Hugging Face's logo

yesj1234
/
koja_mbartLarge_exp1

Text2Text Generation
Transformers
PyTorch
Korean
Japanese
mbart
Generated from Trainer
Inference Endpoints
Model card Files Community
yesj1234 commited on
Commit
65506bc
•
1 Parent(s): fb11025

Upload folder using huggingface_hub

Browse files
Files changed (46) hide show
  1. .gitattributes +3 -0
  2. README.md +78 -0
  3. all_results.json +22 -0
  4. checkpoint-33000/config.json +59 -0
  5. checkpoint-33000/generation_config.json +12 -0
  6. checkpoint-33000/optimizer.pt +3 -0
  7. checkpoint-33000/pytorch_model.bin +3 -0
  8. checkpoint-33000/rng_state_0.pth +3 -0
  9. checkpoint-33000/rng_state_1.pth +3 -0
  10. checkpoint-33000/rng_state_2.pth +3 -0
  11. checkpoint-33000/rng_state_3.pth +3 -0
  12. checkpoint-33000/scheduler.pt +3 -0
  13. checkpoint-33000/sentencepiece.bpe.model +3 -0
  14. checkpoint-33000/special_tokens_map.json +69 -0
  15. checkpoint-33000/tokenizer.json +3 -0
  16. checkpoint-33000/tokenizer_config.json +528 -0
  17. checkpoint-33000/trainer_state.json +475 -0
  18. checkpoint-33000/training_args.bin +3 -0
  19. checkpoint-55000/config.json +59 -0
  20. checkpoint-55000/generation_config.json +12 -0
  21. checkpoint-55000/optimizer.pt +3 -0
  22. checkpoint-55000/pytorch_model.bin +3 -0
  23. checkpoint-55000/rng_state_0.pth +3 -0
  24. checkpoint-55000/rng_state_1.pth +3 -0
  25. checkpoint-55000/rng_state_2.pth +3 -0
  26. checkpoint-55000/rng_state_3.pth +3 -0
  27. checkpoint-55000/scheduler.pt +3 -0
  28. checkpoint-55000/sentencepiece.bpe.model +3 -0
  29. checkpoint-55000/special_tokens_map.json +69 -0
  30. checkpoint-55000/tokenizer.json +3 -0
  31. checkpoint-55000/tokenizer_config.json +528 -0
  32. checkpoint-55000/trainer_state.json +779 -0
  33. checkpoint-55000/training_args.bin +3 -0
  34. config.json +59 -0
  35. eval_results.json +10 -0
  36. generated_predictions.txt +0 -0
  37. generation_config.json +12 -0
  38. predict_results.json +9 -0
  39. pytorch_model.bin +3 -0
  40. sentencepiece.bpe.model +3 -0
  41. special_tokens_map.json +69 -0
  42. tokenizer.json +3 -0
  43. tokenizer_config.json +528 -0
  44. train_results.json +8 -0
  45. trainer_state.json +788 -0
  46. training_args.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,6 @@ 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-33000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-55000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ko
4
+ - ja
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: mbartLarge_koja_37p
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
+ # mbartLarge_koja_37p
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: 0.8995
23
+ - Bleu: 6.2934
24
+ - Gen Len: 17.4862
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: 1500
55
+ - num_epochs: 35
56
+
57
+ ### Training results
58
+
59
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
60
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|
61
+ | 1.2712 | 0.48 | 5500 | 1.2204 | 3.0858 | 18.1085 |
62
+ | 1.0946 | 0.97 | 11000 | 1.0578 | 3.3162 | 17.8651 |
63
+ | 0.9546 | 1.45 | 16500 | 0.9688 | 5.6024 | 17.902 |
64
+ | 0.89 | 1.94 | 22000 | 0.9414 | 5.1453 | 17.6144 |
65
+ | 0.834 | 2.42 | 27500 | 0.9213 | 5.3985 | 17.6899 |
66
+ | 0.7439 | 2.91 | 33000 | 0.8995 | 6.2934 | 17.4862 |
67
+ | 0.6803 | 3.39 | 38500 | 0.9016 | 6.3565 | 17.8899 |
68
+ | 0.733 | 3.88 | 44000 | 0.9226 | 7.0351 | 17.6112 |
69
+ | 0.6601 | 4.36 | 49500 | 0.9807 | 5.3084 | 17.4292 |
70
+ | 0.6933 | 4.84 | 55000 | 0.9238 | 6.8389 | 17.5131 |
71
+
72
+
73
+ ### Framework versions
74
+
75
+ - Transformers 4.34.1
76
+ - Pytorch 2.1.0+cu121
77
+ - Datasets 2.14.6
78
+ - Tokenizers 0.14.1
all_results.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.84,
3
+ "eval_bleu": 6.2934,
4
+ "eval_gen_len": 17.4862,
5
+ "eval_loss": 0.8995155096054077,
6
+ "eval_runtime": 1459.5128,
7
+ "eval_samples": 22707,
8
+ "eval_samples_per_second": 15.558,
9
+ "eval_steps_per_second": 0.973,
10
+ "predict_bleu": 6.6934,
11
+ "predict_gen_len": 17.4855,
12
+ "predict_loss": 0.8966056704521179,
13
+ "predict_runtime": 1467.5245,
14
+ "predict_samples": 22707,
15
+ "predict_samples_per_second": 15.473,
16
+ "predict_steps_per_second": 0.968,
17
+ "train_loss": 0.9352452936345881,
18
+ "train_runtime": 56979.8243,
19
+ "train_samples": 181661,
20
+ "train_samples_per_second": 111.586,
21
+ "train_steps_per_second": 6.974
22
+ }
checkpoint-33000/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": 250012,
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-33000/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": 250012,
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-33000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b16fd660d47caeee798e3aa9340c44e687bc84cc7cefc8b9a9b4865b3e1aa50
3
+ size 4887473903
checkpoint-33000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5c3cd3ec0addbdd6dc608f670bd2e4a00af46ea2a899f929bf197745fa4165c
3
+ size 2444694490
checkpoint-33000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd413a8e79d6fd11adabc06f88c1bf93fd666a53f9036971da0c7ce13ff13af3
3
+ size 15024
checkpoint-33000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbc0c873acb82f57bd5af4a71261b432be2dae7cf5974ac8fb95cf1e835a5732
3
+ size 15024
checkpoint-33000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:127b0da3435adc6c74f1a79d5cfbba0ccb45238dc0dbc22c014b432f21ec50f8
3
+ size 15024
checkpoint-33000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7863fed05d9b23b11b6cdeec5bc2a64668e21341210041e920f4d5c0630205f5
3
+ size 15024
checkpoint-33000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5d403c4a894cc130ade195e13077d2ef94bb4aa0e54c33559a5e46c380e7b7d
3
+ size 1064
checkpoint-33000/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-33000/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-33000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fc2a9a463953c664738ef8571ed97361b8de258b2aca93c0214a89676a73379
3
+ size 17110043
checkpoint-33000/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": "ko_KR",
525
+ "tgt_lang": "ja_XX",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-33000/trainer_state.json ADDED
@@ -0,0 +1,475 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8995155096054077,
3
+ "best_model_checkpoint": "./mbartLarge_koja_37p/checkpoint-33000",
4
+ "epoch": 2.9064646820503786,
5
+ "eval_steps": 5500,
6
+ "global_step": 33000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 1.6666666666666667e-05,
14
+ "loss": 3.5939,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.09,
19
+ "learning_rate": 3.3333333333333335e-05,
20
+ "loss": 3.0254,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.13,
25
+ "learning_rate": 5e-05,
26
+ "loss": 1.9133,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.18,
