Hugging Face's logo

yesj1234
/
mbart-mmt_mid1_zh-ko

Text2Text Generation
Transformers
PyTorch
Chinese
Korean
mbart
Generated from Trainer
Inference Endpoints
Model card Files Community
yesj1234 commited on
Commit
4421a6e
·
1 Parent(s): 3dddb63

Upload folder using huggingface_hub

Browse files
Files changed (49) hide show
  1. .gitattributes +3 -0
  2. README.md +83 -0
  3. added_tokens.json +32 -0
  4. all_results.json +22 -0
  5. checkpoint-12000/added_tokens.json +32 -0
  6. checkpoint-12000/config.json +60 -0
  7. checkpoint-12000/generation_config.json +10 -0
  8. checkpoint-12000/optimizer.pt +3 -0
  9. checkpoint-12000/pytorch_model.bin +3 -0
  10. checkpoint-12000/rng_state_0.pth +3 -0
  11. checkpoint-12000/rng_state_1.pth +3 -0
  12. checkpoint-12000/rng_state_2.pth +3 -0
  13. checkpoint-12000/rng_state_3.pth +3 -0
  14. checkpoint-12000/scheduler.pt +3 -0
  15. checkpoint-12000/sentencepiece.bpe.model +3 -0
  16. checkpoint-12000/special_tokens_map.json +36 -0
  17. checkpoint-12000/tokenizer.json +3 -0
  18. checkpoint-12000/tokenizer_config.json +283 -0
  19. checkpoint-12000/trainer_state.json +283 -0
  20. checkpoint-12000/training_args.bin +3 -0
  21. checkpoint-16000/added_tokens.json +32 -0
  22. checkpoint-16000/config.json +60 -0
  23. checkpoint-16000/generation_config.json +10 -0
  24. checkpoint-16000/optimizer.pt +3 -0
  25. checkpoint-16000/pytorch_model.bin +3 -0
  26. checkpoint-16000/rng_state_0.pth +3 -0
  27. checkpoint-16000/rng_state_1.pth +3 -0
  28. checkpoint-16000/rng_state_2.pth +3 -0
  29. checkpoint-16000/rng_state_3.pth +3 -0
  30. checkpoint-16000/scheduler.pt +3 -0
  31. checkpoint-16000/sentencepiece.bpe.model +3 -0
  32. checkpoint-16000/special_tokens_map.json +36 -0
  33. checkpoint-16000/tokenizer.json +3 -0
  34. checkpoint-16000/tokenizer_config.json +283 -0
  35. checkpoint-16000/trainer_state.json +371 -0
  36. checkpoint-16000/training_args.bin +3 -0
  37. config.json +60 -0
  38. eval_results.json +10 -0
  39. generated_predictions.txt +0 -0
  40. generation_config.json +10 -0
  41. predict_results.json +9 -0
  42. pytorch_model.bin +3 -0
  43. sentencepiece.bpe.model +3 -0
  44. special_tokens_map.json +36 -0
  45. tokenizer.json +3 -0
  46. tokenizer_config.json +283 -0
  47. train_results.json +8 -0
  48. trainer_state.json +380 -0
  49. 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-12000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-16000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - zh
4
+ - ko
5
+ base_model: facebook/mbart-large-cc25
6
+ tags:
7
+ - generated_from_trainer
8
+ metrics:
9
+ - bleu
10
+ model-index:
11
+ - name: zh-kr_mid
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
+ # zh-kr_mid
19
+
20
+ This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 2.5557
23
+ - Bleu: 16.6036
24
+ - Gen Len: 15.4901
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
+ - total_train_batch_size: 16
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: linear
53
+ - lr_scheduler_warmup_steps: 500
54
+ - num_epochs: 40
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
59
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
60
+ | 2.7248 | 0.75 | 1000 | 1.9410 | 3.2381 | 48.6095 |
61
+ | 1.5683 | 1.5 | 2000 | 1.6889 | 10.2345 | 20.4433 |
62
+ | 1.1916 | 2.25 | 3000 | 1.6843 | 13.4571 | 18.8854 |
63
+ | 1.068 | 2.99 | 4000 | 1.6390 | 15.6862 | 15.5054 |
64
+ | 0.7313 | 3.74 | 5000 | 1.7003 | 15.2014 | 16.5938 |
65
+ | 0.4832 | 4.49 | 6000 | 1.8982 | 15.0381 | 16.9068 |
66
+ | 0.3862 | 5.24 | 7000 | 2.1426 | 15.5397 | 15.6451 |
67
+ | 0.3675 | 5.99 | 8000 | 2.1168 | 15.8847 | 15.6926 |
68
+ | 0.2627 | 6.74 | 9000 | 2.2603 | 16.3603 | 15.9671 |
69
+ | 0.1955 | 7.49 | 10000 | 2.4114 | 15.7447 | 15.979 |
70
+ | 0.171 | 8.23 | 11000 | 2.5141 | 15.7852 | 15.9244 |
71
+ | 0.1702 | 8.98 | 12000 | 2.5557 | 16.6036 | 15.4901 |
72
+ | 0.1298 | 9.73 | 13000 | 2.6536 | 16.1319 | 15.5492 |
73
+ | 0.1052 | 10.48 | 14000 | 2.7586 | 16.1807 | 15.8884 |
74
+ | 0.2268 | 11.23 | 15000 | 2.7258 | 15.1752 | 15.5346 |
75
+ | 0.1327 | 11.98 | 16000 | 2.7193 | 15.8563 | 15.7971 |
76
+
77
+
78
+ ### Framework versions
79
+
80
+ - Transformers 4.34.0
81
+ - Pytorch 2.1.0+cu121
82
+ - Datasets 2.14.5
83
+ - Tokenizers 0.14.1
added_tokens.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 250026,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3,
7
+ "ar_AR": 250001,
8
+ "cs_CZ": 250002,
9
+ "de_DE": 250003,
10
+ "en_XX": 250004,
11
+ "es_XX": 250005,
12
+ "et_EE": 250006,
13
+ "fi_FI": 250007,
14
+ "fr_XX": 250008,
15
+ "gu_IN": 250009,
16
+ "hi_IN": 250010,
17
+ "it_IT": 250011,
18
+ "ja_XX": 250012,
19
+ "kk_KZ": 250013,
20
+ "ko_KR": 250014,
21
+ "lt_LT": 250015,
22
+ "lv_LV": 250016,
23
+ "my_MM": 250017,
24
+ "ne_NP": 250018,
25
+ "nl_XX": 250019,
26
+ "ro_RO": 250020,
27
+ "ru_RU": 250021,
28
+ "si_LK": 250022,
29
+ "tr_TR": 250023,
30
+ "vi_VN": 250024,
31
+ "zh_CN": 250025
32
+ }
all_results.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 11.98,
3
+ "eval_bleu": 16.6036,
4
+ "eval_gen_len": 15.4901,
5
+ "eval_loss": 2.5557188987731934,
6
+ "eval_runtime": 167.3814,
7
+ "eval_samples": 2671,
8
+ "eval_samples_per_second": 15.958,
9
+ "eval_steps_per_second": 0.998,
10
+ "predict_bleu": 15.2894,
11
+ "predict_gen_len": 15.6769,
12
+ "predict_loss": 2.6123738288879395,
13
+ "predict_runtime": 173.2964,
14
+ "predict_samples": 2671,
15
