Hugging Face's logo

yesj1234
/
ko-en_mbartLarge_exp15

Text2Text Generation
Transformers
PyTorch
Korean
English
mbart
Generated from Trainer
Inference Endpoints
Model card Files Community
yesj1234 commited on
Commit
f12647c
•
1 Parent(s): a806311

Upload folder using huggingface_hub

Browse files
Files changed (33) hide show
  1. .gitattributes +2 -0
  2. README.md +88 -0
  3. added_tokens.json +59 -0
  4. all_results.json +22 -0
  5. checkpoint-32000/added_tokens.json +59 -0
  6. checkpoint-32000/config.json +59 -0
  7. checkpoint-32000/generation_config.json +12 -0
  8. checkpoint-32000/optimizer.pt +3 -0
  9. checkpoint-32000/pytorch_model.bin +3 -0
  10. checkpoint-32000/rng_state_0.pth +3 -0
  11. checkpoint-32000/rng_state_1.pth +3 -0
  12. checkpoint-32000/rng_state_2.pth +3 -0
  13. checkpoint-32000/rng_state_3.pth +3 -0
  14. checkpoint-32000/scheduler.pt +3 -0
  15. checkpoint-32000/sentencepiece.bpe.model +3 -0
  16. checkpoint-32000/special_tokens_map.json +63 -0
  17. checkpoint-32000/tokenizer.json +3 -0
  18. checkpoint-32000/tokenizer_config.json +528 -0
  19. checkpoint-32000/trainer_state.json +563 -0
  20. checkpoint-32000/training_args.bin +3 -0
  21. config.json +59 -0
  22. eval_results.json +10 -0
  23. generated_predictions.txt +0 -0
  24. generation_config.json +12 -0
  25. predict_results.json +9 -0
  26. pytorch_model.bin +3 -0
  27. sentencepiece.bpe.model +3 -0
  28. special_tokens_map.json +63 -0
  29. tokenizer.json +3 -0
  30. tokenizer_config.json +528 -0
  31. train_results.json +8 -0
  32. trainer_state.json +708 -0
  33. training_args.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ checkpoint-32000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ko
4
+ - en
5
+ base_model: facebook/mbart-large-50-many-to-many-mmt
6
+ tags:
7
+ - generated_from_trainer
8
+ metrics:
9
+ - bleu
10
+ model-index:
11
+ - name: ko-en_mbartLarge_exp10p
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # ko-en_mbartLarge_exp10p
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: 1.1283
23
+ - Bleu: 28.8237
24
+ - Gen Len: 18.5382
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: cosine_with_restarts
54
+ - lr_scheduler_warmup_steps: 1000
55
+ - num_epochs: 40
56
+
57
+ ### Training results
58
+
59
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
60
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
61
+ | 1.4782 | 0.31 | 2000 | 1.4360 | 21.538 | 18.6032 |
62
+ | 1.3618 | 0.62 | 4000 | 1.3226 | 23.8354 | 18.5594 |
63
+ | 1.2983 | 0.93 | 6000 | 1.2637 | 25.0795 | 18.7894 |
64
+ | 1.2065 | 1.24 | 8000 | 1.2371 | 25.7409 | 18.5615 |
65
+ | 1.1926 | 1.55 | 10000 | 1.2116 | 26.0527 | 18.4019 |
66
+ | 1.1734 | 1.86 | 12000 | 1.1907 | 26.9802 | 18.6141 |
67
+ | 1.0677 | 2.17 | 14000 | 1.1802 | 27.1925 | 18.4547 |
68
+ | 1.0773 | 2.48 | 16000 | 1.1655 | 27.5641 | 18.6726 |
69
+ | 1.0688 | 2.78 | 18000 | 1.1521 | 27.6261 | 18.6127 |
70
+ | 0.9542 | 3.09 | 20000 | 1.1709 | 27.16 | 18.3782 |
71
+ | 0.9531 | 3.4 | 22000 | 1.1435 | 28.0684 | 18.436 |
72
+ | 0.9756 | 3.71 | 24000 | 1.1565 | 27.6025 | 18.7284 |
73
+ | 0.9964 | 4.02 | 26000 | 1.2285 | 25.6999 | 18.3255 |
74
+ | 0.9721 | 4.33 | 28000 | 1.1881 | 27.3499 | 18.5409 |
75
+ | 0.9237 | 4.64 | 30000 | 1.1497 | 28.2692 | 18.6614 |
76
+ | 0.9041 | 4.95 | 32000 | 1.1283 | 28.8215 | 18.5493 |
77
+ | 0.6842 | 5.26 | 34000 | 1.1741 | 28.6873 | 18.515 |
78
+ | 0.7101 | 5.57 | 36000 | 1.1876 | 28.0778 | 18.3422 |
79
+ | 0.7697 | 5.88 | 38000 | 1.1898 | 27.6338 | 18.6766 |
80
+ | 0.6028 | 6.19 | 40000 | 1.2393 | 28.0713 | 18.5903 |
81
+
82
+
83
+ ### Framework versions
84
+
85
+ - Transformers 4.34.0
86
+ - Pytorch 2.1.0+cu121
87
+ - Datasets 2.14.5
88
+ - Tokenizers 0.14.1
added_tokens.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 250053,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3,
7
+ "af_ZA": 250026,
8
+ "ar_AR": 250001,
9
+ "az_AZ": 250027,
10
+ "bn_IN": 250028,
11
+ "cs_CZ": 250002,
12
+ "de_DE": 250003,
13
+ "en_XX": 250004,
14
+ "es_XX": 250005,
15
+ "et_EE": 250006,
16
+ "fa_IR": 250029,
17
+ "fi_FI": 250007,
18
+ "fr_XX": 250008,
19
+ "gl_ES": 250051,
20
+ "gu_IN": 250009,
21
+ "he_IL": 250030,
22
+ "hi_IN": 250010,
23
+ "hr_HR": 250031,
24
+ "id_ID": 250032,
25
+ "it_IT": 250011,
26
+ "ja_XX": 250012,
27
+ "ka_GE": 250033,
28
+ "kk_KZ": 250013,
29
+ "km_KH": 250034,
30
+ "ko_KR": 250014,
31
+ "lt_LT": 250015,
32
+ "lv_LV": 250016,
33
+ "mk_MK": 250035,
34
+ "ml_IN": 250036,
35
+ "mn_MN": 250037,
36
+ "mr_IN": 250038,
37
+ "my_MM": 250017,
38
+ "ne_NP": 250018,
39
+ "nl_XX": 250019,
40
+ "pl_PL": 250039,
41
+ "ps_AF": 250040,
42
+ "pt_XX": 250041,
43
+ "ro_RO": 250020,
44
+ "ru_RU": 250021,
45
+ "si_LK": 250022,
46
+ "sl_SI": 250052,
47
+ "sv_SE": 250042,
48
+ "sw_KE": 250043,
49
+ "ta_IN": 250044,
50
+ "te_IN": 250045,
51
+ "th_TH": 250046,
52
+ "tl_XX": 250047,
53
+ "tr_TR": 250023,
54
+ "uk_UA": 250048,
55
+ "ur_PK": 250049,
56
+ "vi_VN": 250024,
57
+ "xh_ZA": 250050,
58
+ "zh_CN": 250025
59
+ }
all_results.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 6.19,
3
+ "eval_bleu": 28.8237,
4
+ "eval_gen_len": 18.5382,
5
+ "eval_loss": 1.1282682418823242,
6
+ "eval_runtime": 859.2797,
7
+ "eval_samples": 12927,
8
+ "eval_samples_per_second": 15.044,
9
+ "eval_steps_per_second": 0.94,
10
+ "predict_bleu": 28.7158,
11
+ "predict_gen_len": 18.6255,
12
+ "predict_loss": 1.1186487674713135,
13
+ "predict_runtime": 864.2967,
14
+ "predict_samples": 12928,
15
+ "predict_samples_per_second": 14.958,
16
+ "predict_steps_per_second": 0.935,
17
+ "train_loss": 1.0430022468566895,
18
+ "train_runtime": 46145.4904,
19
+ "train_samples": 103424,
20
+ "train_samples_per_second": 89.65,
21
+ "train_steps_per_second": 5.603
22
+ }
checkpoint-32000/added_tokens.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 250053,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3,
7
+ "af_ZA": 250026,
8
+ "ar_AR": 250001,
9
+ "az_AZ": 250027,
10
+ "bn_IN": 250028,
11
+ "cs_CZ": 250002,
12
+ "de_DE": 250003,
13
+ "en_XX": 250004,
14
+ "es_XX": 250005,
15
+ "et_EE": 250006,
16
+ "fa_IR": 250029,
17
+ "fi_FI": 250007,
18
+ "fr_XX": 250008,
19
+ "gl_ES": 250051,
20
+ "gu_IN": 250009,
21
+ "he_IL": 250030,
22
+ "hi_IN": 250010,
23
+ "hr_HR": 250031,
24
+ "id_ID": 250032,
25
+ "it_IT": 250011,
26
+ "ja_XX": 250012,
27
+ "ka_GE": 250033,
28
+ "kk_KZ": 250013,
29
+ "km_KH": 250034,
30
+ "ko_KR": 250014,
31
+ "lt_LT": 250015,
32
+ "lv_LV": 250016,
33
+ "mk_MK": 250035,
34
+ "ml_IN": 250036,
35
+ "mn_MN": 250037,
36
+ "mr_IN": 250038,
37
+ "my_MM": 250017,
38
+ "ne_NP": 250018,
39
+ "nl_XX": 250019,
40
+ "pl_PL": 250039,
41
+ "ps_AF": 250040,
42
+ "pt_XX": 250041,
43
+ "ro_RO": 250020,
44
+ "ru_RU": 250021,
45
+ "si_LK": 250022,
46
+ "sl_SI": 250052,
47
+ "sv_SE": 250042,
48
+ "sw_KE": 250043,
49
+ "ta_IN": 250044,
50
+ "te_IN": 250045,
51
+ "th_TH": 250046,
52
+ "tl_XX": 250047,