31
+ "learning_rate": 4.9936851145520225e-05,
32
+ "loss": 1.672,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.22,
37
+ "learning_rate": 4.987370229104044e-05,
38
+ "loss": 1.56,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.26,
43
+ "learning_rate": 4.9810553436560664e-05,
44
+ "loss": 1.4874,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.31,
49
+ "learning_rate": 4.974740458208088e-05,
50
+ "loss": 1.4345,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.35,
55
+ "learning_rate": 4.96842557276011e-05,
56
+ "loss": 1.3676,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.4,
61
+ "learning_rate": 4.9621106873121326e-05,
62
+ "loss": 1.3183,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.44,
67
+ "learning_rate": 4.955795801864154e-05,
68
+ "loss": 1.2987,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.48,
73
+ "learning_rate": 4.9494809164161765e-05,
74
+ "loss": 1.2712,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.48,
79
+ "eval_bleu": 3.0858,
80
+ "eval_gen_len": 18.1085,
81
+ "eval_loss": 1.2203737497329712,
82
+ "eval_runtime": 1528.8304,
83
+ "eval_samples_per_second": 14.853,
84
+ "eval_steps_per_second": 0.929,
85
+ "step": 5500
86
+ },
87
+ {
88
+ "epoch": 0.53,
89
+ "learning_rate": 4.943166030968199e-05,
90
+ "loss": 1.2405,
91
+ "step": 6000
92
+ },
93
+ {
94
+ "epoch": 0.57,
95
+ "learning_rate": 4.9368511455202204e-05,
96
+ "loss": 1.2262,
97
+ "step": 6500
98
+ },
99
+ {
100
+ "epoch": 0.62,
101
+ "learning_rate": 4.930536260072243e-05,
102
+ "loss": 1.1959,
103
+ "step": 7000
104
+ },
105
+ {
106
+ "epoch": 0.66,
107
+ "learning_rate": 4.924221374624264e-05,
108
+ "loss": 1.182,
109
+ "step": 7500
110
+ },
111
+ {
112
+ "epoch": 0.7,
113
+ "learning_rate": 4.9179064891762866e-05,
114
+ "loss": 1.1683,
115
+ "step": 8000
116
+ },
117
+ {
118
+ "epoch": 0.75,
119
+ "learning_rate": 4.911591603728309e-05,
120
+ "loss": 1.1489,
121
+ "step": 8500
122
+ },
123
+ {
124
+ "epoch": 0.79,
125
+ "learning_rate": 4.9052767182803305e-05,
126
+ "loss": 1.1365,
127
+ "step": 9000
128
+ },
129
+ {
130
+ "epoch": 0.84,
131
+ "learning_rate": 4.898961832832353e-05,
132
+ "loss": 1.1313,
133
+ "step": 9500
134
+ },
135
+ {
136
+ "epoch": 0.88,
137
+ "learning_rate": 4.8926469473843744e-05,
138
+ "loss": 1.1172,
139
+ "step": 10000
140
+ },
141
+ {
142
+ "epoch": 0.92,
143
+ "learning_rate": 4.886332061936397e-05,
144
+ "loss": 1.1123,
145
+ "step": 10500
146
+ },
147
+ {
148
+ "epoch": 0.97,
149
+ "learning_rate": 4.880017176488419e-05,
150
+ "loss": 1.0946,
151
+ "step": 11000
152
+ },
153
+ {
154
+ "epoch": 0.97,
155
+ "eval_bleu": 3.3162,
156
+ "eval_gen_len": 17.8651,
157
+ "eval_loss": 1.0577712059020996,
158
+ "eval_runtime": 1538.4057,
159
+ "eval_samples_per_second": 14.76,
160
+ "eval_steps_per_second": 0.923,
161
+ "step": 11000
162
+ },
163
+ {
164
+ "epoch": 1.01,
165
+ "learning_rate": 4.8737022910404406e-05,
166
+ "loss": 1.0856,
167
+ "step": 11500
168
+ },
169
+ {
170
+ "epoch": 1.06,
171
+ "learning_rate": 4.867387405592463e-05,
172
+ "loss": 1.0716,
173
+ "step": 12000
174
+ },
175
+ {
176
+ "epoch": 1.1,
177
+ "learning_rate": 4.8610725201444845e-05,
178
+ "loss": 1.0736,
179
+ "step": 12500
180
+ },
181
+ {
182
+ "epoch": 1.14,
183
+ "learning_rate": 4.854757634696507e-05,
184
+ "loss": 1.0367,
185
+ "step": 13000
186
+ },
187
+ {
188
+ "epoch": 1.19,
189
+ "learning_rate": 4.848442749248529e-05,
190
+ "loss": 1.0204,
191
+ "step": 13500
192
+ },
193
+ {
194
+ "epoch": 1.23,
195
+ "learning_rate": 4.842127863800551e-05,
196
+ "loss": 1.0533,
197
+ "step": 14000
198
+ },
199
+ {
200
+ "epoch": 1.28,
201
+ "learning_rate": 4.835812978352573e-05,
202
+ "loss": 1.0426,
203
+ "step": 14500
204
+ },
205
+ {
206
+ "epoch": 1.32,
207
+ "learning_rate": 4.829498092904595e-05,
208
+ "loss": 1.0261,
209
+ "step": 15000
210
+ },
211
+ {
212
+ "epoch": 1.37,
213
+ "learning_rate": 4.823183207456617e-05,
214
+ "loss": 1.0111,
215
+ "step": 15500
216
+ },
217
+ {
218
+ "epoch": 1.41,
219
+ "learning_rate": 4.816868322008639e-05,
220
+ "loss": 0.9862,
221
+ "step": 16000
222
+ },
223
+ {
224
+ "epoch": 1.45,
225
+ "learning_rate": 4.810553436560661e-05,
226
+ "loss": 0.9546,
227
+ "step": 16500
228
+ },
229
+ {
230
+ "epoch": 1.45,
231
+ "eval_bleu": 5.6024,
232
+ "eval_gen_len": 17.902,
233
+ "eval_loss": 0.9687988758087158,
234
+ "eval_runtime": 1530.7999,
235
+ "eval_samples_per_second": 14.833,
236
+ "eval_steps_per_second": 0.928,
237
+ "step": 16500
238
+ },
239
+ {
240
+ "epoch": 1.5,
241
+ "learning_rate": 4.804238551112683e-05,
242
+ "loss": 0.9642,
243
+ "step": 17000
244
+ },
245
+ {
246
+ "epoch": 1.54,
247
+ "learning_rate": 4.7979236656647054e-05,
248
+ "loss": 0.9364,
249
+ "step": 17500
250
+ },
251
+ {
252
+ "epoch": 1.59,
253
+ "learning_rate": 4.791608780216727e-05,
254
+ "loss": 0.931,
255
+ "step": 18000
256
+ },
257
+ {
258
+ "epoch": 1.63,
259
+ "learning_rate": 4.785293894768749e-05,
260
+ "loss": 0.915,
261
+ "step": 18500
262
+ },
263
+ {
264
+ "epoch": 1.67,
265
+ "learning_rate": 4.778979009320771e-05,
266
+ "loss": 0.9043,
267
+ "step": 19000
268
+ },
269
+ {
270
+ "epoch": 1.72,
271
+ "learning_rate": 4.772664123872793e-05,
272
+ "loss": 0.9158,
273
+ "step": 19500
274
+ },
275
+ {
276
+ "epoch": 1.76,
277
+ "learning_rate": 4.7663492384248155e-05,
278
+ "loss": 0.9068,
279
+ "step": 20000
280
+ },
281
+ {
282
+ "epoch": 1.81,
283
+ "learning_rate": 4.760034352976837e-05,
284
+ "loss": 0.9314,
285
+ "step": 20500
286
+ },
287
+ {
288
+ "epoch": 1.85,
289
+ "learning_rate": 4.7537194675288594e-05,
290
+ "loss": 0.9112,
291
+ "step": 21000
292
+ },
293
+ {
294
+ "epoch": 1.89,
295
+ "learning_rate": 4.747404582080881e-05,
296
+ "loss": 0.8862,
297
+ "step": 21500
298
+ },
299
+ {
300
+ "epoch": 1.94,
301
+ "learning_rate": 4.741089696632903e-05,
302
+ "loss": 0.89,
303
+ "step": 22000
304
+ },
305
+ {
306
+ "epoch": 1.94,
307
+ "eval_bleu": 5.1453,
308
+ "eval_gen_len": 17.6144,
309
+ "eval_loss": 0.941440761089325,
310
+ "eval_runtime": 1490.4507,
311
+ "eval_samples_per_second": 15.235,
312
+ "eval_steps_per_second": 0.953,
313
+ "step": 22000
314
+ },
315
+ {
316
+ "epoch": 1.98,
317
+ "learning_rate": 4.7347748111849256e-05,
318
+ "loss": 0.8726,
319
+ "step": 22500
320
+ },
321
+ {
322
+ "epoch": 2.03,
323
+ "learning_rate": 4.728459925736947e-05,
324
+ "loss": 0.8749,
325
+ "step": 23000
326
+ },
327
+ {
328
+ "epoch": 2.07,
329
+ "learning_rate": 4.7221450402889695e-05,
330
+ "loss": 0.858,
331
+ "step": 23500
332
+ },
333
+ {
334
+ "epoch": 2.11,
335
+ "learning_rate": 4.715830154840992e-05,
336
+ "loss": 0.8479,
337
+ "step": 24000
338
+ },
339
+ {
340
+ "epoch": 2.16,
341
+ "learning_rate": 4.7095152693930134e-05,
342
