10500
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.02,
|
159 |
+
"learning_rate": 4.673801609245395e-05,
|
160 |
+
"loss": 1.3462,
|
161 |
+
"step": 11000
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.07,
|
165 |
+
"learning_rate": 4.6582683525427945e-05,
|
166 |
+
"loss": 1.3051,
|
167 |
+
"step": 11500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.11,
|
171 |
+
"learning_rate": 4.6427350958401935e-05,
|
172 |
+
"loss": 1.2672,
|
173 |
+
"step": 12000
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.16,
|
177 |
+
"learning_rate": 4.627201839137594e-05,
|
178 |
+
"loss": 1.2606,
|
179 |
+
"step": 12500
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.21,
|
183 |
+
"learning_rate": 4.611668582434994e-05,
|
184 |
+
"loss": 1.268,
|
185 |
+
"step": 13000
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.25,
|
189 |
+
"learning_rate": 4.596135325732393e-05,
|
190 |
+
"loss": 1.2733,
|
191 |
+
"step": 13500
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.3,
|
195 |
+
"learning_rate": 4.5806020690297926e-05,
|
196 |
+
"loss": 1.2652,
|
197 |
+
"step": 14000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.35,
|
201 |
+
"learning_rate": 4.565068812327193e-05,
|
202 |
+
"loss": 1.2389,
|
203 |
+
"step": 14500
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.39,
|
207 |
+
"learning_rate": 4.549535555624592e-05,
|
208 |
+
"loss": 1.2506,
|
209 |
+
"step": 15000
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.39,
|
213 |
+
"eval_bleu": 14.3525,
|
214 |
+
"eval_gen_len": 14.9494,
|
215 |
+
"eval_loss": 1.2668566703796387,
|
216 |
+
"eval_runtime": 1245.6055,
|
217 |
+
"eval_samples_per_second": 17.345,
|
218 |
+
"eval_steps_per_second": 1.085,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.44,
|
223 |
+
"learning_rate": 4.534002298921992e-05,
|
224 |
+
"loss": 1.2423,
|
225 |
+
"step": 15500
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.49,
|
229 |
+
"learning_rate": 4.5184690422193924e-05,
|
230 |
+
"loss": 1.2214,
|
231 |
+
"step": 16000
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.53,
|
235 |
+
"learning_rate": 4.5029357855167915e-05,
|
236 |
+
"loss": 1.1984,
|
237 |
+
"step": 16500
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 1.58,
|
241 |
+
"learning_rate": 4.487402528814191e-05,
|
242 |
+
"loss": 1.1922,
|
243 |
+
"step": 17000
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.63,
|
247 |
+
"learning_rate": 4.471869272111591e-05,
|
248 |
+
"loss": 1.1952,
|
249 |
+
"step": 17500
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.67,
|
253 |
+
"learning_rate": 4.456336015408991e-05,
|
254 |
+
"loss": 1.1998,
|
255 |
+
"step": 18000
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.72,
|
259 |
+
"learning_rate": 4.4408027587063905e-05,
|
260 |
+
"loss": 1.1676,
|
261 |
+
"step": 18500
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 1.77,
|
265 |
+
"learning_rate": 4.42526950200379e-05,
|
266 |
+
"loss": 1.175,
|
267 |
+
"step": 19000
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 1.81,
|
271 |
+
"learning_rate": 4.40973624530119e-05,
|
272 |
+
"loss": 1.1674,
|
273 |
+
"step": 19500
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 1.86,
|
277 |
+
"learning_rate": 4.39420298859859e-05,
|
278 |
+
"loss": 1.1603,
|
279 |
+
"step": 20000
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.86,
|
283 |
+
"eval_bleu": 15.248,
|
284 |
+
"eval_gen_len": 15.0062,
|
285 |
+
"eval_loss": 1.2283018827438354,
|
286 |
+
"eval_runtime": 1235.7777,
|
287 |
+