53
+ "tr_TR": 250023,
54
+ "uk_UA": 250048,
55
+ "ur_PK": 250049,
56
+ "vi_VN": 250024,
57
+ "xh_ZA": 250050,
58
+ "zh_CN": 250025
59
+ }
checkpoint-32000/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
6
+ "add_bias_logits": false,
7
+ "add_final_layer_norm": true,
8
+ "architectures": [
9
+ "MBartForConditionalGeneration"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 0,
13
+ "classif_dropout": 0.0,
14
+ "classifier_dropout": 0.0,
15
+ "d_model": 1024,
16
+ "decoder_attention_heads": 16,
17
+ "decoder_ffn_dim": 4096,
18
+ "decoder_layerdrop": 0.0,
19
+ "decoder_layers": 12,
20
+ "decoder_start_token_id": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_bos_token_id": 250004,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.34.0",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
checkpoint-32000/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "forced_bos_token_id": 250004,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.34.0"
12
+ }
checkpoint-32000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f1955a12213801f41b8a0dc7228d0143b41ab28c0aba949e9de581bdc6666db
3
+ size 4887473903
checkpoint-32000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a251f1e8982b753eb8e35886e66cef76ac0d36cef06f5d88311a4051d28811a7
3
+ size 2444694490
checkpoint-32000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2286eaff45f71c1a241f138671bf980fe9acb8ed9af8585dd5764ba961dadfd7
3
+ size 15024
checkpoint-32000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a63f0e29ae5a800745e3af564ac386cf53c5242fe5dbb0109e59180ffb26a1ff
3
+ size 15024
checkpoint-32000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2baa86e0a57bd6d6986197a6ea1746912b4f89cdbc44da931578732ed288eea
3
+ size 15024
checkpoint-32000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:752fb6a032e902b3d02e333f2e574a22412e492f4547bdcf1a057f280c339a5b
3
+ size 15024
checkpoint-32000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbb66a5c4115774ff65265809a78bf9aa1109f5125b8501601ab5e0ef50a5d2a
3
+ size 1064
checkpoint-32000/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-32000/special_tokens_map.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
55
+ ],
56
+ "bos_token": "<s>",
57
+ "cls_token": "<s>",
58
+ "eos_token": "</s>",
59
+ "mask_token": "<mask>",
60
+ "pad_token": "<pad>",
61
+ "sep_token": "</s>",
62
+ "unk_token": "<unk>"
63
+ }
checkpoint-32000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
3
+ size 17109939
checkpoint-32000/tokenizer_config.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": true,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": true,
46
+ "normalized": false,
47
+ "rstrip": true,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": true,
54
+ "normalized": false,
55
+ "rstrip": true,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": true,
62
+ "normalized": false,
63
+ "rstrip": true,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": true,
70
+ "normalized": false,
71
+ "rstrip": true,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": true,
78
+ "normalized": false,
79
+ "rstrip": true,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": true,
86
+ "normalized": false,
87
+ "rstrip": true,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": true,
94
+ "normalized": false,
95
+ "rstrip": true,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": true,
102
+ "normalized": false,
103
+ "rstrip": true,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": true,
110
+ "normalized": false,
111
+ "rstrip": true,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": true,
118
+ "normalized": false,
119
+ "rstrip": true,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": true,
126
+ "normalized": false,
127
+ "rstrip": true,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": true,
134
+ "normalized": false,
135
+ "rstrip": true,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": true,
142
+ "normalized": false,
143
+ "rstrip": true,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": true,
150
+ "normalized": false,
151
+ "rstrip": true,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": true,
158
+ "normalized": false,
159
+ "rstrip": true,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": true,
166
+ "normalized": false,
167
+ "rstrip": true,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": true,
174
+ "normalized": false,
175
+ "rstrip": true,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": true,
182
+ "normalized": false,
183
+ "rstrip": true,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": true,
190
+ "normalized": false,
191
+ "rstrip": true,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": true,
198
+ "normalized": false,
199
+ "rstrip": true,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": true,
206
+ "normalized": false,
207
+ "rstrip": true,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": true,
214
+ "normalized": false,
215
+ "rstrip": true,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": true,
222
+ "normalized": false,
223
+ "rstrip": true,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": true,
230
+ "normalized": false,
231
+ "rstrip": true,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "af_ZA",
237
+ "lstrip": true,
238
+ "normalized": false,
239
+ "rstrip": true,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "250027": {
244
+ "content": "az_AZ",
245
+ "lstrip": true,
246
+ "normalized": false,
247
+ "rstrip": true,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "250028": {
252
+ "content": "bn_IN",
253
+ "lstrip": true,
254
+ "normalized": false,
255
+ "rstrip": true,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "250029": {
260
+ "content": "fa_IR",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": true,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "250030": {
268
+ "content": "he_IL",
269
+ "lstrip": true,
270
+ "normalized": false,
271
+ "rstrip": true,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "250031": {
276
+ "content": "hr_HR",
277
+ "lstrip": true,
278
+ "normalized": false,
279
+ "rstrip": true,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "250032": {
284
+ "content": "id_ID",
285
+ "lstrip": true,
286
+ "normalized": false,
287
+ "rstrip": true,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "250033": {
292
+ "content": "ka_GE",
293
+ "lstrip": true,
294
+ "normalized": false,
295
+ "rstrip": true,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "250034": {
300
+ "content": "km_KH",
301
+ "lstrip": true,
302
+ "normalized": false,
303
+ "rstrip": true,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "250035": {
308
+ "content": "mk_MK",
309
+ "lstrip": true,
310
+ "normalized": false,
311
+ "rstrip": true,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "250036": {
316
+ "content": "ml_IN",
317
+ "lstrip": true,