+ "predict_samples_per_second": 15.413,
16
+ "predict_steps_per_second": 0.964,
17
+ "train_loss": 0.6965520789623261,
18
+ "train_runtime": 15742.7956,
19
+ "train_samples": 21375,
20
+ "train_samples_per_second": 54.311,
21
+ "train_steps_per_second": 3.395
22
+ }
checkpoint-12000/added_tokens.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 250026,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3,
7
+ "ar_AR": 250001,
8
+ "cs_CZ": 250002,
9
+ "de_DE": 250003,
10
+ "en_XX": 250004,
11
+ "es_XX": 250005,
12
+ "et_EE": 250006,
13
+ "fi_FI": 250007,
14
+ "fr_XX": 250008,
15
+ "gu_IN": 250009,
16
+ "hi_IN": 250010,
17
+ "it_IT": 250011,
18
+ "ja_XX": 250012,
19
+ "kk_KZ": 250013,
20
+ "ko_KR": 250014,
21
+ "lt_LT": 250015,
22
+ "lv_LV": 250016,
23
+ "my_MM": 250017,
24
+ "ne_NP": 250018,
25
+ "nl_XX": 250019,
26
+ "ro_RO": 250020,
27
+ "ru_RU": 250021,
28
+ "si_LK": 250022,
29
+ "tr_TR": 250023,
30
+ "vi_VN": 250024,
31
+ "zh_CN": 250025
32
+ }
checkpoint-12000/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-cc25",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
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": 250014,
21
+ "dropout": 0.1,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 4096,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 12,
26
+ "eos_token_id": 2,
27
+ "forced_eos_token_id": 2,
28
+ "id2label": {
29
+ "0": "LABEL_0",
30
+ "1": "LABEL_1",
31
+ "2": "LABEL_2"
32
+ },
33
+ "init_std": 0.02,
34
+ "is_encoder_decoder": true,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1,
38
+ "LABEL_2": 2
39
+ },
40
+ "max_length": 1024,
41
+ "max_position_embeddings": 1024,
42
+ "model_type": "mbart",
43
+ "normalize_before": true,
44
+ "normalize_embedding": true,
45
+ "num_beams": 5,
46
+ "num_hidden_layers": 12,
47
+ "output_past": true,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": true,
50
+ "static_position_embeddings": false,
51
+ "task_specific_params": {
52
+ "translation_en_to_ro": {
53
+ "decoder_start_token_id": 250020
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.34.0",
58
+ "use_cache": true,
59
+ "vocab_size": 250027
60
+ }
checkpoint-12000/generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 250014,
4
+ "eos_token_id": 2,
5
+ "forced_eos_token_id": 2,
6
+ "max_length": 1024,
7
+ "num_beams": 5,
8
+ "pad_token_id": 1,
9
+ "transformers_version": "4.34.0"
10
+ }
checkpoint-12000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5a01af644b201d9b8b5bfb521cc22fac96609f5711e349b886792eeb3aa7923
3
+ size 4887252719
checkpoint-12000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57bc3ec79d14b77e794d98ec08abca470f52cb6ecfdb0ded73411deff2bdea21
3
+ size 2444583770
checkpoint-12000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1ff89d0330228d2a61832395d8f5bb1ad21ea2b7c9a9f6c0d979d9a1fa9b7f3
3
+ size 15024
checkpoint-12000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a2e17feb94359dde3aacd3621296f78dd6bcd7132ced92cd0ac588d72509856
3
+ size 15024
checkpoint-12000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47623681bc7b0278d319e89317815aa45285a280c5a94c6103f7aca3d85b30ae
3
+ size 15024
checkpoint-12000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33aec63437b352ea3e96a2378066c4c1b85da22a9c0154671f0b7920f8ab2431
3
+ size 15024
checkpoint-12000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58f43e7c9ccb7dae1d994eb3470c08cb28e53fa97f0c15e2a55152b945144e8f
3
+ size 1064
checkpoint-12000/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-12000/special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN"
28
+ ],
29
+ "bos_token": "<s>",
30
+ "cls_token": "<s>",
31
+ "eos_token": "</s>",
32
+ "mask_token": "<mask>",
33
+ "pad_token": "<pad>",
34
+ "sep_token": "</s>",
35
+ "unk_token": "<unk>"
36
+ }
checkpoint-12000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94bf3391e3589e53c8b8888fcbe2a854ebe851645c71bf46b7af9335fadf0db5
3
+ size 17088468
checkpoint-12000/tokenizer_config.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "<mask>",
237
+ "lstrip": true,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ }
243
+ },
244
+ "additional_special_tokens": [
245
+ "ar_AR",
246
+ "cs_CZ",
247
+ "de_DE",
248
+ "en_XX",
249
+ "es_XX",
250
+ "et_EE",
251
+ "fi_FI",
252
+ "fr_XX",
253
+ "gu_IN",
254
+ "hi_IN",
255
+ "it_IT",
256
+ "ja_XX",
257
+ "kk_KZ",
258
+ "ko_KR",
259
+ "lt_LT",
260
+ "lv_LV",
261
+ "my_MM",
262
+ "ne_NP",
263
+ "nl_XX",
264
+ "ro_RO",
265
+ "ru_RU",
266
+ "si_LK",
267
+ "tr_TR",
268
+ "vi_VN",
269
+ "zh_CN"
270
+ ],
271
+ "bos_token": "<s>",
272
+ "clean_up_tokenization_spaces": true,
273
+ "cls_token": "<s>",
274
+ "eos_token": "</s>",
275
+ "mask_token": "<mask>",
276
+ "model_max_length": 1024,
277
+ "pad_token": "<pad>",
278
+ "sep_token": "</s>",
279
+ "src_lang": "zh_CN",
280
+ "tgt_lang": "ko_KR",
281
+ "tokenizer_class": "MBartTokenizer",
282
+ "unk_token": "<unk>"
283
+ }
checkpoint-12000/trainer_state.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 16.6036,
3
+ "best_model_checkpoint": "./zh-kr_mid/checkpoint-12000",
4
+ "epoch": 8.982035928143713,
5
+ "eval_steps": 1000,
6
+ "global_step": 12000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.37,
13
+ "learning_rate": 4.91e-05,
14
+ "loss": 4.4024,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.75,
19
+ "learning_rate": 4.95362674726105e-05,
20
+ "loss": 2.7248,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.75,
25