+ "loss": 0.8394,
343
+ "step": 24500
344
+ },
345
+ {
346
+ "epoch": 2.2,
347
+ "learning_rate": 4.703200383945036e-05,
348
+ "loss": 0.8472,
349
+ "step": 25000
350
+ },
351
+ {
352
+ "epoch": 2.25,
353
+ "learning_rate": 4.696885498497057e-05,
354
+ "loss": 0.7793,
355
+ "step": 25500
356
+ },
357
+ {
358
+ "epoch": 2.29,
359
+ "learning_rate": 4.6905706130490796e-05,
360
+ "loss": 0.7753,
361
+ "step": 26000
362
+ },
363
+ {
364
+ "epoch": 2.33,
365
+ "learning_rate": 4.684255727601102e-05,
366
+ "loss": 0.7655,
367
+ "step": 26500
368
+ },
369
+ {
370
+ "epoch": 2.38,
371
+ "learning_rate": 4.6779408421531235e-05,
372
+ "loss": 0.7666,
373
+ "step": 27000
374
+ },
375
+ {
376
+ "epoch": 2.42,
377
+ "learning_rate": 4.671625956705146e-05,
378
+ "loss": 0.834,
379
+ "step": 27500
380
+ },
381
+ {
382
+ "epoch": 2.42,
383
+ "eval_bleu": 5.3985,
384
+ "eval_gen_len": 17.6899,
385
+ "eval_loss": 0.9212985038757324,
386
+ "eval_runtime": 1489.7985,
387
+ "eval_samples_per_second": 15.242,
388
+ "eval_steps_per_second": 0.953,
389
+ "step": 27500
390
+ },
391
+ {
392
+ "epoch": 2.47,
393
+ "learning_rate": 4.6653110712571674e-05,
394
+ "loss": 0.8098,
395
+ "step": 28000
396
+ },
397
+ {
398
+ "epoch": 2.51,
399
+ "learning_rate": 4.65899618580919e-05,
400
+ "loss": 0.8153,
401
+ "step": 28500
402
+ },
403
+ {
404
+ "epoch": 2.55,
405
+ "learning_rate": 4.652681300361212e-05,
406
+ "loss": 0.7837,
407
+ "step": 29000
408
+ },
409
+ {
410
+ "epoch": 2.6,
411
+ "learning_rate": 4.6463664149132336e-05,
412
+ "loss": 0.8255,
413
+ "step": 29500
414
+ },
415
+ {
416
+ "epoch": 2.64,
417
+ "learning_rate": 4.640051529465256e-05,
418
+ "loss": 0.8345,
419
+ "step": 30000
420
+ },
421
+ {
422
+ "epoch": 2.69,
423
+ "learning_rate": 4.6337366440172775e-05,
424
+ "loss": 0.8061,
425
+ "step": 30500
426
+ },
427
+ {
428
+ "epoch": 2.73,
429
+ "learning_rate": 4.6274217585693e-05,
430
+ "loss": 0.7743,
431
+ "step": 31000
432
+ },
433
+ {
434
+ "epoch": 2.77,
435
+ "learning_rate": 4.621106873121322e-05,
436
+ "loss": 0.7759,
437
+ "step": 31500
438
+ },
439
+ {
440
+ "epoch": 2.82,
441
+ "learning_rate": 4.614791987673344e-05,
442
+ "loss": 0.7777,
443
+ "step": 32000
444
+ },
445
+ {
446
+ "epoch": 2.86,
447
+ "learning_rate": 4.608477102225366e-05,
448
+ "loss": 0.7537,
449
+ "step": 32500
450
+ },
451
+ {
452
+ "epoch": 2.91,
453
+ "learning_rate": 4.602162216777388e-05,
454
+ "loss": 0.7439,
455
+ "step": 33000
456
+ },
457
+ {
458
+ "epoch": 2.91,
459
+ "eval_bleu": 6.2934,
460
+ "eval_gen_len": 17.4862,
461
+ "eval_loss": 0.8995155096054077,
462
+ "eval_runtime": 1459.4956,
463
+ "eval_samples_per_second": 15.558,
464
+ "eval_steps_per_second": 0.973,
465
+ "step": 33000
466
+ }
467
+ ],
468
+ "logging_steps": 500,
469
+ "max_steps": 397390,
470
+ "num_train_epochs": 35,
471
+ "save_steps": 5500,
472
+ "total_flos": 1.1443832137753559e+18,
473
+ "trial_name": null,
474
+ "trial_params": null
475
+ }
checkpoint-33000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2dd88a9d2ace72595383bc2566ed0c69120d645a38262a75472796a55099c32
3
+ size 4664
checkpoint-55000/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": 250012,
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-55000/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": 250012,
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-55000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9474ef99c8259a111be6f441ca7237bc1b9777bd8ee9fbfac1f48021268d91d2
3
+ size 4887473903
checkpoint-55000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08f94589cf7a67b2c1fb606e19bd3093f2338ce9e77ed5ddc0e75cb217ce7463
3
+ size 2444694490
checkpoint-55000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5b73927de71fb1f753954b1bf5c26f405e25fce8d9d14dc1b753ac2a1ff95e0
3
+ size 15024
checkpoint-55000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c9696cccc6c7d3aae29278a13419eb992c3e4eb97d92cce737f67b41f6ccffb
3
+ size 15024
checkpoint-55000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d32edf43347d0c6438b0f4eaf11239fac149ed6793a5e44f0ef7cafb352557eb
3
+ size 15024
checkpoint-55000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f56e8f6475c49f9d7d5ba767176b742c0afb9ac1d19766bf956831e02083a829
3
+ size 15024
checkpoint-55000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d48fb43879bbefcdc352c9a648169ec8945d8d7fbb6a5e4ee9868fba6d794ebf
3
+ size 1064
checkpoint-55000/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-55000/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-55000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fc2a9a463953c664738ef8571ed97361b8de258b2aca93c0214a89676a73379
3
+ size 17110043
checkpoint-55000/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": "ko_KR",
525
+ "tgt_lang": "ja_XX",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-55000/trainer_state.json ADDED
@@ -0,0 +1,779 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8995155096054077,
3
+ "best_model_checkpoint": "./mbartLarge_koja_37p/checkpoint-33000",
4
+ "epoch": 4.844107803417298,
5
+ "eval_steps": 5500,
6
+ "global_step": 55000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 1.6666666666666667e-05,
14
+ "loss": 3.5939,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.09,
19
+ "learning_rate": 3.3333333333333335e-05,
20
+ "loss": 3.0254,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.13,
25
+ "learning_rate": 5e-05,
26
+ "loss": 1.9133,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.18,
31
+ "learning_rate": 4.9936851145520225e-05,
32
+ "loss": 1.672,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.22,
37
+ "learning_rate": 4.987370229104044e-05,
38
+ "loss": 1.56,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.26,
43
+ "learning_rate": 4.9810553436560664e-05,
44
+ "loss": 1.4874,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.31,
49
+ "learning_rate": 4.974740458208088e-05,
50
+ "loss": 1.4345,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.35,
55
+ "learning_rate": 4.96842557276011e-05,
56
+ "loss": 1.3676,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.4,
61
+ "learning_rate": 4.9621106873121326e-05,
62
+ "loss": 1.3183,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.44,
67
+ "learning_rate": 4.955795801864154e-05,
68
+ "loss": 1.2987,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.48,
73
+ "learning_rate": 4.9494809164161765e-05,
74
+ "loss": 1.2712,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.48,
79
+ "eval_bleu": 3.0858,
80
+ "eval_gen_len": 18.1085,
81
+ "eval_loss": 1.2203737497329712,
82
+ "eval_runtime": 1528.8304,
83
+ "eval_samples_per_second": 14.853,
84
+ "eval_steps_per_second": 0.929,
85
+ "step": 5500
86
+ },
87
+ {
88
+ "epoch": 0.53,
89
+ "learning_rate": 4.943166030968199e-05,
90