318
+ "normalized": false,
319
+ "rstrip": true,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "250037": {
324
+ "content": "mn_MN",
325
+ "lstrip": true,
326
+ "normalized": false,
327
+ "rstrip": true,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "250038": {
332
+ "content": "mr_IN",
333
+ "lstrip": true,
334
+ "normalized": false,
335
+ "rstrip": true,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "250039": {
340
+ "content": "pl_PL",
341
+ "lstrip": true,
342
+ "normalized": false,
343
+ "rstrip": true,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "250040": {
348
+ "content": "ps_AF",
349
+ "lstrip": true,
350
+ "normalized": false,
351
+ "rstrip": true,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "250041": {
356
+ "content": "pt_XX",
357
+ "lstrip": true,
358
+ "normalized": false,
359
+ "rstrip": true,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "250042": {
364
+ "content": "sv_SE",
365
+ "lstrip": true,
366
+ "normalized": false,
367
+ "rstrip": true,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "250043": {
372
+ "content": "sw_KE",
373
+ "lstrip": true,
374
+ "normalized": false,
375
+ "rstrip": true,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "250044": {
380
+ "content": "ta_IN",
381
+ "lstrip": true,
382
+ "normalized": false,
383
+ "rstrip": true,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "250045": {
388
+ "content": "te_IN",
389
+ "lstrip": true,
390
+ "normalized": false,
391
+ "rstrip": true,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "250046": {
396
+ "content": "th_TH",
397
+ "lstrip": true,
398
+ "normalized": false,
399
+ "rstrip": true,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "250047": {
404
+ "content": "tl_XX",
405
+ "lstrip": true,
406
+ "normalized": false,
407
+ "rstrip": true,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "250048": {
412
+ "content": "uk_UA",
413
+ "lstrip": true,
414
+ "normalized": false,
415
+ "rstrip": true,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "250049": {
420
+ "content": "ur_PK",
421
+ "lstrip": true,
422
+ "normalized": false,
423
+ "rstrip": true,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
+ "lstrip": true,
430
+ "normalized": false,
431
+ "rstrip": true,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "250051": {
436
+ "content": "gl_ES",
437
+ "lstrip": true,
438
+ "normalized": false,
439
+ "rstrip": true,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "250052": {
444
+ "content": "sl_SI",
445
+ "lstrip": true,
446
+ "normalized": false,
447
+ "rstrip": true,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ }
459
+ },
460
+ "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
513
+ ],
514
+ "bos_token": "<s>",
515
+ "clean_up_tokenization_spaces": true,
516
+ "cls_token": "<s>",
517
+ "eos_token": "</s>",
518
+ "language_codes": "ML50",
519
+ "mask_token": "<mask>",
520
+ "model_max_length": 1000000000000000019884624838656,
521
+ "pad_token": "<pad>",
522
+ "sep_token": "</s>",
523
+ "sp_model_kwargs": {},
524
+ "src_lang": "ko_KR",
525
+ "tgt_lang": "en_XX",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-32000/trainer_state.json ADDED
@@ -0,0 +1,563 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 28.8215,
3
+ "best_model_checkpoint": "./ko-en_mbartLarge_exp10p/checkpoint-32000",
4
+ "epoch": 4.9504950495049505,
5
+ "eval_steps": 2000,
6
+ "global_step": 32000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.08,
13
+ "learning_rate": 2.5e-05,
14
+ "loss": 1.9221,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.15,
19
+ "learning_rate": 5e-05,
20
+ "loss": 1.7751,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.23,
25
+ "learning_rate": 4.9999535065698766e-05,
26
+ "loss": 1.5628,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.31,
31
+ "learning_rate": 4.9998140280088176e-05,
32
+ "loss": 1.4782,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.31,
37
+ "eval_bleu": 21.538,
38
+ "eval_gen_len": 18.6032,
39
+ "eval_loss": 1.435985803604126,
40
+ "eval_runtime": 861.7869,
41
+ "eval_samples_per_second": 15.0,
42
+ "eval_steps_per_second": 0.938,
43
+ "step": 2000
44
+ },
45
+ {
46
+ "epoch": 0.39,
47
+ "learning_rate": 4.999581569504692e-05,
48
+ "loss": 1.4515,
49
+ "step": 2500
50
+ },
51
+ {
52
+ "epoch": 0.46,
53
+ "learning_rate": 4.999256139703734e-05,
54
+ "loss": 1.4195,
55
+ "step": 3000
56
+ },
57
+ {
58
+ "epoch": 0.54,
59
+ "learning_rate": 4.998837750710223e-05,
60
+ "loss": 1.3946,
61
+ "step": 3500
62
+ },
63
+ {
64
+ "epoch": 0.62,
65
+ "learning_rate": 4.99832641808603e-05,
66
+ "loss": 1.3618,
67
+ "step": 4000
68
+ },
69
+ {
70
+ "epoch": 0.62,
71
+ "eval_bleu": 23.8354,
72
+ "eval_gen_len": 18.5594,
73
+ "eval_loss": 1.3225533962249756,
74
+ "eval_runtime": 848.5716,
75
+ "eval_samples_per_second": 15.234,
76
+ "eval_steps_per_second": 0.952,
77
+ "step": 4000
78
+ },
79
+ {
80
+ "epoch": 0.7,
81
+ "learning_rate": 4.99772216085004e-05,
82
+ "loss": 1.3398,
83
+ "step": 4500
84
+ },
85
+ {
86
+ "epoch": 0.77,
87
+ "learning_rate": 4.9970250014774486e-05,
88
+ "loss": 1.3278,
89
+ "step": 5000
90
+ },
91
+ {
92
+ "epoch": 0.85,
93
+ "learning_rate": 4.996234965898918e-05,
94
+ "loss": 1.3002,
95
+ "step": 5500
96
+ },
97
+ {
98
+ "epoch": 0.93,
99
+ "learning_rate": 4.9953520834996206e-05,
100
+ "loss": 1.2983,
101
+ "step": 6000
102
+ },
103
+ {
104
+ "epoch": 0.93,
105
+ "eval_bleu": 25.0795,
106
+ "eval_gen_len": 18.7894,
107
+ "eval_loss": 1.2636672258377075,
108
+ "eval_runtime": 866.6867,
109
+ "eval_samples_per_second": 14.915,
110
+ "eval_steps_per_second": 0.932,
111
+ "step": 6000
112
+ },
113
+ {
114
+ "epoch": 1.01,
115
+ "learning_rate": 4.994376387118141e-05,
116
+ "loss": 1.2851,
117
+ "step": 6500
118
+ },
119
+ {
120
+ "epoch": 1.08,
121
+ "learning_rate": 4.993307913045257e-05,
122
+ "loss": 1.2246,
123
+ "step": 7000
124
+ },
125
+ {
126
+ "epoch": 1.16,
127
+ "learning_rate": 4.9921467010225866e-05,
128
+ "loss": 1.2111,
129
+ "step": 7500
130
+ },
131
+ {
132
+ "epoch": 1.24,
133
+ "learning_rate": 4.9908927942411154e-05,
134
+ "loss": 1.2065,
135
+ "step": 8000
136
+ },
137
+ {
138
+ "epoch": 1.24,
139
+ "eval_bleu": 25.7409,
140
+ "eval_gen_len": 18.5615,
141
+ "eval_loss": 1.237057089805603,
142
+ "eval_runtime": 857.6049,
143
+ "eval_samples_per_second": 15.073,
144
+ "eval_steps_per_second": 0.942,
145
+ "step": 8000
146
+ },
147
+ {
148
+ "epoch": 1.31,
149
+ "learning_rate": 4.989546239339585e-05,
150
+ "loss": 1.201,
151
+ "step": 8500
152
+ },
153
+ {
154
+ "epoch": 1.39,
155
+ "learning_rate": 4.98810708640276e-05,
156
+ "loss": 1.2098,
157
+ "step": 9000
158
+ },
159
+ {
160
+ "epoch": 1.47,
161
+ "learning_rate": 4.986575388959566e-05,
162
+ "loss": 1.1832,
163
+ "step": 9500
164
+ },
165
+ {
166
+ "epoch": 1.55,
167
+ "learning_rate": 4.984951203981097e-05,
168
+ "loss": 1.1926,