+ "eval_bleu": 3.2381,
26
+ "eval_gen_len": 48.6095,
27
+ "eval_loss": 1.9409534931182861,
28
+ "eval_runtime": 1917.4714,
29
+ "eval_samples_per_second": 1.393,
30
+ "eval_steps_per_second": 0.087,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 1.12,
35
+ "learning_rate": 4.9064979221760484e-05,
36
+ "loss": 1.8483,
37
+ "step": 1500
38
+ },
39
+ {
40
+ "epoch": 1.5,
41
+ "learning_rate": 4.85927465054779e-05,
42
+ "loss": 1.5683,
43
+ "step": 2000
44
+ },
45
+ {
46
+ "epoch": 1.5,
47
+ "eval_bleu": 10.2345,
48
+ "eval_gen_len": 20.4433,
49
+ "eval_loss": 1.6889078617095947,
50
+ "eval_runtime": 497.5454,
51
+ "eval_samples_per_second": 5.368,
52
+ "eval_steps_per_second": 0.336,
53
+ "step": 2000
54
+ },
55
+ {
56
+ "epoch": 1.87,
57
+ "learning_rate": 4.812051378919532e-05,
58
+ "loss": 1.5205,
59
+ "step": 2500
60
+ },
61
+ {
62
+ "epoch": 2.25,
63
+ "learning_rate": 4.764828107291273e-05,
64
+ "loss": 1.1916,
65
+ "step": 3000
66
+ },
67
+ {
68
+ "epoch": 2.25,
69
+ "eval_bleu": 13.4571,
70
+ "eval_gen_len": 18.8854,
71
+ "eval_loss": 1.6842907667160034,
72
+ "eval_runtime": 538.3261,
73
+ "eval_samples_per_second": 4.962,
74
+ "eval_steps_per_second": 0.31,
75
+ "step": 3000
76
+ },
77
+ {
78
+ "epoch": 2.62,
79
+ "learning_rate": 4.7176048356630146e-05,
80
+ "loss": 1.0458,
81
+ "step": 3500
82
+ },
83
+ {
84
+ "epoch": 2.99,
85
+ "learning_rate": 4.670381564034756e-05,
86
+ "loss": 1.068,
87
+ "step": 4000
88
+ },
89
+ {
90
+ "epoch": 2.99,
91
+ "eval_bleu": 15.6862,
92
+ "eval_gen_len": 15.5054,
93
+ "eval_loss": 1.6389988660812378,
94
+ "eval_runtime": 200.5304,
95
+ "eval_samples_per_second": 13.32,
96
+ "eval_steps_per_second": 0.833,
97
+ "step": 4000
98
+ },
99
+ {
100
+ "epoch": 3.37,
101
+ "learning_rate": 4.623158292406498e-05,
102
+ "loss": 0.7097,
103
+ "step": 4500
104
+ },
105
+ {
106
+ "epoch": 3.74,
107
+ "learning_rate": 4.57593502077824e-05,
108
+ "loss": 0.7313,
109
+ "step": 5000
110
+ },
111
+ {
112
+ "epoch": 3.74,
113
+ "eval_bleu": 15.2014,
114
+ "eval_gen_len": 16.5938,
115
+ "eval_loss": 1.7002748250961304,
116
+ "eval_runtime": 222.4222,
117
+ "eval_samples_per_second": 12.009,
118
+ "eval_steps_per_second": 0.751,
119
+ "step": 5000
120
+ },
121
+ {
122
+ "epoch": 4.12,
123
+ "learning_rate": 4.5287117491499814e-05,
124
+ "loss": 0.6587,
125
+ "step": 5500
126
+ },
127
+ {
128
+ "epoch": 4.49,
129
+ "learning_rate": 4.481488477521723e-05,
130
+ "loss": 0.4832,
131
+ "step": 6000
132
+ },
133
+ {
134
+ "epoch": 4.49,
135
+ "eval_bleu": 15.0381,
136
+ "eval_gen_len": 16.9068,
137
+ "eval_loss": 1.8982346057891846,
138
+ "eval_runtime": 258.8867,
139
+ "eval_samples_per_second": 10.317,
140
+ "eval_steps_per_second": 0.645,
141
+ "step": 6000
142
+ },
143
+ {
144
+ "epoch": 4.87,
145
+ "learning_rate": 4.434265205893465e-05,
146
+ "loss": 0.5089,
147
+ "step": 6500
148
+ },
149
+ {
150
+ "epoch": 5.24,
151
+ "learning_rate": 4.3870419342652066e-05,
152
+ "loss": 0.3862,
153
+ "step": 7000
154
+ },
155
+ {
156
+ "epoch": 5.24,
157
+ "eval_bleu": 15.5397,
158
+ "eval_gen_len": 15.6451,
159
+ "eval_loss": 2.1426022052764893,
160
+ "eval_runtime": 170.9728,
161
+ "eval_samples_per_second": 15.622,
162
+ "eval_steps_per_second": 0.977,
163
+ "step": 7000
164
+ },
165
+ {
166
+ "epoch": 5.61,
167
+ "learning_rate": 4.339913109180204e-05,
168
+ "loss": 0.3495,
169
+ "step": 7500
170
+ },
171
+ {
172
+ "epoch": 5.99,
173
+ "learning_rate": 4.292689837551946e-05,
174
+ "loss": 0.3675,
175
+ "step": 8000
176
+ },
177
+ {
178
+ "epoch": 5.99,
179
+ "eval_bleu": 15.8847,
180
+ "eval_gen_len": 15.6926,
181
+ "eval_loss": 2.1168086528778076,
182
+ "eval_runtime": 169.7878,
183
+ "eval_samples_per_second": 15.731,
184
+ "eval_steps_per_second": 0.984,
185
+ "step": 8000
186
+ },
187
+ {
188
+ "epoch": 6.36,
189
+ "learning_rate": 4.2454665659236876e-05,
190
+ "loss": 0.2443,
191
+ "step": 8500
192
+ },
193
+ {
194
+ "epoch": 6.74,
195
+ "learning_rate": 4.198243294295429e-05,
196
+ "loss": 0.2627,
197
+ "step": 9000
198
+ },
199
+ {
200
+ "epoch": 6.74,
201
+ "eval_bleu": 16.3603,
202
+ "eval_gen_len": 15.9671,
203
+ "eval_loss": 2.260291814804077,
204
+ "eval_runtime": 176.4048,
205
+ "eval_samples_per_second": 15.141,
206
+ "eval_steps_per_second": 0.947,
207
+ "step": 9000
208
+ },
209
+ {
210
+ "epoch": 7.11,
211
+ "learning_rate": 4.151020022667171e-05,
212
+ "loss": 0.2471,
213
+ "step": 9500
214
+ },
215
+ {
216
+ "epoch": 7.49,
217
+ "learning_rate": 4.103796751038912e-05,
218
+ "loss": 0.1955,
219
+ "step": 10000
220
+ },
221
+ {
222
+ "epoch": 7.49,
223
+ "eval_bleu": 15.7447,
224
+ "eval_gen_len": 15.979,
225
+ "eval_loss": 2.4113738536834717,
226
+ "eval_runtime": 200.4127,
227
+ "eval_samples_per_second": 13.327,
228
+ "eval_steps_per_second": 0.833,
229
+ "step": 10000
230
+ },
231
+ {
232
+ "epoch": 7.86,
233
+ "learning_rate": 4.056573479410654e-05,
234
+ "loss": 0.2088,
235
+ "step": 10500
236
+ },
237
+ {
238
+ "epoch": 8.23,
239
+ "learning_rate": 4.0093502077823954e-05,
240
+ "loss": 0.171,
241
+ "step": 11000
242
+ },
243
+ {
244
+ "epoch": 8.23,
245
+ "eval_bleu": 15.7852,
246
+ "eval_gen_len": 15.9244,
247
+ "eval_loss": 2.5141146183013916,
248
+ "eval_runtime": 180.5467,
249
+ "eval_samples_per_second": 14.794,
250
+ "eval_steps_per_second": 0.925,
251
+ "step": 11000
252
+ },
253
+ {
254
+ "epoch": 8.61,
255
+ "learning_rate": 3.962126936154137e-05,
256
+ "loss": 0.157,
257