"eval_samples_per_second": 17.483,
|
288 |
+
"eval_steps_per_second": 1.093,
|
289 |
+
"step": 20000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 1.9,
|
293 |
+
"learning_rate": 4.3786697318959896e-05,
|
294 |
+
"loss": 1.1944,
|
295 |
+
"step": 20500
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 1.95,
|
299 |
+
"learning_rate": 4.3631364751933894e-05,
|
300 |
+
"loss": 1.1644,
|
301 |
+
"step": 21000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 2.0,
|
305 |
+
"learning_rate": 4.347603218490789e-05,
|
306 |
+
"loss": 1.137,
|
307 |
+
"step": 21500
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 2.04,
|
311 |
+
"learning_rate": 4.332069961788189e-05,
|
312 |
+
"loss": 1.1543,
|
313 |
+
"step": 22000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 2.09,
|
317 |
+
"learning_rate": 4.316536705085589e-05,
|
318 |
+
"loss": 1.1359,
|
319 |
+
"step": 22500
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 2.14,
|
323 |
+
"learning_rate": 4.301003448382988e-05,
|
324 |
+
"loss": 1.1373,
|
325 |
+
"step": 23000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 2.18,
|
329 |
+
"learning_rate": 4.285470191680388e-05,
|
330 |
+
"loss": 1.1229,
|
331 |
+
"step": 23500
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 2.23,
|
335 |
+
"learning_rate": 4.269936934977788e-05,
|
336 |
+
"loss": 1.1076,
|
337 |
+
"step": 24000
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 2.28,
|
341 |
+
"learning_rate": 4.254403678275187e-05,
|
342 |
+
"loss": 1.0849,
|
343 |
+
"step": 24500
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 2.32,
|
347 |
+
"learning_rate": 4.2388704215725875e-05,
|
348 |
+
"loss": 1.0765,
|
349 |
+
"step": 25000
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 2.32,
|
353 |
+
"eval_bleu": 15.4063,
|
354 |
+
"eval_gen_len": 14.7808,
|
355 |
+
"eval_loss": 1.2180595397949219,
|
356 |
+
"eval_runtime": 1212.0146,
|
357 |
+
"eval_samples_per_second": 17.826,
|
358 |
+
"eval_steps_per_second": 1.115,
|
359 |
+
"step": 25000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 2.37,
|
363 |
+
"learning_rate": 4.223337164869987e-05,
|
364 |
+
"loss": 1.107,
|
365 |
+
"step": 25500
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 2.42,
|
369 |
+
"learning_rate": 4.2078039081673864e-05,
|
370 |
+
"loss": 1.1183,
|
371 |
+
"step": 26000
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 2.46,
|
375 |
+
"learning_rate": 4.192270651464786e-05,
|
376 |
+
"loss": 1.1004,
|
377 |
+
"step": 26500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 2.51,
|
381 |
+
"learning_rate": 4.1767373947621866e-05,
|
382 |
+
"loss": 1.0793,
|
383 |
+
"step": 27000
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 2.56,
|
387 |
+
"learning_rate": 4.161204138059586e-05,
|
388 |
+
"loss": 1.0901,
|
389 |
+
"step": 27500
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 2.6,
|
393 |
+
"learning_rate": 4.1456708813569854e-05,
|
394 |
+
"loss": 1.1028,
|
395 |
+
"step": 28000
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 2.65,
|
399 |
+
"learning_rate": 4.130137624654385e-05,
|
400 |
+
"loss": 1.1398,
|
401 |
+
"step": 28500
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 2.69,
|
405 |
+
"learning_rate": 4.114604367951785e-05,
|
406 |
+
"loss": 1.088,
|
407 |
+
"step": 29000
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 2.74,
|
411 |
+
"learning_rate": 4.099071111249185e-05,
|
412 |
+
"loss": 1.0459,
|
413 |
+
"step": 29500
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 2.79,
|
417 |
+