169
+ "step": 10000
170
+ },
171
+ {
172
+ "epoch": 1.55,
173
+ "eval_bleu": 26.0527,
174
+ "eval_gen_len": 18.4019,
175
+ "eval_loss": 1.2115644216537476,
176
+ "eval_runtime": 851.6364,
177
+ "eval_samples_per_second": 15.179,
178
+ "eval_steps_per_second": 0.949,
179
+ "step": 10000
180
+ },
181
+ {
182
+ "epoch": 1.62,
183
+ "learning_rate": 4.983234591878498e-05,
184
+ "loss": 1.1858,
185
+ "step": 10500
186
+ },
187
+ {
188
+ "epoch": 1.7,
189
+ "learning_rate": 4.981425616500717e-05,
190
+ "loss": 1.1939,
191
+ "step": 11000
192
+ },
193
+ {
194
+ "epoch": 1.78,
195
+ "learning_rate": 4.9795243451321304e-05,
196
+ "loss": 1.1693,
197
+ "step": 11500
198
+ },
199
+ {
200
+ "epoch": 1.86,
201
+ "learning_rate": 4.977530848490039e-05,
202
+ "loss": 1.1734,
203
+ "step": 12000
204
+ },
205
+ {
206
+ "epoch": 1.86,
207
+ "eval_bleu": 26.9802,
208
+ "eval_gen_len": 18.6141,
209
+ "eval_loss": 1.190738558769226,
210
+ "eval_runtime": 858.138,
211
+ "eval_samples_per_second": 15.064,
212
+ "eval_steps_per_second": 0.942,
213
+ "step": 12000
214
+ },
215
+ {
216
+ "epoch": 1.93,
217
+ "learning_rate": 4.975445200722042e-05,
218
+ "loss": 1.173,
219
+ "step": 12500
220
+ },
221
+ {
222
+ "epoch": 2.01,
223
+ "learning_rate": 4.973267479403273e-05,
224
+ "loss": 1.1541,
225
+ "step": 13000
226
+ },
227
+ {
228
+ "epoch": 2.09,
229
+ "learning_rate": 4.9709977655335196e-05,
230
+ "loss": 1.0598,
231
+ "step": 13500
232
+ },
233
+ {
234
+ "epoch": 2.17,
235
+ "learning_rate": 4.968636143534208e-05,
236
+ "loss": 1.0677,
237
+ "step": 14000
238
+ },
239
+ {
240
+ "epoch": 2.17,
241
+ "eval_bleu": 27.1925,
242
+ "eval_gen_len": 18.4547,
243
+ "eval_loss": 1.1801778078079224,
244
+ "eval_runtime": 850.3421,
245
+ "eval_samples_per_second": 15.202,
246
+ "eval_steps_per_second": 0.95,
247
+ "step": 14000
248
+ },
249
+ {
250
+ "epoch": 2.24,
251
+ "learning_rate": 4.966182701245266e-05,
252
+ "loss": 1.0789,
253
+ "step": 14500
254
+ },
255
+ {
256
+ "epoch": 2.32,
257
+ "learning_rate": 4.9636375299218484e-05,
258
+ "loss": 1.0706,
259
+ "step": 15000
260
+ },
261
+ {
262
+ "epoch": 2.4,
263
+ "learning_rate": 4.961000724230954e-05,
264
+ "loss": 1.082,
265
+ "step": 15500
266
+ },
267
+ {
268
+ "epoch": 2.48,
269
+ "learning_rate": 4.958272382247895e-05,
270
+ "loss": 1.0773,
271
+ "step": 16000
272
+ },
273
+ {
274
+ "epoch": 2.48,
275
+ "eval_bleu": 27.5641,
276
+ "eval_gen_len": 18.6726,
277
+ "eval_loss": 1.1654949188232422,
278
+ "eval_runtime": 856.0869,
279
+ "eval_samples_per_second": 15.1,
280
+ "eval_steps_per_second": 0.944,
281
+ "step": 16000
282
+ },
283
+ {
284
+ "epoch": 2.55,
285
+ "learning_rate": 4.955452605452653e-05,
286
+ "loss": 1.0749,
287
+ "step": 16500
288
+ },
289
+ {
290
+ "epoch": 2.63,
291
+ "learning_rate": 4.952541498726105e-05,
292
+ "loss": 1.0845,
293
+ "step": 17000
294
+ },
295
+ {
296
+ "epoch": 2.71,
297
+ "learning_rate": 4.949539170346119e-05,
298
+ "loss": 1.0725,
299
+ "step": 17500
300
+ },
301
+ {
302
+ "epoch": 2.78,
303
+ "learning_rate": 4.9464457319835334e-05,
304
+ "loss": 1.0688,
305
+ "step": 18000
306
+ },
307
+ {
308
+ "epoch": 2.78,
309
+ "eval_bleu": 27.6261,
310
+ "eval_gen_len": 18.6127,
311
+ "eval_loss": 1.1520819664001465,
312
+ "eval_runtime": 854.4229,
313
+ "eval_samples_per_second": 15.13,
314
+ "eval_steps_per_second": 0.946,
315
+ "step": 18000
316
+ },
317
+ {
318
+ "epoch": 2.86,
319
+ "learning_rate": 4.9432612986979945e-05,
320
+ "loss": 1.0801,
321
+ "step": 18500
322
+ },
323
+ {
324
+ "epoch": 2.94,
325
+ "learning_rate": 4.939985988933683e-05,
326
+ "loss": 1.0993,
327
+ "step": 19000
328
+ },
329
+ {
330
+ "epoch": 3.02,
331
+ "learning_rate": 4.9366199245149095e-05,
332
+ "loss": 1.0534,
333
+ "step": 19500
334
+ },
335
+ {
336
+ "epoch": 3.09,
337
+ "learning_rate": 4.9331632306415776e-05,
338
+ "loss": 0.9542,
339
+ "step": 20000
340
+ },
341
+ {
342
+ "epoch": 3.09,
343
+ "eval_bleu": 27.16,
344
+ "eval_gen_len": 18.3782,
345
+ "eval_loss": 1.1709098815917969,
346
+ "eval_runtime": 856.056,
347
+ "eval_samples_per_second": 15.101,
348
+ "eval_steps_per_second": 0.944,
349
+ "step": 20000
350
+ },
351
+ {
352
+ "epoch": 3.17,
353
+ "learning_rate": 4.929616035884531e-05,
354
+ "loss": 0.9688,
355
+ "step": 20500
356
+ },
357
+ {
358
+ "epoch": 3.25,
359
+ "learning_rate": 4.925978472180771e-05,
360
+ "loss": 0.9453,
361
+ "step": 21000
362
+ },
363
+ {
364
+ "epoch": 3.33,
365
+ "learning_rate": 4.9222506748285495e-05,
366
+ "loss": 0.9471,
367
+ "step": 21500
368
+ },
369
+ {
370
+ "epoch": 3.4,
371
+ "learning_rate": 4.918432782482335e-05,
372
+ "loss": 0.9531,
373
+ "step": 22000
374
+ },
375
+ {
376
+ "epoch": 3.4,
377
+ "eval_bleu": 28.0684,
378
+ "eval_gen_len": 18.436,
379
+ "eval_loss": 1.1434855461120605,
380
+ "eval_runtime": 852.3756,
381
+ "eval_samples_per_second": 15.166,
382
+ "eval_steps_per_second": 0.948,
383
+ "step": 22000
384
+ },
385
+ {
386
+ "epoch": 3.48,
387
+ "learning_rate": 4.914524937147655e-05,
388
+ "loss": 0.9478,
389
+ "step": 22500
390
+ },
391
+ {
392
+ "epoch": 3.56,
393
+ "learning_rate": 4.910527284175818e-05,
394
+ "loss": 0.9594,
395
+ "step": 23000
396
+ },
397
+ {
398
+ "epoch": 3.64,
399
+ "learning_rate": 4.906439972258503e-05,
400
+ "loss": 0.9483,
401
+ "step": 23500
402
+ },
403
+ {
404
+ "epoch": 3.71,
405
+ "learning_rate": 4.902263153422231e-05,
406
+ "loss": 0.9756,
407
+ "step": 24000
408
+ },
409
+ {
410
+ "epoch": 3.71,
411
+ "eval_bleu": 27.6025,
412
+ "eval_gen_len": 18.7284,
413
+ "eval_loss": 1.1564555168151855,
414
+ "eval_runtime": 860.0664,
415
+ "eval_samples_per_second": 15.03,
416
+ "eval_steps_per_second": 0.939,
417
+ "step": 24000
418
+ },
419
+ {
420
+ "epoch": 3.79,
421
+ "learning_rate": 4.8979969830227086e-05,
422
+ "loss": 0.9912,
423
+ "step": 24500
424
+ },
425
+ {
426
+ "epoch": 3.87,
427
+ "learning_rate": 4.893641619739053e-05,
428
+ "loss": 0.9789,
429
+ "step": 25000
430
+ },
431
+ {
432
+ "epoch": 3.94,
433
+ "learning_rate": 4.8891972255678876e-05,
434
+ "loss": 1.0068,
435
+ "step": 25500
436
+ },
437
+ {
438
+ "epoch": 4.02,
439
+ "learning_rate": 4.8846639658173156e-05,
440
+ "loss": 0.9964,
441
+ "step": 26000
442
+ },
443
+ {
444
+ "epoch": 4.02,
445
+ "eval_bleu": 25.6999,
446
+ "eval_gen_len": 18.3255,
447
+ "eval_loss": 1.2285293340682983,
448
+ "eval_runtime": 852.9084,
449
+ "eval_samples_per_second": 15.156,
450
+ "eval_steps_per_second": 0.947,
451
+ "step": 26000
452
+ },
453
+ {
454
+ "epoch": 4.1,
455
+ "learning_rate": 4.880042009100772e-05,
456
+ "loss": 0.9576,
457
+ "step": 26500
458
+ },
459
+ {
460
+ "epoch": 4.18,
461
+ "learning_rate": 4.8753315273307575e-05,
462
+ "loss": 0.9583,
463
+ "step": 27000
464
+ },
465
+ {
466
+ "epoch": 4.25,
467
+ "learning_rate": 4.8705326957124334e-05,
468
+ "loss": 0.9696,
469
+ "step": 27500
470
+ },
471
+ {
472
+ "epoch": 4.33,
473