+ "step": 11500
258
+ },
259
+ {
260
+ "epoch": 8.98,
261
+ "learning_rate": 3.9149981110691354e-05,
262
+ "loss": 0.1702,
263
+ "step": 12000
264
+ },
265
+ {
266
+ "epoch": 8.98,
267
+ "eval_bleu": 16.6036,
268
+ "eval_gen_len": 15.4901,
269
+ "eval_loss": 2.5557188987731934,
270
+ "eval_runtime": 168.9011,
271
+ "eval_samples_per_second": 15.814,
272
+ "eval_steps_per_second": 0.989,
273
+ "step": 12000
274
+ }
275
+ ],
276
+ "logging_steps": 500,
277
+ "max_steps": 53440,
278
+ "num_train_epochs": 40,
279
+ "save_steps": 1000,
280
+ "total_flos": 4.16088915443712e+17,
281
+ "trial_name": null,
282
+ "trial_params": null
283
+ }
checkpoint-12000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f8ee59bef192bac6b7c7efff93ae004410d1f65031577a07e5f97f4a15294d8
3
+ size 4664
checkpoint-16000/added_tokens.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</s>": 2,
3
+ "<mask>": 250026,
4
+ "<pad>": 1,
5
+ "<s>": 0,
6
+ "<unk>": 3,
7
+ "ar_AR": 250001,
8
+ "cs_CZ": 250002,
9
+ "de_DE": 250003,
10
+ "en_XX": 250004,
11
+ "es_XX": 250005,
12
+ "et_EE": 250006,
13
+ "fi_FI": 250007,
14
+ "fr_XX": 250008,
15
+ "gu_IN": 250009,
16
+ "hi_IN": 250010,
17
+ "it_IT": 250011,
18
+ "ja_XX": 250012,
19
+ "kk_KZ": 250013,
20
+ "ko_KR": 250014,
21
+ "lt_LT": 250015,
22
+ "lv_LV": 250016,
23
+ "my_MM": 250017,
24
+ "ne_NP": 250018,
25
+ "nl_XX": 250019,
26
+ "ro_RO": 250020,
27
+ "ru_RU": 250021,
28
+ "si_LK": 250022,
29
+ "tr_TR": 250023,
30
+ "vi_VN": 250024,
31
+ "zh_CN": 250025
32
+ }
checkpoint-16000/config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-cc25",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
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": 250014,
21
+ "dropout": 0.1,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 4096,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 12,
26
+ "eos_token_id": 2,
27
+ "forced_eos_token_id": 2,
28
+ "id2label": {
29
+ "0": "LABEL_0",
30
+ "1": "LABEL_1",
31
+ "2": "LABEL_2"
32
+ },
33
+ "init_std": 0.02,
34
+ "is_encoder_decoder": true,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1,
38
+ "LABEL_2": 2
39
+ },
40
+ "max_length": 1024,
41
+ "max_position_embeddings": 1024,
42
+ "model_type": "mbart",
43
+ "normalize_before": true,
44
+ "normalize_embedding": true,
45
+ "num_beams": 5,
46
+ "num_hidden_layers": 12,
47
+ "output_past": true,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": true,
50
+ "static_position_embeddings": false,
51
+ "task_specific_params": {
52
+ "translation_en_to_ro": {
53
+ "decoder_start_token_id": 250020
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.34.0",
58
+ "use_cache": true,
59
+ "vocab_size": 250027
60
+ }
checkpoint-16000/generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 250014,
4
+ "eos_token_id": 2,
5
+ "forced_eos_token_id": 2,
6
+ "max_length": 1024,
7
+ "num_beams": 5,
8
+ "pad_token_id": 1,
9
+ "transformers_version": "4.34.0"
10
+ }
checkpoint-16000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfa30d655bf97d02c0e9d54c0a00ccfd2384061d6a8361c47dce839600afedf0
3
+ size 4887252719
checkpoint-16000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed920044c3258bc64085b49d7027552bdcdc8bd1eeb6c2e108a0382b959a3b07
3
+ size 2444583770
checkpoint-16000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60b1be36ac15bab9072d95250fcd90780272d2742a77f50a417d566283029562
3
+ size 15024
checkpoint-16000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:741292a7a3f4901d413ac17c30ead6c54d0bb0ee8f5ef839a937f6049cc90683
3
+ size 15024
checkpoint-16000/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41d98f98af5dd33d78f959e268c0042577113bca190b868d91ef13f4992b1a77
3
+ size 15024
checkpoint-16000/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8df1ff97fccfdb101efdb8e0724d1521e8e9e054ec8f83f7ddb1a903c086dc4
3
+ size 15024
checkpoint-16000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:364fc9b79f751001a85f7d2d09857a5812dd9b7a582c25ff4bfb162e51d5d1bd
3
+ size 1064
checkpoint-16000/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-16000/special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN"
28
+ ],
29
+ "bos_token": "<s>",
30
+ "cls_token": "<s>",
31
+ "eos_token": "</s>",
32
+ "mask_token": "<mask>",
33
+ "pad_token": "<pad>",
34
+ "sep_token": "</s>",
35
+ "unk_token": "<unk>"
36
+ }
checkpoint-16000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94bf3391e3589e53c8b8888fcbe2a854ebe851645c71bf46b7af9335fadf0db5
3
+ size 17088468
checkpoint-16000/tokenizer_config.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "<mask>",
237
+ "lstrip": true,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ }
243
+ },
244
+ "additional_special_tokens": [
245
+ "ar_AR",
246
+ "cs_CZ",
247
+ "de_DE",
248
+ "en_XX",
249
+ "es_XX",
250
+ "et_EE",
251
+ "fi_FI",
252
+ "fr_XX",
253
+ "gu_IN",
254
+ "hi_IN",
255
+ "it_IT",
256
+ "ja_XX",
257
+ "kk_KZ",
258
+ "ko_KR",
259
+ "lt_LT",
260
+ "lv_LV",
261
+ "my_MM",
262
+ "ne_NP",
263
+ "nl_XX",
264
+ "ro_RO",
265
+ "ru_RU",
266
+ "si_LK",
267
+ "tr_TR",
268
+ "vi_VN",
269
+ "zh_CN"
270
+ ],
271
+ "bos_token": "<s>",
272
+ "clean_up_tokenization_spaces": true,
273
+ "cls_token": "<s>",
274
+ "eos_token": "</s>",
275
+ "mask_token": "<mask>",
276
+ "model_max_length": 1024,
277
+ "pad_token": "<pad>",
278
+ "sep_token": "</s>",
279
+ "src_lang": "zh_CN",
280
+ "tgt_lang": "ko_KR",
281
+ "tokenizer_class": "MBartTokenizer",
282