"learning_rate": 4.0835378545465845e-05,
|
418 |
+
"loss": 1.1019,
|
419 |
+
"step": 30000
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 2.79,
|
423 |
+
"eval_bleu": 14.3608,
|
424 |
+
"eval_gen_len": 14.9014,
|
425 |
+
"eval_loss": 1.2753331661224365,
|
426 |
+
"eval_runtime": 1243.9853,
|
427 |
+
"eval_samples_per_second": 17.368,
|
428 |
+
"eval_steps_per_second": 1.086,
|
429 |
+
"step": 30000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 2.83,
|
433 |
+
"learning_rate": 4.068004597843984e-05,
|
434 |
+
"loss": 1.0986,
|
435 |
+
"step": 30500
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 2.88,
|
439 |
+
"learning_rate": 4.052471341141384e-05,
|
440 |
+
"loss": 1.107,
|
441 |
+
"step": 31000
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 2.93,
|
445 |
+
"learning_rate": 4.036938084438784e-05,
|
446 |
+
"loss": 1.1234,
|
447 |
+
"step": 31500
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 2.97,
|
451 |
+
"learning_rate": 4.021404827736183e-05,
|
452 |
+
"loss": 1.138,
|
453 |
+
"step": 32000
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 3.02,
|
457 |
+
"learning_rate": 4.0058715710335834e-05,
|
458 |
+
"loss": 1.1106,
|
459 |
+
"step": 32500
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 3.07,
|
463 |
+
"learning_rate": 3.990338314330983e-05,
|
464 |
+
"loss": 1.0363,
|
465 |
+
"step": 33000
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 3.11,
|
469 |
+
"learning_rate": 3.974805057628382e-05,
|
470 |
+
"loss": 1.01,
|
471 |
+
"step": 33500
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 3.16,
|
475 |
+
"learning_rate": 3.959271800925782e-05,
|
476 |
+
"loss": 1.0107,
|
477 |
+
"step": 34000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 3.21,
|
481 |
+
"learning_rate": 3.9437385442231824e-05,
|
482 |
+
"loss": 1.0122,
|
483 |
+
"step": 34500
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 3.25,
|
487 |
+
"learning_rate": 3.9282052875205815e-05,
|
488 |
+
"loss": 1.0504,
|
489 |
+
"step": 35000
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 3.25,
|
493 |
+
"eval_bleu": 15.3253,
|
494 |
+
"eval_gen_len": 14.7948,
|
495 |
+
"eval_loss": 1.2334309816360474,
|
496 |
+
"eval_runtime": 1248.5109,
|
497 |
+
"eval_samples_per_second": 17.305,
|
498 |
+
"eval_steps_per_second": 1.082,
|
499 |
+
"step": 35000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 3.3,
|
503 |
+
"learning_rate": 3.912672030817981e-05,
|
504 |
+
"loss": 1.0407,
|
505 |
+
"step": 35500
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 3.34,
|
509 |
+
"learning_rate": 3.897138774115382e-05,
|
510 |
+
"loss": 1.0186,
|
511 |
+
"step": 36000
|
512 |
+
},
|
513 |
+
{
|
514 |
+
"epoch": 3.39,
|
515 |
+
"learning_rate": 3.881605517412781e-05,
|
516 |
+
"loss": 0.9753,
|
517 |
+
"step": 36500
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 3.44,
|
521 |
+
"learning_rate": 3.8660722607101806e-05,
|
522 |
+
"loss": 0.9714,
|
523 |
+
"step": 37000
|
524 |
+
},
|
525 |
+
{
|
526 |
+
"epoch": 3.48,
|
527 |
+
"learning_rate": 3.8505390040075804e-05,
|
528 |
+
"loss": 0.9734,
|
529 |
+
"step": 37500
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 3.53,
|
533 |
+
"learning_rate": 3.83500574730498e-05,
|
534 |
+
"loss": 0.963,
|
535 |
+
"step": 38000
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 3.58,
|
539 |
+
"learning_rate": 3.81947249060238e-05,
|
540 |
+
"loss": 1.0455,
|
541 |
+
"step": 38500
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 3.62,
|
545 |
+
"learning_rate": 3.80393923389978e-05,
|