+ "learning_rate": 4.865645692737114e-05,
474
+ "loss": 0.9721,
475
+ "step": 28000
476
+ },
477
+ {
478
+ "epoch": 4.33,
479
+ "eval_bleu": 27.3499,
480
+ "eval_gen_len": 18.5409,
481
+ "eval_loss": 1.1880507469177246,
482
+ "eval_runtime": 849.7595,
483
+ "eval_samples_per_second": 15.213,
484
+ "eval_steps_per_second": 0.951,
485
+ "step": 28000
486
+ },
487
+ {
488
+ "epoch": 4.41,
489
+ "learning_rate": 4.860670700175625e-05,
490
+ "loss": 0.9355,
491
+ "step": 28500
492
+ },
493
+ {
494
+ "epoch": 4.49,
495
+ "learning_rate": 4.855607903071542e-05,
496
+ "loss": 0.9314,
497
+ "step": 29000
498
+ },
499
+ {
500
+ "epoch": 4.56,
501
+ "learning_rate": 4.850457489734306e-05,
502
+ "loss": 0.9142,
503
+ "step": 29500
504
+ },
505
+ {
506
+ "epoch": 4.64,
507
+ "learning_rate": 4.845219651732225e-05,
508
+ "loss": 0.9237,
509
+ "step": 30000
510
+ },
511
+ {
512
+ "epoch": 4.64,
513
+ "eval_bleu": 28.2692,
514
+ "eval_gen_len": 18.6614,
515
+ "eval_loss": 1.1497083902359009,
516
+ "eval_runtime": 859.3006,
517
+ "eval_samples_per_second": 15.044,
518
+ "eval_steps_per_second": 0.94,
519
+ "step": 30000
520
+ },
521
+ {
522
+ "epoch": 4.72,
523
+ "learning_rate": 4.839894583885341e-05,
524
+ "loss": 0.9102,
525
+ "step": 30500
526
+ },
527
+ {
528
+ "epoch": 4.8,
529
+ "learning_rate": 4.834482484258193e-05,
530
+ "loss": 0.8956,
531
+ "step": 31000
532
+ },
533
+ {
534
+ "epoch": 4.87,
535
+ "learning_rate": 4.8289835541524394e-05,
536
+ "loss": 0.8902,
537
+ "step": 31500
538
+ },
539
+ {
540
+ "epoch": 4.95,
541
+ "learning_rate": 4.8233979980993785e-05,
542
+ "loss": 0.9041,
543
+ "step": 32000
544
+ },
545
+ {
546
+ "epoch": 4.95,
547
+ "eval_bleu": 28.8215,
548
+ "eval_gen_len": 18.5493,
549
+ "eval_loss": 1.1282682418823242,
550
+ "eval_runtime": 861.0284,
551
+ "eval_samples_per_second": 15.013,
552
+ "eval_steps_per_second": 0.938,
553
+ "step": 32000
554
+ }
555
+ ],
556
+ "logging_steps": 500,
557
+ "max_steps": 258560,
558
+ "num_train_epochs": 40,
559
+ "save_steps": 2000,
560
+ "total_flos": 1.1096744631436575e+18,
561
+ "trial_name": null,
562
+ "trial_params": null
563
+ }
checkpoint-32000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd02550cbfb03e9a7ae7cf1a2ddb8160686adc96bf410e0e0c42f69feadc663e
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": 250004,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.34.0",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 6.19,
3
+ "eval_bleu": 28.8237,
4
+ "eval_gen_len": 18.5382,
5
+ "eval_loss": 1.1282682418823242,
6
+ "eval_runtime": 859.2797,
7
+ "eval_samples": 12927,
8
+ "eval_samples_per_second": 15.044,
9
+ "eval_steps_per_second": 0.94
10
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "forced_bos_token_id": 250004,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.34.0"
12
+ }
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 28.7158,
3
+ "predict_gen_len": 18.6255,
4
+ "predict_loss": 1.1186487674713135,
5
+ "predict_runtime": 864.2967,
6
+ "predict_samples": 12928,
7
+ "predict_samples_per_second": 14.958,
8
+ "predict_steps_per_second": 0.935
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a251f1e8982b753eb8e35886e66cef76ac0d36cef06f5d88311a4051d28811a7
3
+ size 2444694490
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
55
+ ],
56
+ "bos_token": "<s>",
57
+ "cls_token": "<s>",
58
+ "eos_token": "</s>",
59
+ "mask_token": "<mask>",
60
+ "pad_token": "<pad>",
61
+ "sep_token": "</s>",
62
+ "unk_token": "<unk>"
63
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3a71a6430f26d8667787032844b9efe0d820460f1eb2e56a96b5f171813bc1a
3
+ size 17109939
tokenizer_config.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": true,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": true,
46
+ "normalized": false,
47
+ "rstrip": true,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": true,
54
+ "normalized": false,
55
+ "rstrip": true,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": true,
62
+ "normalized": false,
63
+ "rstrip": true,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": true,
70
+ "normalized": false,
71
+ "rstrip": true,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": true,
78
+ "normalized": false,
79
+ "rstrip": true,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": true,
86
+ "normalized": false,
87
+ "rstrip": true,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": true,
94
+ "normalized": false,
95
+ "rstrip": true,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": true,
102
+ "normalized": false,
103
+ "rstrip": true,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": true,
110
+ "normalized": false,
111
+ "rstrip": true,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": true,
118
+ "normalized": false,
119
+ "rstrip": true,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": true,
126
+ "normalized": false,
127
+ "rstrip": true,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": true,
134
+ "normalized": false,
135
+ "rstrip": true,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": true,
142
+ "normalized": false,
143
+ "rstrip": true,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": true,
150
+ "normalized": false,
151
+ "rstrip": true,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": true,
158
+ "normalized": false,
159
+ "rstrip": true,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": true,
166
+ "normalized": false,
167
+ "rstrip": true,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": true,
174
+ "normalized": false,
175
+ "rstrip": true,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": true,
182
+ "normalized": false,
183
+ "rstrip": true,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": true,
190
+ "normalized": false,
191
+ "rstrip": true,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": true,
198
+ "normalized": false,
199
+ "rstrip": true,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": true,
206
+ "normalized": false,
207
+ "rstrip": true,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": true,
214
+ "normalized": false,
215
+ "rstrip": true,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": true,
222
+ "normalized": false,
223
+ "rstrip": true,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": true,
230
+ "normalized": false,
231
+ "rstrip": true,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "af_ZA",
237
+ "lstrip": true,
238
+ "normalized": false,
239
+ "rstrip": true,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "250027": {
244
+ "content": "az_AZ",
245
+ "lstrip": true,
246
+ "normalized": false,
247
+ "rstrip": true,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "250028": {
252
+ "content": "bn_IN",
253
+ "lstrip": true,
254
+ "normalized": false,
255
+ "rstrip": true,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "250029": {
260