+ "loss": 1.2405,
91
+ "step": 6000
92
+ },
93
+ {
94
+ "epoch": 0.57,
95
+ "learning_rate": 4.9368511455202204e-05,
96
+ "loss": 1.2262,
97
+ "step": 6500
98
+ },
99
+ {
100
+ "epoch": 0.62,
101
+ "learning_rate": 4.930536260072243e-05,
102
+ "loss": 1.1959,
103
+ "step": 7000
104
+ },
105
+ {
106
+ "epoch": 0.66,
107
+ "learning_rate": 4.924221374624264e-05,
108
+ "loss": 1.182,
109
+ "step": 7500
110
+ },
111
+ {
112
+ "epoch": 0.7,
113
+ "learning_rate": 4.9179064891762866e-05,
114
+ "loss": 1.1683,
115
+ "step": 8000
116
+ },
117
+ {
118
+ "epoch": 0.75,
119
+ "learning_rate": 4.911591603728309e-05,
120
+ "loss": 1.1489,
121
+ "step": 8500
122
+ },
123
+ {
124
+ "epoch": 0.79,
125
+ "learning_rate": 4.9052767182803305e-05,
126
+ "loss": 1.1365,
127
+ "step": 9000
128
+ },
129
+ {
130
+ "epoch": 0.84,
131
+ "learning_rate": 4.898961832832353e-05,
132
+ "loss": 1.1313,
133
+ "step": 9500
134
+ },
135
+ {
136
+ "epoch": 0.88,
137
+ "learning_rate": 4.8926469473843744e-05,
138
+ "loss": 1.1172,
139
+ "step": 10000
140
+ },
141
+ {
142
+ "epoch": 0.92,
143
+ "learning_rate": 4.886332061936397e-05,
144
+ "loss": 1.1123,
145
+ "step": 10500
146
+ },
147
+ {
148
+ "epoch": 0.97,
149
+ "learning_rate": 4.880017176488419e-05,
150
+ "loss": 1.0946,
151
+ "step": 11000
152
+ },
153
+ {
154
+ "epoch": 0.97,
155
+ "eval_bleu": 3.3162,
156
+ "eval_gen_len": 17.8651,
157
+ "eval_loss": 1.0577712059020996,
158
+ "eval_runtime": 1538.4057,
159
+ "eval_samples_per_second": 14.76,
160
+ "eval_steps_per_second": 0.923,
161
+ "step": 11000
162
+ },
163
+ {
164
+ "epoch": 1.01,
165
+ "learning_rate": 4.8737022910404406e-05,
166
+ "loss": 1.0856,
167
+ "step": 11500
168
+ },
169
+ {
170
+ "epoch": 1.06,
171
+ "learning_rate": 4.867387405592463e-05,
172
+ "loss": 1.0716,
173
+ "step": 12000
174
+ },
175
+ {
176
+ "epoch": 1.1,
177
+ "learning_rate": 4.8610725201444845e-05,
178
+ "loss": 1.0736,
179
+ "step": 12500
180
+ },
181
+ {
182
+ "epoch": 1.14,
183
+ "learning_rate": 4.854757634696507e-05,
184
+ "loss": 1.0367,
185
+ "step": 13000
186
+ },
187
+ {
188
+ "epoch": 1.19,
189
+ "learning_rate": 4.848442749248529e-05,
190
+ "loss": 1.0204,
191
+ "step": 13500
192
+ },
193
+ {
194
+ "epoch": 1.23,
195
+ "learning_rate": 4.842127863800551e-05,
196
+ "loss": 1.0533,
197
+ "step": 14000
198
+ },
199
+ {
200
+ "epoch": 1.28,
201
+ "learning_rate": 4.835812978352573e-05,
202
+ "loss": 1.0426,
203
+ "step": 14500
204
+ },
205
+ {
206
+ "epoch": 1.32,
207
+ "learning_rate": 4.829498092904595e-05,
208
+ "loss": 1.0261,
209
+ "step": 15000
210
+ },
211
+ {
212
+ "epoch": 1.37,
213
+ "learning_rate": 4.823183207456617e-05,
214
+ "loss": 1.0111,
215
+ "step": 15500
216
+ },
217
+ {
218
+ "epoch": 1.41,
219
+ "learning_rate": 4.816868322008639e-05,
220
+ "loss": 0.9862,
221
+ "step": 16000
222
+ },
223
+ {
224
+ "epoch": 1.45,
225
+ "learning_rate": 4.810553436560661e-05,
226
+ "loss": 0.9546,
227
+ "step": 16500
228
+ },
229
+ {
230
+ "epoch": 1.45,
231
+ "eval_bleu": 5.6024,
232
+ "eval_gen_len": 17.902,
233
+ "eval_loss": 0.9687988758087158,
234
+ "eval_runtime": 1530.7999,
235
+ "eval_samples_per_second": 14.833,
236
+ "eval_steps_per_second": 0.928,
237
+ "step": 16500
238
+ },
239
+ {
240
+ "epoch": 1.5,
241
+ "learning_rate": 4.804238551112683e-05,
242
+ "loss": 0.9642,
243
+ "step": 17000
244
+ },
245
+ {
246
+ "epoch": 1.54,
247
+ "learning_rate": 4.7979236656647054e-05,
248
+ "loss": 0.9364,
249
+ "step": 17500
250
+ },
251
+ {
252
+ "epoch": 1.59,
253
+ "learning_rate": 4.791608780216727e-05,
254
+ "loss": 0.931,
255
+ "step": 18000
256
+ },
257
+ {
258
+ "epoch": 1.63,
259
+ "learning_rate": 4.785293894768749e-05,
260
+ "loss": 0.915,
261
+ "step": 18500
262
+ },
263
+ {
264
+ "epoch": 1.67,
265
+ "learning_rate": 4.778979009320771e-05,
266
+ "loss": 0.9043,
267
+ "step": 19000
268
+ },
269
+ {
270
+ "epoch": 1.72,
271
+ "learning_rate": 4.772664123872793e-05,
272
+ "loss": 0.9158,
273
+ "step": 19500
274
+ },
275
+ {
276
+ "epoch": 1.76,
277
+ "learning_rate": 4.7663492384248155e-05,
278
+ "loss": 0.9068,
279
+ "step": 20000
280
+ },
281
+ {
282
+ "epoch": 1.81,
283
+ "learning_rate": 4.760034352976837e-05,
284
+ "loss": 0.9314,
285
+ "step": 20500
286
+ },
287
+ {
288
+ "epoch": 1.85,
289
+ "learning_rate": 4.7537194675288594e-05,
290
+ "loss": 0.9112,
291
+ "step": 21000
292
+ },
293
+ {
294
+ "epoch": 1.89,
295
+ "learning_rate": 4.747404582080881e-05,
296
+ "loss": 0.8862,
297
+ "step": 21500
298
+ },
299
+ {
300
+ "epoch": 1.94,
301
+ "learning_rate": 4.741089696632903e-05,
302
+ "loss": 0.89,
303
+ "step": 22000
304
+ },
305
+ {
306
+ "epoch": 1.94,
307
+ "eval_bleu": 5.1453,
308
+ "eval_gen_len": 17.6144,
309
+ "eval_loss": 0.941440761089325,
310
+ "eval_runtime": 1490.4507,
311
+ "eval_samples_per_second": 15.235,
312
+ "eval_steps_per_second": 0.953,
313
+ "step": 22000
314
+ },
315
+ {
316
+ "epoch": 1.98,
317
+ "learning_rate": 4.7347748111849256e-05,
318
+ "loss": 0.8726,
319
+ "step": 22500
320
+ },
321
+ {
322
+ "epoch": 2.03,
323
+ "learning_rate": 4.728459925736947e-05,
324
+ "loss": 0.8749,
325
+ "step": 23000
326
+ },
327
+ {
328
+ "epoch": 2.07,
329
+ "learning_rate": 4.7221450402889695e-05,
330
+ "loss": 0.858,
331
+ "step": 23500
332
+ },
333
+ {
334
+ "epoch": 2.11,
335
+ "learning_rate": 4.715830154840992e-05,
336
+ "loss": 0.8479,
337
+ "step": 24000
338
+ },
339
+ {
340
+ "epoch": 2.16,
341
+ "learning_rate": 4.7095152693930134e-05,
342
+ "loss": 0.8394,
343
+ "step": 24500
344
+ },
345
+ {
346
+ "epoch": 2.2,
347
+ "learning_rate": 4.703200383945036e-05,
348
+ "loss": 0.8472,
349
+ "step": 25000
350
+ },
351
+ {
352
+ "epoch": 2.25,
353
+ "learning_rate": 4.696885498497057e-05,
354
+ "loss": 0.7793,
355
+ "step": 25500
356
+ },
357
+ {
358
+ "epoch": 2.29,
359
+ "learning_rate": 4.6905706130490796e-05,
360
+ "loss": 0.7753,
361
+ "step": 26000
362
+ },
363
+ {
364
+ "epoch": 2.33,
365
+ "learning_rate": 4.684255727601102e-05,
366
+ "loss": 0.7655,
367
+ "step": 26500
368
+ },
369
+ {
370
+ "epoch": 2.38,
371
+ "learning_rate": 4.6779408421531235e-05,
372
+ "loss": 0.7666,
373
+ "step": 27000
374
+ },
375
+ {
376
+ "epoch": 2.42,
377
+ "learning_rate": 4.671625956705146e-05,
378
+ "loss": 0.834,
379
+ "step": 27500
380
+ },
381
+ {
382
+ "epoch": 2.42,
383
+ "eval_bleu": 5.3985,
384
+ "eval_gen_len": 17.6899,
385
+ "eval_loss": 0.9212985038757324,
386
+ "eval_runtime": 1489.7985,
387
+ "eval_samples_per_second": 15.242,
388
+ "eval_steps_per_second": 0.953,
389
+ "step": 27500
390
+ },
391
+ {
392
+ "epoch": 2.47,
393
+ "learning_rate": 4.6653110712571674e-05,
394
+ "loss": 0.8098,
395
+ "step": 28000
396
+ },
397
+ {
398
+ "epoch": 2.51,
399
+ "learning_rate": 4.65899618580919e-05,
400
+ "loss": 0.8153,
401
+ "step": 28500
402
+ },
403
+ {