+ "unk_token": "<unk>"
283
+ }
checkpoint-16000/trainer_state.json ADDED
@@ -0,0 +1,371 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 16.6036,
3
+ "best_model_checkpoint": "./zh-kr_mid/checkpoint-12000",
4
+ "epoch": 11.976047904191617,
5
+ "eval_steps": 1000,
6
+ "global_step": 16000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.37,
13
+ "learning_rate": 4.91e-05,
14
+ "loss": 4.4024,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.75,
19
+ "learning_rate": 4.95362674726105e-05,
20
+ "loss": 2.7248,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.75,
25
+ "eval_bleu": 3.2381,
26
+ "eval_gen_len": 48.6095,
27
+ "eval_loss": 1.9409534931182861,
28
+ "eval_runtime": 1917.4714,
29
+ "eval_samples_per_second": 1.393,
30
+ "eval_steps_per_second": 0.087,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 1.12,
35
+ "learning_rate": 4.9064979221760484e-05,
36
+ "loss": 1.8483,
37
+ "step": 1500
38
+ },
39
+ {
40
+ "epoch": 1.5,
41
+ "learning_rate": 4.85927465054779e-05,
42
+ "loss": 1.5683,
43
+ "step": 2000
44
+ },
45
+ {
46
+ "epoch": 1.5,
47
+ "eval_bleu": 10.2345,
48
+ "eval_gen_len": 20.4433,
49
+ "eval_loss": 1.6889078617095947,
50
+ "eval_runtime": 497.5454,
51
+ "eval_samples_per_second": 5.368,
52
+ "eval_steps_per_second": 0.336,
53
+ "step": 2000
54
+ },
55
+ {
56
+ "epoch": 1.87,
57
+ "learning_rate": 4.812051378919532e-05,
58
+ "loss": 1.5205,
59
+ "step": 2500
60
+ },
61
+ {
62
+ "epoch": 2.25,
63
+ "learning_rate": 4.764828107291273e-05,
64
+ "loss": 1.1916,
65
+ "step": 3000
66
+ },
67
+ {
68
+ "epoch": 2.25,
69
+ "eval_bleu": 13.4571,
70
+ "eval_gen_len": 18.8854,
71
+ "eval_loss": 1.6842907667160034,
72
+ "eval_runtime": 538.3261,
73
+ "eval_samples_per_second": 4.962,
74
+ "eval_steps_per_second": 0.31,
75
+ "step": 3000
76
+ },
77
+ {
78
+ "epoch": 2.62,
79
+ "learning_rate": 4.7176048356630146e-05,
80
+ "loss": 1.0458,
81
+ "step": 3500
82
+ },
83
+ {
84
+ "epoch": 2.99,
85
+ "learning_rate": 4.670381564034756e-05,
86
+ "loss": 1.068,
87
+ "step": 4000
88
+ },
89
+ {
90
+ "epoch": 2.99,
91
+ "eval_bleu": 15.6862,
92
+ "eval_gen_len": 15.5054,
93
+ "eval_loss": 1.6389988660812378,
94
+ "eval_runtime": 200.5304,
95
+ "eval_samples_per_second": 13.32,
96
+ "eval_steps_per_second": 0.833,
97
+ "step": 4000
98
+ },
99
+ {
100
+ "epoch": 3.37,
101
+ "learning_rate": 4.623158292406498e-05,
102
+ "loss": 0.7097,
103
+ "step": 4500
104
+ },
105
+ {
106
+ "epoch": 3.74,
107
+ "learning_rate": 4.57593502077824e-05,
108
+ "loss": 0.7313,
109
+ "step": 5000
110
+ },
111
+ {
112
+ "epoch": 3.74,
113
+ "eval_bleu": 15.2014,
114
+ "eval_gen_len": 16.5938,
115
+ "eval_loss": 1.7002748250961304,
116
+ "eval_runtime": 222.4222,
117
+ "eval_samples_per_second": 12.009,
118
+ "eval_steps_per_second": 0.751,
119
+ "step": 5000
120
+ },
121
+ {
122
+ "epoch": 4.12,
123
+ "learning_rate": 4.5287117491499814e-05,
124
+ "loss": 0.6587,
125
+ "step": 5500
126
+ },
127
+ {
128
+ "epoch": 4.49,
129
+ "learning_rate": 4.481488477521723e-05,
130
+ "loss": 0.4832,
131
+ "step": 6000
132
+ },
133
+ {
134
+ "epoch": 4.49,
135
+ "eval_bleu": 15.0381,
136
+ "eval_gen_len": 16.9068,
137
+ "eval_loss": 1.8982346057891846,
138
+ "eval_runtime": 258.8867,
139
+ "eval_samples_per_second": 10.317,
140
+ "eval_steps_per_second": 0.645,
141
+ "step": 6000
142
+ },
143
+ {
144
+ "epoch": 4.87,
145
+ "learning_rate": 4.434265205893465e-05,
146
+ "loss": 0.5089,
147
+ "step": 6500
148
+ },
149
+ {
150
+ "epoch": 5.24,
151
+ "learning_rate": 4.3870419342652066e-05,
152
+ "loss": 0.3862,
153
+ "step": 7000
154
+ },
155
+ {
156
+ "epoch": 5.24,
157
+ "eval_bleu": 15.5397,
158
+ "eval_gen_len": 15.6451,
159
+ "eval_loss": 2.1426022052764893,
160
+ "eval_runtime": 170.9728,
161
+ "eval_samples_per_second": 15.622,
162
+ "eval_steps_per_second": 0.977,
163
+ "step": 7000
164
+ },
165
+ {
166
+ "epoch": 5.61,
167
+ "learning_rate": 4.339913109180204e-05,
168
+ "loss": 0.3495,
169
+ "step": 7500
170
+ },
171
+ {
172
+ "epoch": 5.99,
173
+ "learning_rate": 4.292689837551946e-05,
174
+ "loss": 0.3675,
175
+ "step": 8000
176
+ },
177
+ {
178
+ "epoch": 5.99,
179
+ "eval_bleu": 15.8847,
180
+ "eval_gen_len": 15.6926,
181
+ "eval_loss": 2.1168086528778076,
182
+ "eval_runtime": 169.7878,
183
+ "eval_samples_per_second": 15.731,
184
+ "eval_steps_per_second": 0.984,
185
+ "step": 8000
186
+ },
187
+ {
188
+ "epoch": 6.36,
189
+ "learning_rate": 4.2454665659236876e-05,
190
+ "loss": 0.2443,
191
+ "step": 8500
192
+ },
193
+ {
194
+ "epoch": 6.74,
195
+ "learning_rate": 4.198243294295429e-05,
196
+ "loss": 0.2627,
197
+ "step": 9000
198
+ },
199
+ {
200
+ "epoch": 6.74,
201
+ "eval_bleu": 16.3603,
202
+ "eval_gen_len": 15.9671,
203
+ "eval_loss": 2.260291814804077,
204
+ "eval_runtime": 176.4048,
205
+ "eval_samples_per_second": 15.141,
206
+ "eval_steps_per_second": 0.947,
207
+ "step": 9000
208
+ },
209
+ {
210
+ "epoch": 7.11,
211
+ "learning_rate": 4.151020022667171e-05,
212
+ "loss": 0.2471,
213
+ "step": 9500
214
+ },
215
+ {
216
+ "epoch": 7.49,
217
+ "learning_rate": 4.103796751038912e-05,
218
+ "loss": 0.1955,
219
+ "step": 10000
220
+ },
221
+ {
222
+ "epoch": 7.49,
223
+ "eval_bleu": 15.7447,
224
+ "eval_gen_len": 15.979,
225
+ "eval_loss": 2.4113738536834717,
226
+ "eval_runtime": 200.4127,
227
+ "eval_samples_per_second": 13.327,
228
+ "eval_steps_per_second": 0.833,
229
+ "step": 10000
230
+ },
231
+ {
232
+ "epoch": 7.86,
233