+ "content": "fa_IR",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": true,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "250030": {
268
+ "content": "he_IL",
269
+ "lstrip": true,
270
+ "normalized": false,
271
+ "rstrip": true,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "250031": {
276
+ "content": "hr_HR",
277
+ "lstrip": true,
278
+ "normalized": false,
279
+ "rstrip": true,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "250032": {
284
+ "content": "id_ID",
285
+ "lstrip": true,
286
+ "normalized": false,
287
+ "rstrip": true,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "250033": {
292
+ "content": "ka_GE",
293
+ "lstrip": true,
294
+ "normalized": false,
295
+ "rstrip": true,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "250034": {
300
+ "content": "km_KH",
301
+ "lstrip": true,
302
+ "normalized": false,
303
+ "rstrip": true,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "250035": {
308
+ "content": "mk_MK",
309
+ "lstrip": true,
310
+ "normalized": false,
311
+ "rstrip": true,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "250036": {
316
+ "content": "ml_IN",
317
+ "lstrip": true,
318
+ "normalized": false,
319
+ "rstrip": true,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "250037": {
324
+ "content": "mn_MN",
325
+ "lstrip": true,
326
+ "normalized": false,
327
+ "rstrip": true,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "250038": {
332
+ "content": "mr_IN",
333
+ "lstrip": true,
334
+ "normalized": false,
335
+ "rstrip": true,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "250039": {
340
+ "content": "pl_PL",
341
+ "lstrip": true,
342
+ "normalized": false,
343
+ "rstrip": true,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "250040": {
348
+ "content": "ps_AF",
349
+ "lstrip": true,
350
+ "normalized": false,
351
+ "rstrip": true,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "250041": {
356
+ "content": "pt_XX",
357
+ "lstrip": true,
358
+ "normalized": false,
359
+ "rstrip": true,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "250042": {
364
+ "content": "sv_SE",
365
+ "lstrip": true,
366
+ "normalized": false,
367
+ "rstrip": true,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "250043": {
372
+ "content": "sw_KE",
373
+ "lstrip": true,
374
+ "normalized": false,
375
+ "rstrip": true,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "250044": {
380
+ "content": "ta_IN",
381
+ "lstrip": true,
382
+ "normalized": false,
383
+ "rstrip": true,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "250045": {
388
+ "content": "te_IN",
389
+ "lstrip": true,
390
+ "normalized": false,
391
+ "rstrip": true,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "250046": {
396
+ "content": "th_TH",
397
+ "lstrip": true,
398
+ "normalized": false,
399
+ "rstrip": true,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "250047": {
404
+ "content": "tl_XX",
405
+ "lstrip": true,
406
+ "normalized": false,
407
+ "rstrip": true,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "250048": {
412
+ "content": "uk_UA",
413
+ "lstrip": true,
414
+ "normalized": false,
415
+ "rstrip": true,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "250049": {
420
+ "content": "ur_PK",
421
+ "lstrip": true,
422
+ "normalized": false,
423
+ "rstrip": true,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
+ "lstrip": true,
430
+ "normalized": false,
431
+ "rstrip": true,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "250051": {
436
+ "content": "gl_ES",
437
+ "lstrip": true,
438
+ "normalized": false,
439
+ "rstrip": true,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "250052": {
444
+ "content": "sl_SI",
445
+ "lstrip": true,
446
+ "normalized": false,
447
+ "rstrip": true,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ }
459
+ },
460
+ "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
513
+ ],
514
+ "bos_token": "<s>",
515
+ "clean_up_tokenization_spaces": true,
516
+ "cls_token": "<s>",
517
+ "eos_token": "</s>",
518
+ "language_codes": "ML50",
519
+ "mask_token": "<mask>",
520
+ "model_max_length": 1000000000000000019884624838656,
521
+ "pad_token": "<pad>",
522
+ "sep_token": "</s>",
523
+ "sp_model_kwargs": {},
524
+ "src_lang": "ko_KR",
525
+ "tgt_lang": "en_XX",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 6.19,
3
+ "train_loss": 1.0430022468566895,
4
+ "train_runtime": 46145.4904,
5
+ "train_samples": 103424,
6
+ "train_samples_per_second": 89.65,
7
+ "train_steps_per_second": 5.603
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,708 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 28.8215,
3
+ "best_model_checkpoint": "./ko-en_mbartLarge_exp10p/checkpoint-32000",
4
+ "epoch": 6.188118811881188,
5
+ "eval_steps": 2000,
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": 0.08,
13
+ "learning_rate": 2.5e-05,
14
+ "loss": 1.9221,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.15,
19
+ "learning_rate": 5e-05,
20
+ "loss": 1.7751,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.23,
25
+ "learning_rate": 4.9999535065698766e-05,
26
+ "loss": 1.5628,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.31,
31
+ "learning_rate": 4.9998140280088176e-05,
32
+ "loss": 1.4782,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.31,
37
+ "eval_bleu": 21.538,
38
+ "eval_gen_len": 18.6032,
39
+ "eval_loss": 1.435985803604126,
40
+ "eval_runtime": 861.7869,
41
+ "eval_samples_per_second": 15.0,
42
+ "eval_steps_per_second": 0.938,
43
+ "step": 2000
44
+ },
45
+ {
46
+ "epoch": 0.39,
47
+ "learning_rate": 4.999581569504692e-05,
48
+ "loss": 1.4515,
49
+ "step": 2500
50
+ },
51
+ {
52
+ "epoch": 0.46,
53
+ "learning_rate": 4.999256139703734e-05,
54
+ "loss": 1.4195,
55
+ "step": 3000
56
+ },
57
+ {
58
+ "epoch": 0.54,
59
+ "learning_rate": 4.998837750710223e-05,
60
+ "loss": 1.3946,
61
+ "step": 3500
62
+ },
63
+ {
64
+ "epoch": 0.62,
65
+ "learning_rate": 4.99832641808603e-05,
66
+ "loss": 1.3618,
67
+ "step": 4000
68
+ },
69
+ {
70
+ "epoch": 0.62,
71
+ "eval_bleu": 23.8354,
72
+ "eval_gen_len": 18.5594,
73
+ "eval_loss": 1.3225533962249756,
74
+ "eval_runtime": 848.5716,
75
+ "eval_samples_per_second": 15.234,
76
+ "eval_steps_per_second": 0.952,
77
+ "step": 4000
78
+ },
79
+ {
80
+ "epoch": 0.7,
81
+ "learning_rate": 4.99772216085004e-05,
82
+ "loss": 1.3398,
83
+ "step": 4500
84
+ },
85
+ {
86
+ "epoch": 0.77,
87
+ "learning_rate": 4.9970250014774486e-05,
88
+ "loss": 1.3278,
89
+ "step": 5000
90
+ },
91
+ {
92
+ "epoch": 0.85,
93
+ "learning_rate": 4.996234965898918e-05,
94
+ "loss": 1.3002,
95
+ "step": 5500
96
+ },
97
+ {
98
+ "epoch": 0.93,
99
+ "learning_rate": 4.9953520834996206e-05,
100
+ "loss": 1.2983,
101
+ "step": 6000
102
+ },
103
+ {
104
+ "epoch": 0.93,
105
+ "eval_bleu": 25.0795,
106
+ "eval_gen_len": 18.7894,
107
+ "eval_loss": 1.2636672258377075,
108
+ "eval_runtime": 866.6867,
109
+ "eval_samples_per_second": 14.915,
110