404
+ "epoch": 2.55,
405
+ "learning_rate": 4.652681300361212e-05,
406
+ "loss": 0.7837,
407
+ "step": 29000
408
+ },
409
+ {
410
+ "epoch": 2.6,
411
+ "learning_rate": 4.6463664149132336e-05,
412
+ "loss": 0.8255,
413
+ "step": 29500
414
+ },
415
+ {
416
+ "epoch": 2.64,
417
+ "learning_rate": 4.640051529465256e-05,
418
+ "loss": 0.8345,
419
+ "step": 30000
420
+ },
421
+ {
422
+ "epoch": 2.69,
423
+ "learning_rate": 4.6337366440172775e-05,
424
+ "loss": 0.8061,
425
+ "step": 30500
426
+ },
427
+ {
428
+ "epoch": 2.73,
429
+ "learning_rate": 4.6274217585693e-05,
430
+ "loss": 0.7743,
431
+ "step": 31000
432
+ },
433
+ {
434
+ "epoch": 2.77,
435
+ "learning_rate": 4.621106873121322e-05,
436
+ "loss": 0.7759,
437
+ "step": 31500
438
+ },
439
+ {
440
+ "epoch": 2.82,
441
+ "learning_rate": 4.614791987673344e-05,
442
+ "loss": 0.7777,
443
+ "step": 32000
444
+ },
445
+ {
446
+ "epoch": 2.86,
447
+ "learning_rate": 4.608477102225366e-05,
448
+ "loss": 0.7537,
449
+ "step": 32500
450
+ },
451
+ {
452
+ "epoch": 2.91,
453
+ "learning_rate": 4.602162216777388e-05,
454
+ "loss": 0.7439,
455
+ "step": 33000
456
+ },
457
+ {
458
+ "epoch": 2.91,
459
+ "eval_bleu": 6.2934,
460
+ "eval_gen_len": 17.4862,
461
+ "eval_loss": 0.8995155096054077,
462
+ "eval_runtime": 1459.4956,
463
+ "eval_samples_per_second": 15.558,
464
+ "eval_steps_per_second": 0.973,
465
+ "step": 33000
466
+ },
467
+ {
468
+ "epoch": 2.95,
469
+ "learning_rate": 4.59584733132941e-05,
470
+ "loss": 0.7424,
471
+ "step": 33500
472
+ },
473
+ {
474
+ "epoch": 2.99,
475
+ "learning_rate": 4.589532445881432e-05,
476
+ "loss": 0.7416,
477
+ "step": 34000
478
+ },
479
+ {
480
+ "epoch": 3.04,
481
+ "learning_rate": 4.583217560433454e-05,
482
+ "loss": 0.7722,
483
+ "step": 34500
484
+ },
485
+ {
486
+ "epoch": 3.08,
487
+ "learning_rate": 4.576902674985476e-05,
488
+ "loss": 0.7967,
489
+ "step": 35000
490
+ },
491
+ {
492
+ "epoch": 3.13,
493
+ "learning_rate": 4.5705877895374984e-05,
494
+ "loss": 0.7323,
495
+ "step": 35500
496
+ },
497
+ {
498
+ "epoch": 3.17,
499
+ "learning_rate": 4.56427290408952e-05,
500
+ "loss": 0.7171,
501
+ "step": 36000
502
+ },
503
+ {
504
+ "epoch": 3.21,
505
+ "learning_rate": 4.557958018641542e-05,
506
+ "loss": 0.7044,
507
+ "step": 36500
508
+ },
509
+ {
510
+ "epoch": 3.26,
511
+ "learning_rate": 4.551643133193564e-05,
512
+ "loss": 0.6828,
513
+ "step": 37000
514
+ },
515
+ {
516
+ "epoch": 3.3,
517
+ "learning_rate": 4.545328247745586e-05,
518
+ "loss": 0.7149,
519
+ "step": 37500
520
+ },
521
+ {
522
+ "epoch": 3.35,
523
+ "learning_rate": 4.5390133622976085e-05,
524
+ "loss": 0.6622,
525
+ "step": 38000
526
+ },
527
+ {
528
+ "epoch": 3.39,
529
+ "learning_rate": 4.53269847684963e-05,
530
+ "loss": 0.6803,
531
+ "step": 38500
532
+ },
533
+ {
534
+ "epoch": 3.39,
535
+ "eval_bleu": 6.3565,
536
+ "eval_gen_len": 17.8899,
537
+ "eval_loss": 0.9016386270523071,
538
+ "eval_runtime": 1493.3684,
539
+ "eval_samples_per_second": 15.205,
540
+ "eval_steps_per_second": 0.951,
541
+ "step": 38500
542
+ },
543
+ {
544
+ "epoch": 3.43,
545
+ "learning_rate": 4.5263835914016524e-05,
546
+ "loss": 0.7079,
547
+ "step": 39000
548
+ },
549
+ {
550
+ "epoch": 3.48,
551
+ "learning_rate": 4.520068705953675e-05,
552
+ "loss": 0.6978,
553
+ "step": 39500
554
+ },
555
+ {
556
+ "epoch": 3.52,
557
+ "learning_rate": 4.513753820505696e-05,
558
+ "loss": 0.7528,
559
+ "step": 40000
560
+ },
561
+ {
562
+ "epoch": 3.57,
563
+ "learning_rate": 4.5074389350577186e-05,
564
+ "loss": 0.7558,
565
+ "step": 40500
566
+ },
567
+ {
568
+ "epoch": 3.61,
569
+ "learning_rate": 4.50112404960974e-05,
570
+ "loss": 0.7742,
571
+ "step": 41000
572
+ },
573
+ {
574
+ "epoch": 3.66,
575
+ "learning_rate": 4.4948091641617625e-05,
576
+ "loss": 0.7505,
577
+ "step": 41500
578
+ },
579
+ {
580
+ "epoch": 3.7,
581
+ "learning_rate": 4.488494278713785e-05,
582
+ "loss": 0.7185,
583
+ "step": 42000
584
+ },
585
+ {
586
+ "epoch": 3.74,
587
+ "learning_rate": 4.4821793932658064e-05,
588
+ "loss": 0.7024,
589
+ "step": 42500
590
+ },
591
+ {
592
+ "epoch": 3.79,
593
+ "learning_rate": 4.475864507817829e-05,
594
+ "loss": 0.7262,
595
+ "step": 43000
596
+ },
597
+ {
598
+ "epoch": 3.83,
599
+ "learning_rate": 4.46954962236985e-05,
600
+ "loss": 0.7394,
601
+ "step": 43500
602
+ },
603
+ {
604
+ "epoch": 3.88,
605
+ "learning_rate": 4.4632347369218726e-05,
606
+ "loss": 0.733,
607
+ "step": 44000
608
+ },
609
+ {
610
+ "epoch": 3.88,
611
+ "eval_bleu": 7.0351,
612
+ "eval_gen_len": 17.6112,
613
+ "eval_loss": 0.9225591421127319,
614
+ "eval_runtime": 1483.9686,
615
+ "eval_samples_per_second": 15.302,
616
+ "eval_steps_per_second": 0.957,
617
+ "step": 44000
618
+ },
619
+ {
620
+ "epoch": 3.92,
621
+ "learning_rate": 4.456919851473895e-05,
622
+ "loss": 0.742,
623
+ "step": 44500
624
+ },
625
+ {
626
+ "epoch": 3.96,
627
+ "learning_rate": 4.4506049660259165e-05,
628
+ "loss": 0.735,
629
+ "step": 45000
630
+ },
631
+ {
632
+ "epoch": 4.01,
633
+ "learning_rate": 4.444290080577939e-05,
634
+ "loss": 0.6992,
635
+ "step": 45500
636
+ },
637
+ {
638
+ "epoch": 4.05,
639
+ "learning_rate": 4.4379751951299604e-05,
640
+ "loss": 0.7007,
641
+ "step": 46000
642
+ },
643
+ {
644
+ "epoch": 4.1,
645
+ "learning_rate": 4.431660309681983e-05,
646
+ "loss": 0.6979,
647
+ "step": 46500
648
+ },
649
+ {
650
+ "epoch": 4.14,
651
+ "learning_rate": 4.425345424234005e-05,
652
+ "loss": 0.6515,
653
+ "step": 47000
654
+ },
655
+ {
656
+ "epoch": 4.18,
657
+ "learning_rate": 4.4190305387860266e-05,
658
+ "loss": 0.6919,
659
+ "step": 47500
660
+ },
661
+ {
662
+ "epoch": 4.23,
663
+ "learning_rate": 4.412715653338049e-05,
664
+ "loss": 0.6241,
665
+ "step": 48000
666
+ },
667
+ {
668
+ "epoch": 4.27,
669
+ "learning_rate": 4.406400767890071e-05,
670
+ "loss": 0.6196,
671
+ "step": 48500
672
+ },
673
+ {
674
+ "epoch": 4.32,
675
+ "learning_rate": 4.400085882442093e-05,
676
+ "loss": 0.6557,
677
+ "step": 49000
678
+ },
679
+ {
680
+ "epoch": 4.36,
681
+ "learning_rate": 4.393770996994115e-05,
682
+ "loss": 0.6601,
683
+ "step": 49500
684
+ },
685
+ {
686
+ "epoch": 4.36,
687
+ "eval_bleu": 5.3084,
688
+ "eval_gen_len": 17.4292,
689
+ "eval_loss": 0.9806875586509705,
690
+ "eval_runtime": 1461.811,
691
+ "eval_samples_per_second": 15.533,
692
+ "eval_steps_per_second": 0.971,
693
+ "step": 49500
694
+ },
695
+ {
696
+ "epoch": 4.4,
697
+ "learning_rate": 4.387456111546137e-05,
698
+ "loss": 0.7354,
699
+ "step": 50000
700
+ },
701
+ {
702
+ "epoch": 4.45,
703
+ "learning_rate": 4.381141226098159e-05,
704
+ "loss": 0.7452,
705
+ "step": 50500
706
+ },
707
+ {
708
+ "epoch": 4.49,
709
+ "learning_rate": 4.374826340650181e-05,
710
+ "loss": 0.7175,
711
+ "step": 51000
712
+ },
713
+ {
714
+ "epoch": 4.54,
715