+ "learning_rate": 4.056573479410654e-05,
234
+ "loss": 0.2088,
235
+ "step": 10500
236
+ },
237
+ {
238
+ "epoch": 8.23,
239
+ "learning_rate": 4.0093502077823954e-05,
240
+ "loss": 0.171,
241
+ "step": 11000
242
+ },
243
+ {
244
+ "epoch": 8.23,
245
+ "eval_bleu": 15.7852,
246
+ "eval_gen_len": 15.9244,
247
+ "eval_loss": 2.5141146183013916,
248
+ "eval_runtime": 180.5467,
249
+ "eval_samples_per_second": 14.794,
250
+ "eval_steps_per_second": 0.925,
251
+ "step": 11000
252
+ },
253
+ {
254
+ "epoch": 8.61,
255
+ "learning_rate": 3.962126936154137e-05,
256
+ "loss": 0.157,
257
+ "step": 11500
258
+ },
259
+ {
260
+ "epoch": 8.98,
261
+ "learning_rate": 3.9149981110691354e-05,
262
+ "loss": 0.1702,
263
+ "step": 12000
264
+ },
265
+ {
266
+ "epoch": 8.98,
267
+ "eval_bleu": 16.6036,
268
+ "eval_gen_len": 15.4901,
269
+ "eval_loss": 2.5557188987731934,
270
+ "eval_runtime": 168.9011,
271
+ "eval_samples_per_second": 15.814,
272
+ "eval_steps_per_second": 0.989,
273
+ "step": 12000
274
+ },
275
+ {
276
+ "epoch": 9.36,
277
+ "learning_rate": 3.8677748394408764e-05,
278
+ "loss": 0.1206,
279
+ "step": 12500
280
+ },
281
+ {
282
+ "epoch": 9.73,
283
+ "learning_rate": 3.820551567812618e-05,
284
+ "loss": 0.1298,
285
+ "step": 13000
286
+ },
287
+ {
288
+ "epoch": 9.73,
289
+ "eval_bleu": 16.1319,
290
+ "eval_gen_len": 15.5492,
291
+ "eval_loss": 2.653599739074707,
292
+ "eval_runtime": 168.2024,
293
+ "eval_samples_per_second": 15.88,
294
+ "eval_steps_per_second": 0.993,
295
+ "step": 13000
296
+ },
297
+ {
298
+ "epoch": 10.1,
299
+ "learning_rate": 3.77332829618436e-05,
300
+ "loss": 0.1263,
301
+ "step": 13500
302
+ },
303
+ {
304
+ "epoch": 10.48,
305
+ "learning_rate": 3.7261050245561016e-05,
306
+ "loss": 0.1052,
307
+ "step": 14000
308
+ },
309
+ {
310
+ "epoch": 10.48,
311
+ "eval_bleu": 16.1807,
312
+ "eval_gen_len": 15.8884,
313
+ "eval_loss": 2.7585606575012207,
314
+ "eval_runtime": 171.7593,
315
+ "eval_samples_per_second": 15.551,
316
+ "eval_steps_per_second": 0.972,
317
+ "step": 14000
318
+ },
319
+ {
320
+ "epoch": 10.85,
321
+ "learning_rate": 3.678881752927843e-05,
322
+ "loss": 0.1137,
323
+ "step": 14500
324
+ },
325
+ {
326
+ "epoch": 11.23,
327
+ "learning_rate": 3.6320362674726105e-05,
328
+ "loss": 0.2268,
329
+ "step": 15000
330
+ },
331
+ {
332
+ "epoch": 11.23,
333
+ "eval_bleu": 15.1752,
334
+ "eval_gen_len": 15.5346,
335
+ "eval_loss": 2.7257988452911377,
336
+ "eval_runtime": 171.5928,
337
+ "eval_samples_per_second": 15.566,
338
+ "eval_steps_per_second": 0.973,
339
+ "step": 15000
340
+ },
341
+ {
342
+ "epoch": 11.6,
343
+ "learning_rate": 3.584812995844352e-05,
344
+ "loss": 0.1133,
345
+ "step": 15500
346
+ },
347
+ {
348
+ "epoch": 11.98,
349
+ "learning_rate": 3.5376841707593505e-05,
350
+ "loss": 0.1327,
351
+ "step": 16000
352
+ },
353
+ {
354
+ "epoch": 11.98,
355
+ "eval_bleu": 15.8563,
356
+ "eval_gen_len": 15.7971,
357
+ "eval_loss": 2.719254493713379,
358
+ "eval_runtime": 171.7089,
359
+ "eval_samples_per_second": 15.555,
360
+ "eval_steps_per_second": 0.973,
361
+ "step": 16000
362
+ }
363
+ ],
364
+ "logging_steps": 500,
365
+ "max_steps": 53440,
366
+ "num_train_epochs": 40,
367
+ "save_steps": 1000,
368
+ "total_flos": 5.54785220591616e+17,
369
+ "trial_name": null,
370
+ "trial_params": null
371
+ }
checkpoint-16000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f8ee59bef192bac6b7c7efff93ae004410d1f65031577a07e5f97f4a15294d8
3
+ size 4664
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-cc25",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
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": 250014,
21
+ "dropout": 0.1,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 4096,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 12,
26
+ "eos_token_id": 2,
27
+ "forced_eos_token_id": 2,
28
+ "id2label": {
29
+ "0": "LABEL_0",
30
+ "1": "LABEL_1",
31
+ "2": "LABEL_2"
32
+ },
33
+ "init_std": 0.02,
34
+ "is_encoder_decoder": true,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1,
38
+ "LABEL_2": 2
39
+ },
40
+ "max_length": 1024,
41
+ "max_position_embeddings": 1024,
42
+ "model_type": "mbart",
43
+ "normalize_before": true,
44
+ "normalize_embedding": true,
45
+ "num_beams": 5,
46
+ "num_hidden_layers": 12,
47
+ "output_past": true,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": true,
50
+ "static_position_embeddings": false,
51
+ "task_specific_params": {
52
+ "translation_en_to_ro": {
53
+ "decoder_start_token_id": 250020
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.34.0",
58
+ "use_cache": true,
59
+ "vocab_size": 250027
60
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 11.98,
3
+ "eval_bleu": 16.6036,
4
+ "eval_gen_len": 15.4901,
5
+ "eval_loss": 2.5557188987731934,
6
+ "eval_runtime": 167.3814,
7
+ "eval_samples": 2671,
8
+ "eval_samples_per_second": 15.958,
9
+ "eval_steps_per_second": 0.998
10
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 250014,
4
+ "eos_token_id": 2,
5
+ "forced_eos_token_id": 2,
6
+ "max_length": 1024,
7
+ "num_beams": 5,
8
+ "pad_token_id": 1,
9
+ "transformers_version": "4.34.0"
10
+ }
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 15.2894,
3
+ "predict_gen_len": 15.6769,
4
+ "predict_loss": 2.6123738288879395,
5
+ "predict_runtime": 173.2964,
6
+ "predict_samples": 2671,
7
+ "predict_samples_per_second": 15.413,
8
+ "predict_steps_per_second": 0.964