+ "eval_steps_per_second": 0.932,
111
+ "step": 6000
112
+ },
113
+ {
114
+ "epoch": 1.01,
115
+ "learning_rate": 4.994376387118141e-05,
116
+ "loss": 1.2851,
117
+ "step": 6500
118
+ },
119
+ {
120
+ "epoch": 1.08,
121
+ "learning_rate": 4.993307913045257e-05,
122
+ "loss": 1.2246,
123
+ "step": 7000
124
+ },
125
+ {
126
+ "epoch": 1.16,
127
+ "learning_rate": 4.9921467010225866e-05,
128
+ "loss": 1.2111,
129
+ "step": 7500
130
+ },
131
+ {
132
+ "epoch": 1.24,
133
+ "learning_rate": 4.9908927942411154e-05,
134
+ "loss": 1.2065,
135
+ "step": 8000
136
+ },
137
+ {
138
+ "epoch": 1.24,
139
+ "eval_bleu": 25.7409,
140
+ "eval_gen_len": 18.5615,
141
+ "eval_loss": 1.237057089805603,
142
+ "eval_runtime": 857.6049,
143
+ "eval_samples_per_second": 15.073,
144
+ "eval_steps_per_second": 0.942,
145
+ "step": 8000
146
+ },
147
+ {
148
+ "epoch": 1.31,
149
+ "learning_rate": 4.989546239339585e-05,
150
+ "loss": 1.201,
151
+ "step": 8500
152
+ },
153
+ {
154
+ "epoch": 1.39,
155
+ "learning_rate": 4.98810708640276e-05,
156
+ "loss": 1.2098,
157
+ "step": 9000
158
+ },
159
+ {
160
+ "epoch": 1.47,
161
+ "learning_rate": 4.986575388959566e-05,
162
+ "loss": 1.1832,
163
+ "step": 9500
164
+ },
165
+ {
166
+ "epoch": 1.55,
167
+ "learning_rate": 4.984951203981097e-05,
168
+ "loss": 1.1926,
169
+ "step": 10000
170
+ },
171
+ {
172
+ "epoch": 1.55,
173
+ "eval_bleu": 26.0527,
174
+ "eval_gen_len": 18.4019,
175
+ "eval_loss": 1.2115644216537476,
176
+ "eval_runtime": 851.6364,
177
+ "eval_samples_per_second": 15.179,
178
+ "eval_steps_per_second": 0.949,
179
+ "step": 10000
180
+ },
181
+ {
182
+ "epoch": 1.62,
183
+ "learning_rate": 4.983234591878498e-05,
184
+ "loss": 1.1858,
185
+ "step": 10500
186
+ },
187
+ {
188
+ "epoch": 1.7,
189
+ "learning_rate": 4.981425616500717e-05,
190
+ "loss": 1.1939,
191
+ "step": 11000
192
+ },
193
+ {
194
+ "epoch": 1.78,
195
+ "learning_rate": 4.9795243451321304e-05,
196
+ "loss": 1.1693,
197
+ "step": 11500
198
+ },
199
+ {
200
+ "epoch": 1.86,
201
+ "learning_rate": 4.977530848490039e-05,
202
+ "loss": 1.1734,
203
+ "step": 12000
204
+ },
205
+ {
206
+ "epoch": 1.86,
207
+ "eval_bleu": 26.9802,
208
+ "eval_gen_len": 18.6141,
209
+ "eval_loss": 1.190738558769226,
210
+ "eval_runtime": 858.138,
211
+ "eval_samples_per_second": 15.064,
212
+ "eval_steps_per_second": 0.942,
213
+ "step": 12000
214
+ },
215
+ {
216
+ "epoch": 1.93,
217
+ "learning_rate": 4.975445200722042e-05,
218
+ "loss": 1.173,
219
+ "step": 12500
220
+ },
221
+ {
222
+ "epoch": 2.01,
223
+ "learning_rate": 4.973267479403273e-05,
224
+ "loss": 1.1541,
225
+ "step": 13000
226
+ },
227
+ {
228
+ "epoch": 2.09,
229
+ "learning_rate": 4.9709977655335196e-05,
230
+ "loss": 1.0598,
231
+ "step": 13500
232
+ },
233
+ {
234
+ "epoch": 2.17,
235
+ "learning_rate": 4.968636143534208e-05,
236
+ "loss": 1.0677,
237
+ "step": 14000
238
+ },
239
+ {
240
+ "epoch": 2.17,
241
+ "eval_bleu": 27.1925,
242
+ "eval_gen_len": 18.4547,
243
+ "eval_loss": 1.1801778078079224,
244
+ "eval_runtime": 850.3421,
245
+ "eval_samples_per_second": 15.202,
246
+ "eval_steps_per_second": 0.95,
247
+ "step": 14000
248
+ },
249
+ {
250
+ "epoch": 2.24,
251
+ "learning_rate": 4.966182701245266e-05,
252
+ "loss": 1.0789,
253
+ "step": 14500
254
+ },
255
+ {
256
+ "epoch": 2.32,
257
+ "learning_rate": 4.9636375299218484e-05,
258
+ "loss": 1.0706,
259
+ "step": 15000
260
+ },
261
+ {
262
+ "epoch": 2.4,
263
+ "learning_rate": 4.961000724230954e-05,
264
+ "loss": 1.082,
265
+ "step": 15500
266
+ },
267
+ {
268
+ "epoch": 2.48,
269
+ "learning_rate": 4.958272382247895e-05,
270
+ "loss": 1.0773,
271
+ "step": 16000
272
+ },
273
+ {
274
+ "epoch": 2.48,
275
+ "eval_bleu": 27.5641,
276
+ "eval_gen_len": 18.6726,
277
+ "eval_loss": 1.1654949188232422,
278
+ "eval_runtime": 856.0869,
279
+ "eval_samples_per_second": 15.1,
280
+ "eval_steps_per_second": 0.944,
281
+ "step": 16000
282
+ },
283
+ {
284
+ "epoch": 2.55,
285
+ "learning_rate": 4.955452605452653e-05,
286
+ "loss": 1.0749,
287
+ "step": 16500
288
+ },
289
+ {
290
+ "epoch": 2.63,
291
+ "learning_rate": 4.952541498726105e-05,
292
+ "loss": 1.0845,
293
+ "step": 17000
294
+ },
295
+ {
296
+ "epoch": 2.71,
297
+ "learning_rate": 4.949539170346119e-05,
298
+ "loss": 1.0725,
299
+ "step": 17500
300
+ },
301
+ {
302
+ "epoch": 2.78,
303
+ "learning_rate": 4.9464457319835334e-05,
304
+ "loss": 1.0688,
305
+ "step": 18000
306
+ },
307
+ {
308
+ "epoch": 2.78,
309
+ "eval_bleu": 27.6261,
310
+ "eval_gen_len": 18.6127,
311
+ "eval_loss": 1.1520819664001465,
312
+ "eval_runtime": 854.4229,
313
+ "eval_samples_per_second": 15.13,
314
+ "eval_steps_per_second": 0.946,
315
+ "step": 18000
316
+ },
317
+ {
318
+ "epoch": 2.86,
319
+ "learning_rate": 4.9432612986979945e-05,
320
+ "loss": 1.0801,
321
+ "step": 18500
322
+ },
323
+ {
324
+ "epoch": 2.94,
325
+ "learning_rate": 4.939985988933683e-05,
326
+ "loss": 1.0993,
327
+ "step": 19000
328
+ },
329
+ {
330
+ "epoch": 3.02,
331
+ "learning_rate": 4.9366199245149095e-05,
332
+ "loss": 1.0534,
333
+ "step": 19500
334
+ },
335
+ {
336
+ "epoch": 3.09,
337
+ "learning_rate": 4.9331632306415776e-05,
338
+ "loss": 0.9542,
339
+ "step": 20000
340
+ },
341
+ {
342
+ "epoch": 3.09,
343
+ "eval_bleu": 27.16,
344
+ "eval_gen_len": 18.3782,
345
+ "eval_loss": 1.1709098815917969,
346
+ "eval_runtime": 856.056,
347
+ "eval_samples_per_second": 15.101,
348
+ "eval_steps_per_second": 0.944,
349
+ "step": 20000
350
+ },
351
+ {
352
+ "epoch": 3.17,
353
+ "learning_rate": 4.929616035884531e-05,
354
+ "loss": 0.9688,
355
+ "step": 20500
356
+ },
357
+ {
358
+ "epoch": 3.25,
359
+ "learning_rate": 4.925978472180771e-05,
360
+ "loss": 0.9453,
361
+ "step": 21000
362
+ },
363
+ {
364
+ "epoch": 3.33,
365
+ "learning_rate": 4.9222506748285495e-05,
366
+ "loss": 0.9471,
367
+ "step": 21500
368
+ },
369
+ {
370
+ "epoch": 3.4,
371
+ "learning_rate": 4.918432782482335e-05,
372
+ "loss": 0.9531,
373
+ "step": 22000
374
+ },
375
+ {
376
+ "epoch": 3.4,
377
+ "eval_bleu": 28.0684,
378
+ "eval_gen_len": 18.436,
379
+ "eval_loss": 1.1434855461120605,
380
+ "eval_runtime": 852.3756,
381
+ "eval_samples_per_second": 15.166,
382
+ "eval_steps_per_second": 0.948,
383
+ "step": 22000
384
+ },
385
+ {
386
+ "epoch": 3.48,
387
+ "learning_rate": 4.914524937147655e-05,
388
+ "loss": 0.9478,
389
+ "step": 22500
390
+ },
391
+ {
392
+ "epoch": 3.56,
393
+ "learning_rate": 4.910527284175818e-05,
394
+ "loss": 0.9594,
395
+ "step": 23000
396
+ },
397
+ {
398
+ "epoch": 3.64,
399
+ "learning_rate": 4.906439972258503e-05,
400
+ "loss": 0.9483,
401
+ "step": 23500
402
+ },
403
+ {
404
+ "epoch": 3.71,
405
+ "learning_rate": 4.902263153422231e-05,
406
+ "loss": 0.9756,
407
+ "step": 24000
408
+ },
409
+ {
410
+ "epoch": 3.71,
411
+ "eval_bleu": 27.6025,
412
+ "eval_gen_len": 18.7284,
413
+ "eval_loss": 1.1564555168151855,
414
+ "eval_runtime": 860.0664,
415