+ "learning_rate": 4.368511455202203e-05,
716
+ "loss": 0.7101,
717
+ "step": 51500
718
+ },
719
+ {
720
+ "epoch": 4.58,
721
+ "learning_rate": 4.362196569754225e-05,
722
+ "loss": 0.7068,
723
+ "step": 52000
724
+ },
725
+ {
726
+ "epoch": 4.62,
727
+ "learning_rate": 4.355881684306247e-05,
728
+ "loss": 0.6404,
729
+ "step": 52500
730
+ },
731
+ {
732
+ "epoch": 4.67,
733
+ "learning_rate": 4.349566798858269e-05,
734
+ "loss": 0.6316,
735
+ "step": 53000
736
+ },
737
+ {
738
+ "epoch": 4.71,
739
+ "learning_rate": 4.3432519134102913e-05,
740
+ "loss": 0.598,
741
+ "step": 53500
742
+ },
743
+ {
744
+ "epoch": 4.76,
745
+ "learning_rate": 4.336937027962313e-05,
746
+ "loss": 0.6091,
747
+ "step": 54000
748
+ },
749
+ {
750
+ "epoch": 4.8,
751
+ "learning_rate": 4.330622142514335e-05,
752
+ "loss": 0.695,
753
+ "step": 54500
754
+ },
755
+ {
756
+ "epoch": 4.84,
757
+ "learning_rate": 4.324307257066357e-05,
758
+ "loss": 0.6933,
759
+ "step": 55000
760
+ },
761
+ {
762
+ "epoch": 4.84,
763
+ "eval_bleu": 6.8389,
764
+ "eval_gen_len": 17.5131,
765
+ "eval_loss": 0.923828661441803,
766
+ "eval_runtime": 1455.6892,
767
+ "eval_samples_per_second": 15.599,
768
+ "eval_steps_per_second": 0.975,
769
+ "step": 55000
770
+ }
771
+ ],
772
+ "logging_steps": 500,
773
+ "max_steps": 397390,
774
+ "num_train_epochs": 35,
775
+ "save_steps": 5500,
776
+ "total_flos": 1.907212892088828e+18,
777
+ "trial_name": null,
778
+ "trial_params": null
779
+ }
checkpoint-55000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2dd88a9d2ace72595383bc2566ed0c69120d645a38262a75472796a55099c32
3
+ size 4664
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": 250012,
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": 4.84,
3
+ "eval_bleu": 6.2934,
4
+ "eval_gen_len": 17.4862,
5
+ "eval_loss": 0.8995155096054077,
6
+ "eval_runtime": 1459.5128,
7
+ "eval_samples": 22707,
8
+ "eval_samples_per_second": 15.558,
9
+ "eval_steps_per_second": 0.973
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": 250012,
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
+ }
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 6.6934,
3
+ "predict_gen_len": 17.4855,
4
+ "predict_loss": 0.8966056704521179,
5
+ "predict_runtime": 1467.5245,
6
+ "predict_samples": 22707,
7
+ "predict_samples_per_second": 15.473,
8
+ "predict_steps_per_second": 0.968
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5c3cd3ec0addbdd6dc608f670bd2e4a00af46ea2a899f929bf197745fa4165c
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 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fc2a9a463953c664738ef8571ed97361b8de258b2aca93c0214a89676a73379
3
+ size 17110043
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": "ko_KR",
525
+ "tgt_lang": "ja_XX",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.84,
3
+ "train_loss": 0.9352452936345881,
4
+ "train_runtime": 56979.8243,
5
+ "train_samples": 181661,
6
+ "train_samples_per_second": 111.586,
7
+ "train_steps_per_second": 6.974
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,788 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8995155096054077,
3
+ "best_model_checkpoint": "./mbartLarge_koja_37p/checkpoint-33000",
4
+ "epoch": 4.844107803417298,
5
+ "eval_steps": 5500,
6
+ "global_step": 55000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 1.6666666666666667e-05,
14
+ "loss": 3.5939,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.09,
19
+ "learning_rate": 3.3333333333333335e-05,
20
+ "loss": 3.0254,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.13,
25
+ "learning_rate": 5e-05,
26
+ "loss": 1.9133,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.18,
31
+ "learning_rate": 4.9936851145520225e-05,
32
+ "loss": 1.672,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.22,
37
+ "learning_rate": 4.987370229104044e-05,
38
+ "loss": 1.56,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.26,
43
+ "learning_rate": 4.9810553436560664e-05,
44
+ "loss": 1.4874,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.31,
49
+ "learning_rate": 4.974740458208088e-05,
50
+ "loss": 1.4345,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.35,
55
+ "learning_rate": 4.96842557276011e-05,
56
+ "loss": 1.3676,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.4,
61
+ "learning_rate": 4.9621106873121326e-05,
62
+ "loss": 1.3183,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.44,
67
+ "learning_rate": 4.955795801864154e-05,
68
+ "loss": 1.2987,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.48,
73
+ "learning_rate": 4.9494809164161765e-05,
74
+ "loss": 1.2712,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.48,
79
+ "eval_bleu": 3.0858,
80
+ "eval_gen_len": 18.1085,
81
+ "eval_loss": 1.2203737497329712,
82
+ "eval_runtime": 1528.8304,
83
+ "eval_samples_per_second": 14.853,
84
+ "eval_steps_per_second": 0.929,
85
+ "step": 5500
86
+ },
87
+ {
88
+ "epoch": 0.53,
89
+ "learning_rate": 4.943166030968199e-05,
90
+ "loss": 1.2405,
91
+ "step": 6000
92
+ },
93
+ {
94
+ "epoch": 0.57,
95
+ "learning_rate": 4.9368511455202204e-05,
96
+ "loss": 1.2262,
97
+ "step": 6500
98
+ },
99
+ {
100
+ "epoch": 0.62,
101
+ "learning_rate": 4.930536260072243e-05,
102
+ "loss": 1.1959,
103
+ "step": 7000
104
+ },
105
+ {
106
+ "epoch": 0.66,
107
+ "learning_rate": 4.924221374624264e-05,
108
+ "loss": 1.182,
109
+ "step": 7500
110
+ },
111
+ {
112
+ "epoch": 0.7,
113
+ "learning_rate": 4.9179064891762866e-05,
114
+ "loss": 1.1683,
115
+ "step": 8000
116
+ },
117
+ {
118
+ "epoch": 0.75,
119
+ "learning_rate": 4.911591603728309e-05,
120
+ "loss": 1.1489,
121
+ "step": 8500
122
+ },
123
+ {
124
+ "epoch": 0.79,
125
+ "learning_rate": 4.9052767182803305e-05,
126
+ "loss": 1.1365,
127
+ "step": 9000
128
+ },
129
+ {
130
+ "epoch": 0.84,
131
+ "learning_rate": 4.898961832832353e-05,
132
+ "loss": 1.1313,
133
+ "step": 9500
134
+ },
135
+ {
136
+ "epoch": 0.88,
137
+ "learning_rate": 4.8926469473843744e-05,
138
+ "loss": 1.1172,
139
+ "step": 10000
140
+ },
141
+ {
142
+ "epoch": 0.92,
143
+ "learning_rate": 4.886332061936397e-05,
144
+ "loss": 1.1123,
145
+ "step": 10500
146
+ },
147
+ {
148
+ "epoch": 0.97,
149
+ "learning_rate": 4.880017176488419e-05,
150
+ "loss": 1.0946,
151
+ "step": 11000
152
+ },
153
+ {
154
+ "epoch": 0.97,
155
+ "eval_bleu": 3.3162,
156
+ "eval_gen_len": 17.8651,
157
+ "eval_loss": 1.0577712059020996,
158
+ "eval_runtime": 1538.4057,
159
+ "eval_samples_per_second": 14.76,
160
+ "eval_steps_per_second": 0.923,
161
+ "step": 11000
162
+ },
163
+ {
164
+ "epoch": 1.01,
165
+ "learning_rate": 4.8737022910404406e-05,
166
+ "loss": 1.0856,
167
+ "step": 11500
168
+ },
169
+ {
170
+ "epoch": 1.06,
171
+ "learning_rate": 4.867387405592463e-05,
172
+ "loss": 1.0716,
173
+ "step": 12000
174
+ },
175
+ {
176
+ "epoch": 1.1,