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57bc3ec79d14b77e794d98ec08abca470f52cb6ecfdb0ded73411deff2bdea21
3
+ size 2444583770
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 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN"
28
+ ],
29
+ "bos_token": "<s>",
30
+ "cls_token": "<s>",
31
+ "eos_token": "</s>",
32
+ "mask_token": "<mask>",
33
+ "pad_token": "<pad>",
34
+ "sep_token": "</s>",
35
+ "unk_token": "<unk>"
36
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94bf3391e3589e53c8b8888fcbe2a854ebe851645c71bf46b7af9335fadf0db5
3
+ size 17088468
tokenizer_config.json ADDED
@@ -0,0 +1,283 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "<mask>",
237
+ "lstrip": true,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ }
243
+ },
244
+ "additional_special_tokens": [
245
+ "ar_AR",
246
+ "cs_CZ",
247
+ "de_DE",
248
+ "en_XX",
249
+ "es_XX",
250
+ "et_EE",
251
+ "fi_FI",
252
+ "fr_XX",
253
+ "gu_IN",
254
+ "hi_IN",
255
+ "it_IT",
256
+ "ja_XX",
257
+ "kk_KZ",
258
+ "ko_KR",
259
+ "lt_LT",
260
+ "lv_LV",
261
+ "my_MM",
262
+ "ne_NP",
263
+ "nl_XX",
264
+ "ro_RO",
265
+ "ru_RU",
266
+ "si_LK",
267
+ "tr_TR",
268
+ "vi_VN",
269
+ "zh_CN"
270
+ ],
271
+ "bos_token": "<s>",
272
+ "clean_up_tokenization_spaces": true,
273
+ "cls_token": "<s>",
274
+ "eos_token": "</s>",
275
+ "mask_token": "<mask>",
276
+ "model_max_length": 1024,
277
+ "pad_token": "<pad>",
278
+ "sep_token": "</s>",
279
+ "src_lang": "zh_CN",
280
+ "tgt_lang": "ko_KR",
281
+ "tokenizer_class": "MBartTokenizer",
282
+ "unk_token": "<unk>"
283
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 11.98,
3
+ "train_loss": 0.6965520789623261,
4
+ "train_runtime": 15742.7956,
5
+ "train_samples": 21375,
6
+ "train_samples_per_second": 54.311,
7
+ "train_steps_per_second": 3.395
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,380 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 16.6036,
3
+ "best_model_checkpoint": "./zh-kr_mid/checkpoint-12000",
4
+ "epoch": 11.976047904191617,
5
+ "eval_steps": 1000,
6
+ "global_step": 16000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.37,
13
+ "learning_rate": 4.91e-05,
14
+ "loss": 4.4024,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.75,
19
+ "learning_rate": 4.95362674726105e-05,
20
+ "loss": 2.7248,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.75,
25
+ "eval_bleu": 3.2381,
26
+ "eval_gen_len": 48.6095,
27
+ "eval_loss": 1.9409534931182861,
28
+ "eval_runtime": 1917.4714,
29
+ "eval_samples_per_second": 1.393,
30
+ "eval_steps_per_second": 0.087,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 1.12,
35
+ "learning_rate": 4.9064979221760484e-05,
36
+ "loss": 1.8483,
37
+ "step": 1500
38
+ },
39
+ {
40
+ "epoch": 1.5,
41
+ "learning_rate": 4.85927465054779e-05,
42
+ "loss": 1.5683,
43
+ "step": 2000
44
+ },
45
+ {
46
+ "epoch": 1.5,
47
+ "eval_bleu": 10.2345,
48
+ "eval_gen_len": 20.4433,
49
+ "eval_loss": 1.6889078617095947,
50
+ "eval_runtime": 497.5454,
51
+ "eval_samples_per_second": 5.368,
52
+ "eval_steps_per_second": 0.336,
53
+ "step": 2000
54
+ },
55
+ {
56
+ "epoch": 1.87,
57
+ "learning_rate": 4.812051378919532e-05,
58
+ "loss": 1.5205,
59
+ "step": 2500
60
+ },
61
+ {
62
+ "epoch": 2.25,
63
+ "learning_rate": 4.764828107291273e-05,
64
+ "loss": 1.1916,
65
+ "step": 3000
66
+ },
67
+ {
68
+ "epoch": 2.25,
69
+ "eval_bleu": 13.4571,
70
+ "eval_gen_len": 18.8854,
71
+ "eval_loss": 1.6842907667160034,
72
+ "eval_runtime": 538.3261,
73
+ "eval_samples_per_second": 4.962,
74
+ "eval_steps_per_second": 0.31,
75
+ "step": 3000
76
+ },
77
+ {
78
+ "epoch": 2.62,
79
+ "learning_rate": 4.7176048356630146e-05,
80
+ "loss": 1.0458,
81
+ "step": 3500
82
+ },
83
+ {
84
+ "epoch": 2.99,
85
+ "learning_rate": 4.670381564034756e-05,
86
+ "loss": 1.068,
87
+ "step": 4000
88
+ },
89
+ {
90
+ "epoch": 2.99,
91
+ "eval_bleu": 15.6862,
92
+ "eval_gen_len": 15.5054,
93
+ "eval_loss": 1.6389988660812378,
94
+ "eval_runtime": 200.5304,
95
+ "eval_samples_per_second": 13.32,
96
+ "eval_steps_per_second": 0.833,
97
+ "step": 4000
98
+ },
99
+ {
100
+ "epoch": 3.37,
101
+ "learning_rate": 4.623158292406498e-05,
102
+ "loss": 0.7097,
103
+ "step": 4500
104
+ },
105
+ {
106
+ "epoch": 3.74,
107
+ "learning_rate": 4.57593502077824e-05,
108
+ "loss": 0.7313,
109
+ "step": 5000
110
+ },
111
+ {
112
+ "epoch": 3.74,
113
+ "eval_bleu": 15.2014,
114
+ "eval_gen_len": 16.5938,
115
+ "eval_loss": 1.7002748250961304,
116
+ "eval_runtime": 222.4222,
117
+ "eval_samples_per_second": 12.009,
118
+ "eval_steps_per_second": 0.751,
119
+ "step": 5000
120
+ },
121
+ {
122
+ "epoch": 4.12,
123
+ "learning_rate": 4.5287117491499814e-05,
124
+ "loss": 0.6587,
125
+ "step": 5500
126
+ },
127
+ {
128
+ "epoch": 4.49,
129
+ "learning_rate": 4.481488477521723e-05,
130
+ "loss": 0.4832,
131
+ "step": 6000
132
+ },
133
+ {
134
+ "epoch": 4.49,
135
+ "eval_bleu": 15.0381,
136
+ "eval_gen_len": 16.9068,
137
+ "eval_loss": 1.8982346057891846,
138
+ "eval_runtime": 258.8867,
139
+ "eval_samples_per_second": 10.317,
140
+ "eval_steps_per_second": 0.645,
141
+ "step": 6000
142
+ },
143
+ {
144
+ "epoch": 4.87,
145
+ "learning_rate": 4.434265205893465e-05,
146
+ "loss": 0.5089,
147
+ "step": 6500
148
+ },
149
+ {
150
+ "epoch": 5.24,
151
+ "learning_rate": 4.3870419342652066e-05,
152
+ "loss": 0.3862,
153
+ "step": 7000
154
+ },
155
+ {
156
+ "epoch": 5.24,
157
+ "eval_bleu": 15.5397,
158
+ "eval_gen_len": 15.6451,
159
+ "eval_loss": 2.1426022052764893,