+ "eval_samples_per_second": 15.03,
416
+ "eval_steps_per_second": 0.939,
417
+ "step": 24000
418
+ },
419
+ {
420
+ "epoch": 3.79,
421
+ "learning_rate": 4.8979969830227086e-05,
422
+ "loss": 0.9912,
423
+ "step": 24500
424
+ },
425
+ {
426
+ "epoch": 3.87,
427
+ "learning_rate": 4.893641619739053e-05,
428
+ "loss": 0.9789,
429
+ "step": 25000
430
+ },
431
+ {
432
+ "epoch": 3.94,
433
+ "learning_rate": 4.8891972255678876e-05,
434
+ "loss": 1.0068,
435
+ "step": 25500
436
+ },
437
+ {
438
+ "epoch": 4.02,
439
+ "learning_rate": 4.8846639658173156e-05,
440
+ "loss": 0.9964,
441
+ "step": 26000
442
+ },
443
+ {
444
+ "epoch": 4.02,
445
+ "eval_bleu": 25.6999,
446
+ "eval_gen_len": 18.3255,
447
+ "eval_loss": 1.2285293340682983,
448
+ "eval_runtime": 852.9084,
449
+ "eval_samples_per_second": 15.156,
450
+ "eval_steps_per_second": 0.947,
451
+ "step": 26000
452
+ },
453
+ {
454
+ "epoch": 4.1,
455
+ "learning_rate": 4.880042009100772e-05,
456
+ "loss": 0.9576,
457
+ "step": 26500
458
+ },
459
+ {
460
+ "epoch": 4.18,
461
+ "learning_rate": 4.8753315273307575e-05,
462
+ "loss": 0.9583,
463
+ "step": 27000
464
+ },
465
+ {
466
+ "epoch": 4.25,
467
+ "learning_rate": 4.8705326957124334e-05,
468
+ "loss": 0.9696,
469
+ "step": 27500
470
+ },
471
+ {
472
+ "epoch": 4.33,
473
+ "learning_rate": 4.865645692737114e-05,
474
+ "loss": 0.9721,
475
+ "step": 28000
476
+ },
477
+ {
478
+ "epoch": 4.33,
479
+ "eval_bleu": 27.3499,
480
+ "eval_gen_len": 18.5409,
481
+ "eval_loss": 1.1880507469177246,
482
+ "eval_runtime": 849.7595,
483
+ "eval_samples_per_second": 15.213,
484
+ "eval_steps_per_second": 0.951,
485
+ "step": 28000
486
+ },
487
+ {
488
+ "epoch": 4.41,
489
+ "learning_rate": 4.860670700175625e-05,
490
+ "loss": 0.9355,
491
+ "step": 28500
492
+ },
493
+ {
494
+ "epoch": 4.49,
495
+ "learning_rate": 4.855607903071542e-05,
496
+ "loss": 0.9314,
497
+ "step": 29000
498
+ },
499
+ {
500
+ "epoch": 4.56,
501
+ "learning_rate": 4.850457489734306e-05,
502
+ "loss": 0.9142,
503
+ "step": 29500
504
+ },
505
+ {
506
+ "epoch": 4.64,
507
+ "learning_rate": 4.845219651732225e-05,
508
+ "loss": 0.9237,
509
+ "step": 30000
510
+ },
511
+ {
512
+ "epoch": 4.64,
513
+ "eval_bleu": 28.2692,
514
+ "eval_gen_len": 18.6614,
515
+ "eval_loss": 1.1497083902359009,
516
+ "eval_runtime": 859.3006,
517
+ "eval_samples_per_second": 15.044,
518
+ "eval_steps_per_second": 0.94,
519
+ "step": 30000
520
+ },
521
+ {
522
+ "epoch": 4.72,
523
+ "learning_rate": 4.839894583885341e-05,
524
+ "loss": 0.9102,
525
+ "step": 30500
526
+ },
527
+ {
528
+ "epoch": 4.8,
529
+ "learning_rate": 4.834482484258193e-05,
530
+ "loss": 0.8956,
531
+ "step": 31000
532
+ },
533
+ {
534
+ "epoch": 4.87,
535
+ "learning_rate": 4.8289835541524394e-05,
536
+ "loss": 0.8902,
537
+ "step": 31500
538
+ },
539
+ {
540
+ "epoch": 4.95,
541
+ "learning_rate": 4.8233979980993785e-05,
542
+ "loss": 0.9041,
543
+ "step": 32000
544
+ },
545
+ {
546
+ "epoch": 4.95,
547
+ "eval_bleu": 28.8215,
548
+ "eval_gen_len": 18.5493,
549
+ "eval_loss": 1.1282682418823242,
550
+ "eval_runtime": 861.0284,
551
+ "eval_samples_per_second": 15.013,
552
+ "eval_steps_per_second": 0.938,
553
+ "step": 32000
554
+ },
555
+ {
556
+ "epoch": 5.03,
557
+ "learning_rate": 4.817726023852338e-05,
558
+ "loss": 0.8124,
559
+ "step": 32500
560
+ },
561
+ {
562
+ "epoch": 5.11,
563
+ "learning_rate": 4.81196784237895e-05,
564
+ "loss": 0.6719,
565
+ "step": 33000
566
+ },
567
+ {
568
+ "epoch": 5.18,
569
+ "learning_rate": 4.8061236678533e-05,
570
+ "loss": 0.6779,
571
+ "step": 33500
572
+ },
573
+ {
574
+ "epoch": 5.26,
575
+ "learning_rate": 4.800193717647964e-05,
576
+ "loss": 0.6842,
577
+ "step": 34000
578
+ },
579
+ {
580
+ "epoch": 5.26,
581
+ "eval_bleu": 28.6873,
582
+ "eval_gen_len": 18.515,
583
+ "eval_loss": 1.174131989479065,
584
+ "eval_runtime": 856.2355,
585
+ "eval_samples_per_second": 15.097,
586
+ "eval_steps_per_second": 0.944,
587
+ "step": 34000
588
+ },
589
+ {
590
+ "epoch": 5.34,
591
+ "learning_rate": 4.794178212325922e-05,
592
+ "loss": 0.6763,
593
+ "step": 34500
594
+ },
595
+ {
596
+ "epoch": 5.41,
597
+ "learning_rate": 4.7880773756323556e-05,
598
+ "loss": 0.6887,
599
+ "step": 35000
600
+ },
601
+ {
602
+ "epoch": 5.49,
603
+ "learning_rate": 4.781891434486324e-05,
604
+ "loss": 0.7097,
605
+ "step": 35500
606
+ },
607
+ {
608
+ "epoch": 5.57,
609
+ "learning_rate": 4.775620618972326e-05,
610
+ "loss": 0.7101,
611
+ "step": 36000
612
+ },
613
+ {
614
+ "epoch": 5.57,
615
+ "eval_bleu": 28.0778,
616
+ "eval_gen_len": 18.3422,
617
+ "eval_loss": 1.1875863075256348,
618
+ "eval_runtime": 848.3505,
619
+ "eval_samples_per_second": 15.238,
620
+ "eval_steps_per_second": 0.952,
621
+ "step": 36000
622
+ },
623
+ {
624
+ "epoch": 5.65,
625
+ "learning_rate": 4.7692651623317395e-05,
626
+ "loss": 0.7219,
627
+ "step": 36500
628
+ },
629
+ {
630
+ "epoch": 5.72,
631
+ "learning_rate": 4.762825300954147e-05,
632
+ "loss": 0.7489,
633
+ "step": 37000
634
+ },
635
+ {
636
+ "epoch": 5.8,
637
+ "learning_rate": 4.756301274368545e-05,
638
+ "loss": 0.7671,
639
+ "step": 37500
640
+ },
641
+ {
642
+ "epoch": 5.88,
643
+ "learning_rate": 4.749693325234434e-05,
644
+ "loss": 0.7697,
645
+ "step": 38000
646
+ },
647
+ {
648
+ "epoch": 5.88,
649
+ "eval_bleu": 27.6338,
650
+ "eval_gen_len": 18.6766,
651
+ "eval_loss": 1.1897605657577515,
652
+ "eval_runtime": 865.3805,
653
+ "eval_samples_per_second": 14.938,
654
+ "eval_steps_per_second": 0.934,
655
+ "step": 38000
656
+ },
657
+ {
658
+ "epoch": 5.96,
659
+ "learning_rate": 4.74300169933279e-05,
660
+ "loss": 0.7846,
661
+ "step": 38500
662
+ },
663
+ {
664
+ "epoch": 6.03,
665
+ "learning_rate": 4.736226645556926e-05,
666
+ "loss": 0.6989,
667
+ "step": 39000
668
+ },
669
+ {
670
+ "epoch": 6.11,
671
+ "learning_rate": 4.729368415903233e-05,
672
+ "loss": 0.615,
673
+ "step": 39500
674
+ },
675
+ {
676
+ "epoch": 6.19,
677
+ "learning_rate": 4.722427265461809e-05,
678
+ "loss": 0.6028,
679
+ "step": 40000
680
+ },
681
+ {
682
+ "epoch": 6.19,
683
+ "eval_bleu": 28.0713,
684
+ "eval_gen_len": 18.5903,
685
+ "eval_loss": 1.2392680644989014,
686
+ "eval_runtime": 864.5773,
687
+ "eval_samples_per_second": 14.952,
688
+ "eval_steps_per_second": 0.935,
689
+ "step": 40000
690
+ },
691
+ {
692
+ "epoch": 6.19,
693
+ "step": 40000,
694
+ "total_flos": 1.3870670734394655e+18,
695
+ "train_loss": 1.0430022468566895,
696
+ "train_runtime": 46145.4904,
697
+ "train_samples_per_second": 89.65,
698
+ "train_steps_per_second": 5.603
699
+ }
700
+ ],
701
+ "logging_steps": 500,
702
+ "max_steps": 258560,
703
+ "num_train_epochs": 40,
704
+ "save_steps": 2000,
705
+ "total_flos": 1.3870670734394655e+18,
706
+ "trial_name": null,
707
+ "trial_params": null
708
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd02550cbfb03e9a7ae7cf1a2ddb8160686adc96bf410e0e0c42f69feadc663e
3
+ size 4728