177
+ "learning_rate": 4.8610725201444845e-05,
178
+ "loss": 1.0736,
179
+ "step": 12500
180
+ },
181
+ {
182
+ "epoch": 1.14,
183
+ "learning_rate": 4.854757634696507e-05,
184
+ "loss": 1.0367,
185
+ "step": 13000
186
+ },
187
+ {
188
+ "epoch": 1.19,
189
+ "learning_rate": 4.848442749248529e-05,
190
+ "loss": 1.0204,
191
+ "step": 13500
192
+ },
193
+ {
194
+ "epoch": 1.23,
195
+ "learning_rate": 4.842127863800551e-05,
196
+ "loss": 1.0533,
197
+ "step": 14000
198
+ },
199
+ {
200
+ "epoch": 1.28,
201
+ "learning_rate": 4.835812978352573e-05,
202
+ "loss": 1.0426,
203
+ "step": 14500
204
+ },
205
+ {
206
+ "epoch": 1.32,
207
+ "learning_rate": 4.829498092904595e-05,
208
+ "loss": 1.0261,
209
+ "step": 15000
210
+ },
211
+ {
212
+ "epoch": 1.37,
213
+ "learning_rate": 4.823183207456617e-05,
214
+ "loss": 1.0111,
215
+ "step": 15500
216
+ },
217
+ {
218
+ "epoch": 1.41,
219
+ "learning_rate": 4.816868322008639e-05,
220
+ "loss": 0.9862,
221
+ "step": 16000
222
+ },
223
+ {
224
+ "epoch": 1.45,
225
+ "learning_rate": 4.810553436560661e-05,
226
+ "loss": 0.9546,
227
+ "step": 16500
228
+ },
229
+ {
230
+ "epoch": 1.45,
231
+ "eval_bleu": 5.6024,
232
+ "eval_gen_len": 17.902,
233
+ "eval_loss": 0.9687988758087158,
234
+ "eval_runtime": 1530.7999,
235
+ "eval_samples_per_second": 14.833,
236
+ "eval_steps_per_second": 0.928,
237
+ "step": 16500
238
+ },
239
+ {
240
+ "epoch": 1.5,
241
+ "learning_rate": 4.804238551112683e-05,
242
+ "loss": 0.9642,
243
+ "step": 17000
244
+ },
245
+ {
246
+ "epoch": 1.54,
247
+ "learning_rate": 4.7979236656647054e-05,
248
+ "loss": 0.9364,
249
+ "step": 17500
250
+ },
251
+ {
252
+ "epoch": 1.59,
253
+ "learning_rate": 4.791608780216727e-05,
254
+ "loss": 0.931,
255
+ "step": 18000
256
+ },
257
+ {
258
+ "epoch": 1.63,
259
+ "learning_rate": 4.785293894768749e-05,
260
+ "loss": 0.915,
261
+ "step": 18500
262
+ },
263
+ {
264
+ "epoch": 1.67,
265
+ "learning_rate": 4.778979009320771e-05,
266
+ "loss": 0.9043,
267
+ "step": 19000
268
+ },
269
+ {
270
+ "epoch": 1.72,
271
+ "learning_rate": 4.772664123872793e-05,
272
+ "loss": 0.9158,
273
+ "step": 19500
274
+ },
275
+ {
276
+ "epoch": 1.76,
277
+ "learning_rate": 4.7663492384248155e-05,
278
+ "loss": 0.9068,
279
+ "step": 20000
280
+ },
281
+ {
282
+ "epoch": 1.81,
283
+ "learning_rate": 4.760034352976837e-05,
284
+ "loss": 0.9314,
285
+ "step": 20500
286
+ },
287
+ {
288
+ "epoch": 1.85,
289
+ "learning_rate": 4.7537194675288594e-05,
290
+ "loss": 0.9112,
291
+ "step": 21000
292
+ },
293
+ {
294
+ "epoch": 1.89,
295
+ "learning_rate": 4.747404582080881e-05,
296
+ "loss": 0.8862,
297
+ "step": 21500
298
+ },
299
+ {
300
+ "epoch": 1.94,
301
+ "learning_rate": 4.741089696632903e-05,
302
+ "loss": 0.89,
303
+ "step": 22000
304
+ },
305
+ {
306
+ "epoch": 1.94,
307
+ "eval_bleu": 5.1453,
308
+ "eval_gen_len": 17.6144,
309
+ "eval_loss": 0.941440761089325,
310
+ "eval_runtime": 1490.4507,
311
+ "eval_samples_per_second": 15.235,
312
+ "eval_steps_per_second": 0.953,
313
+ "step": 22000
314
+ },
315
+ {
316
+ "epoch": 1.98,
317
+ "learning_rate": 4.7347748111849256e-05,
318
+ "loss": 0.8726,
319
+ "step": 22500
320
+ },
321
+ {
322
+ "epoch": 2.03,
323
+ "learning_rate": 4.728459925736947e-05,
324
+ "loss": 0.8749,
325
+ "step": 23000
326
+ },
327
+ {
328
+ "epoch": 2.07,
329
+ "learning_rate": 4.7221450402889695e-05,
330
+ "loss": 0.858,
331
+ "step": 23500
332
+ },
333
+ {
334
+ "epoch": 2.11,
335
+ "learning_rate": 4.715830154840992e-05,
336
+ "loss": 0.8479,
337
+ "step": 24000
338
+ },
339
+ {
340
+ "epoch": 2.16,
341
+ "learning_rate": 4.7095152693930134e-05,
342
+ "loss": 0.8394,
343
+ "step": 24500
344
+ },
345
+ {
346
+ "epoch": 2.2,
347
+ "learning_rate": 4.703200383945036e-05,
348
+ "loss": 0.8472,
349
+ "step": 25000
350
+ },
351
+ {
352
+ "epoch": 2.25,
353
+ "learning_rate": 4.696885498497057e-05,
354
+ "loss": 0.7793,
355
+ "step": 25500
356
+ },
357
+ {
358
+ "epoch": 2.29,
359
+ "learning_rate": 4.6905706130490796e-05,
360
+ "loss": 0.7753,
361
+ "step": 26000
362
+ },
363
+ {
364
+ "epoch": 2.33,
365
+ "learning_rate": 4.684255727601102e-05,
366
+ "loss": 0.7655,
367
+ "step": 26500
368
+ },
369
+ {
370
+ "epoch": 2.38,
371
+ "learning_rate": 4.6779408421531235e-05,
372
+ "loss": 0.7666,
373
+ "step": 27000
374
+ },
375
+ {
376
+ "epoch": 2.42,
377
+ "learning_rate": 4.671625956705146e-05,
378
+ "loss": 0.834,
379
+ "step": 27500
380
+ },
381
+ {
382
+ "epoch": 2.42,
383
+ "eval_bleu": 5.3985,
384
+ "eval_gen_len": 17.6899,
385
+ "eval_loss": 0.9212985038757324,
386
+ "eval_runtime": 1489.7985,
387
+ "eval_samples_per_second": 15.242,
388
+ "eval_steps_per_second": 0.953,
389
+ "step": 27500
390
+ },
391
+ {
392
+ "epoch": 2.47,
393
+ "learning_rate": 4.6653110712571674e-05,
394
+ "loss": 0.8098,
395
+ "step": 28000
396
+ },
397
+ {
398
+ "epoch": 2.51,
399
+ "learning_rate": 4.65899618580919e-05,
400
+ "loss": 0.8153,
401
+ "step": 28500
402
+ },
403
+ {
404
+ "epoch": 2.55,
405
+ "learning_rate": 4.652681300361212e-05,
406
+ "loss": 0.7837,
407
+ "step": 29000
408
+ },
409
+ {
410
+ "epoch": 2.6,
411
+ "learning_rate": 4.6463664149132336e-05,
412
+ "loss": 0.8255,
413
+ "step": 29500
414
+ },
415
+ {
416
+ "epoch": 2.64,
417
+ "learning_rate": 4.640051529465256e-05,
418
+ "loss": 0.8345,
419
+ "step": 30000
420
+ },
421
+ {
422
+ "epoch": 2.69,
423
+ "learning_rate": 4.6337366440172775e-05,
424
+ "loss": 0.8061,
425
+ "step": 30500
426
+ },
427
+ {
428
+ "epoch": 2.73,
429
+ "learning_rate": 4.6274217585693e-05,
430
+ "loss": 0.7743,
431
+ "step": 31000
432
+ },
433
+ {
434
+ "epoch": 2.77,
435
+ "learning_rate": 4.621106873121322e-05,
436
+ "loss": 0.7759,
437
+ "step": 31500
438
+ },
439
+ {
440
+ "epoch": 2.82,
441
+ "learning_rate": 4.614791987673344e-05,
442
+ "loss": 0.7777,
443
+ "step": 32000
444
+ },
445
+ {
446
+ "epoch": 2.86,
447
+ "learning_rate": 4.608477102225366e-05,
448
+ "loss": 0.7537,
449
+ "step": 32500
450
+ },
451
+ {
452
+ "epoch": 2.91,
453
+ "learning_rate": 4.602162216777388e-05,
454
+ "loss": 0.7439,
455
+ "step": 33000
456
+ },
457
+ {
458
+ "epoch": 2.91,
459
+ "eval_bleu": 6.2934,
460
+ "eval_gen_len": 17.4862,
461
+ "eval_loss": 0.8995155096054077,
462
+ "eval_runtime": 1459.4956,
463
+ "eval_samples_per_second": 15.558,
464
+ "eval_steps_per_second": 0.973,
465
+ "step": 33000
466
+ },
467
+ {
468
+ "epoch": 2.95,
469
+ "learning_rate": 4.59584733132941e-05,
470
+ "loss": 0.7424,
471
+ "step": 33500
472
+ },
473
+ {
474
+ "epoch": 2.99,
475
+ "learning_rate": 4.589532445881432e-05,
476
+ "loss": 0.7416,
477
+ "step": 34000
478
+ },
479
+ {
480
+ "epoch": 3.04,
481
+ "learning_rate": 4.583217560433454e-05,
482
+ "loss": 0.7722,
483
+ "step": 34500
484
+ },
485
+ {
486
+ "epoch": 3.08,
487
+ "learning_rate": 4.576902674985476e-05,