160
+ "eval_runtime": 170.9728,
161
+ "eval_samples_per_second": 15.622,
162
+ "eval_steps_per_second": 0.977,
163
+ "step": 7000
164
+ },
165
+ {
166
+ "epoch": 5.61,
167
+ "learning_rate": 4.339913109180204e-05,
168
+ "loss": 0.3495,
169
+ "step": 7500
170
+ },
171
+ {
172
+ "epoch": 5.99,
173
+ "learning_rate": 4.292689837551946e-05,
174
+ "loss": 0.3675,
175
+ "step": 8000
176
+ },
177
+ {
178
+ "epoch": 5.99,
179
+ "eval_bleu": 15.8847,
180
+ "eval_gen_len": 15.6926,
181
+ "eval_loss": 2.1168086528778076,
182
+ "eval_runtime": 169.7878,
183
+ "eval_samples_per_second": 15.731,
184
+ "eval_steps_per_second": 0.984,
185
+ "step": 8000
186
+ },
187
+ {
188
+ "epoch": 6.36,
189
+ "learning_rate": 4.2454665659236876e-05,
190
+ "loss": 0.2443,
191
+ "step": 8500
192
+ },
193
+ {
194
+ "epoch": 6.74,
195
+ "learning_rate": 4.198243294295429e-05,
196
+ "loss": 0.2627,
197
+ "step": 9000
198
+ },
199
+ {
200
+ "epoch": 6.74,
201
+ "eval_bleu": 16.3603,
202
+ "eval_gen_len": 15.9671,
203
+ "eval_loss": 2.260291814804077,
204
+ "eval_runtime": 176.4048,
205
+ "eval_samples_per_second": 15.141,
206
+ "eval_steps_per_second": 0.947,
207
+ "step": 9000
208
+ },
209
+ {
210
+ "epoch": 7.11,
211
+ "learning_rate": 4.151020022667171e-05,
212
+ "loss": 0.2471,
213
+ "step": 9500
214
+ },
215
+ {
216
+ "epoch": 7.49,
217
+ "learning_rate": 4.103796751038912e-05,
218
+ "loss": 0.1955,
219
+ "step": 10000
220
+ },
221
+ {
222
+ "epoch": 7.49,
223
+ "eval_bleu": 15.7447,
224
+ "eval_gen_len": 15.979,
225
+ "eval_loss": 2.4113738536834717,
226
+ "eval_runtime": 200.4127,
227
+ "eval_samples_per_second": 13.327,
228
+ "eval_steps_per_second": 0.833,
229
+ "step": 10000
230
+ },
231
+ {
232
+ "epoch": 7.86,
233
+ "learning_rate": 4.056573479410654e-05,
234
+ "loss": 0.2088,
235
+ "step": 10500
236
+ },
237
+ {
238
+ "epoch": 8.23,
239
+ "learning_rate": 4.0093502077823954e-05,
240
+ "loss": 0.171,
241
+ "step": 11000
242
+ },
243
+ {
244
+ "epoch": 8.23,
245
+ "eval_bleu": 15.7852,
246
+ "eval_gen_len": 15.9244,
247
+ "eval_loss": 2.5141146183013916,
248
+ "eval_runtime": 180.5467,
249
+ "eval_samples_per_second": 14.794,
250
+ "eval_steps_per_second": 0.925,
251
+ "step": 11000
252
+ },
253
+ {
254
+ "epoch": 8.61,
255
+ "learning_rate": 3.962126936154137e-05,
256
+ "loss": 0.157,
257
+ "step": 11500
258
+ },
259
+ {
260
+ "epoch": 8.98,
261
+ "learning_rate": 3.9149981110691354e-05,
262
+ "loss": 0.1702,
263
+ "step": 12000
264
+ },
265
+ {
266
+ "epoch": 8.98,
267
+ "eval_bleu": 16.6036,
268
+ "eval_gen_len": 15.4901,
269
+ "eval_loss": 2.5557188987731934,
270
+ "eval_runtime": 168.9011,
271
+ "eval_samples_per_second": 15.814,
272
+ "eval_steps_per_second": 0.989,
273
+ "step": 12000
274
+ },
275
+ {
276
+ "epoch": 9.36,
277
+ "learning_rate": 3.8677748394408764e-05,
278
+ "loss": 0.1206,
279
+ "step": 12500
280
+ },
281
+ {
282
+ "epoch": 9.73,
283
+ "learning_rate": 3.820551567812618e-05,
284
+ "loss": 0.1298,
285
+ "step": 13000
286
+ },
287
+ {
288
+ "epoch": 9.73,
289
+ "eval_bleu": 16.1319,
290
+ "eval_gen_len": 15.5492,
291
+ "eval_loss": 2.653599739074707,
292
+ "eval_runtime": 168.2024,
293
+ "eval_samples_per_second": 15.88,
294
+ "eval_steps_per_second": 0.993,
295
+ "step": 13000
296
+ },
297
+ {
298
+ "epoch": 10.1,
299
+ "learning_rate": 3.77332829618436e-05,
300
+ "loss": 0.1263,
301
+ "step": 13500
302
+ },
303
+ {
304
+ "epoch": 10.48,
305
+ "learning_rate": 3.7261050245561016e-05,
306
+ "loss": 0.1052,
307
+ "step": 14000
308
+ },
309
+ {
310
+ "epoch": 10.48,
311
+ "eval_bleu": 16.1807,
312
+ "eval_gen_len": 15.8884,
313
+ "eval_loss": 2.7585606575012207,
314
+ "eval_runtime": 171.7593,
315
+ "eval_samples_per_second": 15.551,
316
+ "eval_steps_per_second": 0.972,
317
+ "step": 14000
318
+ },
319
+ {
320
+ "epoch": 10.85,
321
+ "learning_rate": 3.678881752927843e-05,
322
+ "loss": 0.1137,
323
+ "step": 14500
324
+ },
325
+ {
326
+ "epoch": 11.23,
327
+ "learning_rate": 3.6320362674726105e-05,
328
+ "loss": 0.2268,
329
+ "step": 15000
330
+ },
331
+ {
332
+ "epoch": 11.23,
333
+ "eval_bleu": 15.1752,
334
+ "eval_gen_len": 15.5346,
335
+ "eval_loss": 2.7257988452911377,
336
+ "eval_runtime": 171.5928,
337
+ "eval_samples_per_second": 15.566,
338
+ "eval_steps_per_second": 0.973,
339
+ "step": 15000
340
+ },
341
+ {
342
+ "epoch": 11.6,
343
+ "learning_rate": 3.584812995844352e-05,
344
+ "loss": 0.1133,
345
+ "step": 15500
346
+ },
347
+ {
348
+ "epoch": 11.98,
349
+ "learning_rate": 3.5376841707593505e-05,
350
+ "loss": 0.1327,
351
+ "step": 16000
352
+ },
353
+ {
354
+ "epoch": 11.98,
355
+ "eval_bleu": 15.8563,
356
+ "eval_gen_len": 15.7971,
357
+ "eval_loss": 2.719254493713379,
358
+ "eval_runtime": 171.7089,
359
+ "eval_samples_per_second": 15.555,
360
+ "eval_steps_per_second": 0.973,
361
+ "step": 16000
362
+ },
363
+ {
364
+ "epoch": 11.98,
365
+ "step": 16000,
366
+ "total_flos": 5.54785220591616e+17,
367
+ "train_loss": 0.6965520789623261,
368
+ "train_runtime": 15742.7956,
369
+ "train_samples_per_second": 54.311,
370
+ "train_steps_per_second": 3.395
371
+ }
372
+ ],
373
+ "logging_steps": 500,
374
+ "max_steps": 53440,
375
+ "num_train_epochs": 40,
376
+ "save_steps": 1000,
377
+ "total_flos": 5.54785220591616e+17,
378
+ "trial_name": null,
379
+ "trial_params": null
380
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f8ee59bef192bac6b7c7efff93ae004410d1f65031577a07e5f97f4a15294d8
3
+ size 4664