546 |
+
"loss": 1.064,
|
547 |
+
"step": 39000
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 3.67,
|
551 |
+
"learning_rate": 3.7884059771971794e-05,
|
552 |
+
"loss": 1.005,
|
553 |
+
"step": 39500
|
554 |
+
},
|
555 |
+
{
|
556 |
+
"epoch": 3.72,
|
557 |
+
"learning_rate": 3.772872720494579e-05,
|
558 |
+
"loss": 0.9431,
|
559 |
+
"step": 40000
|
560 |
+
},
|
561 |
+
{
|
562 |
+
"epoch": 3.72,
|
563 |
+
"eval_bleu": 15.2534,
|
564 |
+
"eval_gen_len": 14.7293,
|
565 |
+
"eval_loss": 1.2512198686599731,
|
566 |
+
"eval_runtime": 1212.804,
|
567 |
+
"eval_samples_per_second": 17.814,
|
568 |
+
"eval_steps_per_second": 1.114,
|
569 |
+
"step": 40000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 3.76,
|
573 |
+
"learning_rate": 3.757339463791979e-05,
|
574 |
+
"loss": 0.8958,
|
575 |
+
"step": 40500
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 3.81,
|
579 |
+
"learning_rate": 3.741806207089378e-05,
|
580 |
+
"loss": 0.865,
|
581 |
+
"step": 41000
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 3.86,
|
585 |
+
"learning_rate": 3.7262729503867785e-05,
|
586 |
+
"loss": 0.8829,
|
587 |
+
"step": 41500
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 3.9,
|
591 |
+
"learning_rate": 3.710739693684178e-05,
|
592 |
+
"loss": 0.9097,
|
593 |
+
"step": 42000
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 3.95,
|
597 |
+
"learning_rate": 3.6952064369815774e-05,
|
598 |
+
"loss": 0.8159,
|
599 |
+
"step": 42500
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 4.0,
|
603 |
+
"learning_rate": 3.679673180278977e-05,
|
604 |
+
"loss": 0.8056,
|
605 |
+
"step": 43000
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 4.04,
|
609 |
+
"learning_rate": 3.6641399235763776e-05,
|
610 |
+
"loss": 0.7523,
|
611 |
+
"step": 43500
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 4.09,
|
615 |
+
"learning_rate": 3.648606666873777e-05,
|
616 |
+
"loss": 0.7307,
|
617 |
+
"step": 44000
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 4.13,
|
621 |
+
"learning_rate": 3.6330734101711764e-05,
|
622 |
+
"loss": 0.7998,
|
623 |
+
"step": 44500
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 4.18,
|
627 |
+
"learning_rate": 3.617540153468577e-05,
|
628 |
+
"loss": 0.8394,
|
629 |
+
"step": 45000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 4.18,
|
633 |
+
"eval_bleu": 14.9999,
|
634 |
+
"eval_gen_len": 14.7993,
|
635 |
+
"eval_loss": 1.2971030473709106,
|
636 |
+
"eval_runtime": 1213.4816,
|
637 |
+
"eval_samples_per_second": 17.804,
|
638 |
+
"eval_steps_per_second": 1.113,
|
639 |
+
"step": 45000
|
640 |
+
}
|
641 |
+
],
|
642 |
+
"logging_steps": 500,
|
643 |
+
"max_steps": 161445,
|
644 |
+
"num_train_epochs": 15,
|
645 |
+
"save_steps": 5000,
|
646 |
+
"total_flos": 1.5605414779085128e+18,
|
647 |
+
"trial_name": null,
|
648 |
+
"trial_params": null
|
649 |
+
}
|
checkpoint-45000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae0f38c3140ce223a1dea38a47eeea582e94c01a40ce282a2ae0ea647f40877c
|
3 |
+
size 4728
|
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": 250014,
|
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": "4.34.1",
|
57 |
+
"use_cache": true,
|
58 |
+
"vocab_size": 250054
|
59 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.18,
|
3 |
+
"eval_bleu": 15.4063,
|
4 |
+
"eval_gen_len": 14.7808,
|
5 |
+
"eval_loss": 1.2180595397949219,
|
6 |
+
"eval_runtime": 1209.5725,
|
7 |
+
"eval_samples": 21605,
|
8 |
+
"eval_samples_per_second": 17.862,
|
9 |
+
"eval_steps_per_second": 1.117
|
10 |
+
}
|