488
+ "loss": 0.7967,
489
+ "step": 35000
490
+ },
491
+ {
492
+ "epoch": 3.13,
493
+ "learning_rate": 4.5705877895374984e-05,
494
+ "loss": 0.7323,
495
+ "step": 35500
496
+ },
497
+ {
498
+ "epoch": 3.17,
499
+ "learning_rate": 4.56427290408952e-05,
500
+ "loss": 0.7171,
501
+ "step": 36000
502
+ },
503
+ {
504
+ "epoch": 3.21,
505
+ "learning_rate": 4.557958018641542e-05,
506
+ "loss": 0.7044,
507
+ "step": 36500
508
+ },
509
+ {
510
+ "epoch": 3.26,
511
+ "learning_rate": 4.551643133193564e-05,
512
+ "loss": 0.6828,
513
+ "step": 37000
514
+ },
515
+ {
516
+ "epoch": 3.3,
517
+ "learning_rate": 4.545328247745586e-05,
518
+ "loss": 0.7149,
519
+ "step": 37500
520
+ },
521
+ {
522
+ "epoch": 3.35,
523
+ "learning_rate": 4.5390133622976085e-05,
524
+ "loss": 0.6622,
525
+ "step": 38000
526
+ },
527
+ {
528
+ "epoch": 3.39,
529
+ "learning_rate": 4.53269847684963e-05,
530
+ "loss": 0.6803,
531
+ "step": 38500
532
+ },
533
+ {
534
+ "epoch": 3.39,
535
+ "eval_bleu": 6.3565,
536
+ "eval_gen_len": 17.8899,
537
+ "eval_loss": 0.9016386270523071,
538
+ "eval_runtime": 1493.3684,
539
+ "eval_samples_per_second": 15.205,
540
+ "eval_steps_per_second": 0.951,
541
+ "step": 38500
542
+ },
543
+ {
544
+ "epoch": 3.43,
545
+ "learning_rate": 4.5263835914016524e-05,
546
+ "loss": 0.7079,
547
+ "step": 39000
548
+ },
549
+ {
550
+ "epoch": 3.48,
551
+ "learning_rate": 4.520068705953675e-05,
552
+ "loss": 0.6978,
553
+ "step": 39500
554
+ },
555
+ {
556
+ "epoch": 3.52,
557
+ "learning_rate": 4.513753820505696e-05,
558
+ "loss": 0.7528,
559
+ "step": 40000
560
+ },
561
+ {
562
+ "epoch": 3.57,
563
+ "learning_rate": 4.5074389350577186e-05,
564
+ "loss": 0.7558,
565
+ "step": 40500
566
+ },
567
+ {
568
+ "epoch": 3.61,
569
+ "learning_rate": 4.50112404960974e-05,
570
+ "loss": 0.7742,
571
+ "step": 41000
572
+ },
573
+ {
574
+ "epoch": 3.66,
575
+ "learning_rate": 4.4948091641617625e-05,
576
+ "loss": 0.7505,
577
+ "step": 41500
578
+ },
579
+ {
580
+ "epoch": 3.7,
581
+ "learning_rate": 4.488494278713785e-05,
582
+ "loss": 0.7185,
583
+ "step": 42000
584
+ },
585
+ {
586
+ "epoch": 3.74,
587
+ "learning_rate": 4.4821793932658064e-05,
588
+ "loss": 0.7024,
589
+ "step": 42500
590
+ },
591
+ {
592
+ "epoch": 3.79,
593
+ "learning_rate": 4.475864507817829e-05,
594
+ "loss": 0.7262,
595
+ "step": 43000
596
+ },
597
+ {
598
+ "epoch": 3.83,
599
+ "learning_rate": 4.46954962236985e-05,
600
+ "loss": 0.7394,
601
+ "step": 43500
602
+ },
603
+ {
604
+ "epoch": 3.88,
605
+ "learning_rate": 4.4632347369218726e-05,
606
+ "loss": 0.733,
607
+ "step": 44000
608
+ },
609
+ {
610
+ "epoch": 3.88,
611
+ "eval_bleu": 7.0351,
612
+ "eval_gen_len": 17.6112,
613
+ "eval_loss": 0.9225591421127319,
614
+ "eval_runtime": 1483.9686,
615
+ "eval_samples_per_second": 15.302,
616
+ "eval_steps_per_second": 0.957,
617
+ "step": 44000
618
+ },
619
+ {
620
+ "epoch": 3.92,
621
+ "learning_rate": 4.456919851473895e-05,
622
+ "loss": 0.742,
623
+ "step": 44500
624
+ },
625
+ {
626
+ "epoch": 3.96,
627
+ "learning_rate": 4.4506049660259165e-05,
628
+ "loss": 0.735,
629
+ "step": 45000
630
+ },
631
+ {
632
+ "epoch": 4.01,
633
+ "learning_rate": 4.444290080577939e-05,
634
+ "loss": 0.6992,
635
+ "step": 45500
636
+ },
637
+ {
638
+ "epoch": 4.05,
639
+ "learning_rate": 4.4379751951299604e-05,
640
+ "loss": 0.7007,
641
+ "step": 46000
642
+ },
643
+ {
644
+ "epoch": 4.1,
645
+ "learning_rate": 4.431660309681983e-05,
646
+ "loss": 0.6979,
647
+ "step": 46500
648
+ },
649
+ {
650
+ "epoch": 4.14,
651
+ "learning_rate": 4.425345424234005e-05,
652
+ "loss": 0.6515,
653
+ "step": 47000
654
+ },
655
+ {
656
+ "epoch": 4.18,
657
+ "learning_rate": 4.4190305387860266e-05,
658
+ "loss": 0.6919,
659
+ "step": 47500
660
+ },
661
+ {
662
+ "epoch": 4.23,
663
+ "learning_rate": 4.412715653338049e-05,
664
+ "loss": 0.6241,
665
+ "step": 48000
666
+ },
667
+ {
668
+ "epoch": 4.27,
669
+ "learning_rate": 4.406400767890071e-05,
670
+ "loss": 0.6196,
671
+ "step": 48500
672
+ },
673
+ {
674
+ "epoch": 4.32,
675
+ "learning_rate": 4.400085882442093e-05,
676
+ "loss": 0.6557,
677
+ "step": 49000
678
+ },
679
+ {
680
+ "epoch": 4.36,
681
+ "learning_rate": 4.393770996994115e-05,
682
+ "loss": 0.6601,
683
+ "step": 49500
684
+ },
685
+ {
686
+ "epoch": 4.36,
687
+ "eval_bleu": 5.3084,
688
+ "eval_gen_len": 17.4292,
689
+ "eval_loss": 0.9806875586509705,
690
+ "eval_runtime": 1461.811,
691
+ "eval_samples_per_second": 15.533,
692
+ "eval_steps_per_second": 0.971,
693
+ "step": 49500
694
+ },
695
+ {
696
+ "epoch": 4.4,
697
+ "learning_rate": 4.387456111546137e-05,
698
+ "loss": 0.7354,
699
+ "step": 50000
700
+ },
701
+ {
702
+ "epoch": 4.45,
703
+ "learning_rate": 4.381141226098159e-05,
704
+ "loss": 0.7452,
705
+ "step": 50500
706
+ },
707
+ {
708
+ "epoch": 4.49,
709
+ "learning_rate": 4.374826340650181e-05,
710
+ "loss": 0.7175,
711
+ "step": 51000
712
+ },
713
+ {
714
+ "epoch": 4.54,
715
+ "learning_rate": 4.368511455202203e-05,
716
+ "loss": 0.7101,
717
+ "step": 51500
718
+ },
719
+ {
720
+ "epoch": 4.58,
721
+ "learning_rate": 4.362196569754225e-05,
722
+ "loss": 0.7068,
723
+ "step": 52000
724
+ },
725
+ {
726
+ "epoch": 4.62,
727
+ "learning_rate": 4.355881684306247e-05,
728
+ "loss": 0.6404,
729
+ "step": 52500
730
+ },
731
+ {
732
+ "epoch": 4.67,
733
+ "learning_rate": 4.349566798858269e-05,
734
+ "loss": 0.6316,
735
+ "step": 53000
736
+ },
737
+ {
738
+ "epoch": 4.71,
739
+ "learning_rate": 4.3432519134102913e-05,
740
+ "loss": 0.598,
741
+ "step": 53500
742
+ },
743
+ {
744
+ "epoch": 4.76,
745
+ "learning_rate": 4.336937027962313e-05,
746
+ "loss": 0.6091,
747
+ "step": 54000
748
+ },
749
+ {
750
+ "epoch": 4.8,
751
+ "learning_rate": 4.330622142514335e-05,
752
+ "loss": 0.695,
753
+ "step": 54500
754
+ },
755
+ {
756
+ "epoch": 4.84,
757
+ "learning_rate": 4.324307257066357e-05,
758
+ "loss": 0.6933,
759
+ "step": 55000
760
+ },
761
+ {
762
+ "epoch": 4.84,
763
+ "eval_bleu": 6.8389,
764
+ "eval_gen_len": 17.5131,
765
+ "eval_loss": 0.923828661441803,
766
+ "eval_runtime": 1455.6892,
767
+ "eval_samples_per_second": 15.599,
768
+ "eval_steps_per_second": 0.975,
769
+ "step": 55000
770
+ },
771
+ {
772
+ "epoch": 4.84,
773
+ "step": 55000,
774
+ "total_flos": 1.907212892088828e+18,
775
+ "train_loss": 0.9352452936345881,
776
+ "train_runtime": 56979.8243,
777
+ "train_samples_per_second": 111.586,
778
+ "train_steps_per_second": 6.974
779
+ }
780
+ ],
781
+ "logging_steps": 500,
782
+ "max_steps": 397390,
783
+ "num_train_epochs": 35,
784
+ "save_steps": 5500,
785
+ "total_flos": 1.907212892088828e+18,
786
+ "trial_name": null,
787
+ "trial_params": null
788
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2dd88a9d2ace72595383bc2566ed0c69120d645a38262a75472796a55099c32
3
+ size 4664