cvapict commited on
Commit
9edb9ff
·
verified ·
1 Parent(s): ba9c40c

End of training

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +70 -0
  2. config.json +26 -0
  3. model.safetensors +3 -0
  4. run-0/checkpoint-2616/config.json +26 -0
  5. run-0/checkpoint-2616/model.safetensors +3 -0
  6. run-0/checkpoint-2616/optimizer.pt +3 -0
  7. run-0/checkpoint-2616/rng_state.pth +3 -0
  8. run-0/checkpoint-2616/scheduler.pt +3 -0
  9. run-0/checkpoint-2616/special_tokens_map.json +7 -0
  10. run-0/checkpoint-2616/tokenizer.json +0 -0
  11. run-0/checkpoint-2616/tokenizer_config.json +55 -0
  12. run-0/checkpoint-2616/trainer_state.json +1913 -0
  13. run-0/checkpoint-2616/training_args.bin +3 -0
  14. run-0/checkpoint-2616/vocab.txt +0 -0
  15. run-1/checkpoint-654/config.json +26 -0
  16. run-1/checkpoint-654/model.safetensors +3 -0
  17. run-1/checkpoint-654/optimizer.pt +3 -0
  18. run-1/checkpoint-654/rng_state.pth +3 -0
  19. run-1/checkpoint-654/scheduler.pt +3 -0
  20. run-1/checkpoint-654/special_tokens_map.json +7 -0
  21. run-1/checkpoint-654/tokenizer.json +0 -0
  22. run-1/checkpoint-654/tokenizer_config.json +55 -0
  23. run-1/checkpoint-654/trainer_state.json +505 -0
  24. run-1/checkpoint-654/training_args.bin +3 -0
  25. run-1/checkpoint-654/vocab.txt +0 -0
  26. run-2/checkpoint-492/config.json +26 -0
  27. run-2/checkpoint-492/model.safetensors +3 -0
  28. run-2/checkpoint-492/optimizer.pt +3 -0
  29. run-2/checkpoint-492/rng_state.pth +3 -0
  30. run-2/checkpoint-492/scheduler.pt +3 -0
  31. run-2/checkpoint-492/special_tokens_map.json +7 -0
  32. run-2/checkpoint-492/tokenizer.json +0 -0
  33. run-2/checkpoint-492/tokenizer_config.json +55 -0
  34. run-2/checkpoint-492/trainer_state.json +417 -0
  35. run-2/checkpoint-492/training_args.bin +3 -0
  36. run-2/checkpoint-492/vocab.txt +0 -0
  37. run-3/checkpoint-328/config.json +26 -0
  38. run-3/checkpoint-328/model.safetensors +3 -0
  39. run-3/checkpoint-328/optimizer.pt +3 -0
  40. run-3/checkpoint-328/rng_state.pth +3 -0
  41. run-3/checkpoint-328/scheduler.pt +3 -0
  42. run-3/checkpoint-328/special_tokens_map.json +7 -0
  43. run-3/checkpoint-328/tokenizer.json +0 -0
  44. run-3/checkpoint-328/tokenizer_config.json +55 -0
  45. run-3/checkpoint-328/trainer_state.json +310 -0
  46. run-3/checkpoint-328/training_args.bin +3 -0
  47. run-3/checkpoint-328/vocab.txt +0 -0
  48. run-4/checkpoint-1308/config.json +26 -0
  49. run-4/checkpoint-1308/model.safetensors +3 -0
  50. run-4/checkpoint-1308/optimizer.pt +3 -0
README.md ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: distilbert-base-multilingual-cased
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ - recall
10
+ - precision
11
+ - f1
12
+ model-index:
13
+ - name: distilbert-base-multilingual-cased-aoe-hyper
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ # distilbert-base-multilingual-cased-aoe-hyper
21
+
22
+ This model is a fine-tuned version of [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased) on the None dataset.
23
+ It achieves the following results on the evaluation set:
24
+ - Loss: 0.4021
25
+ - Accuracy: 0.8799
26
+ - Recall: 0.8013
27
+ - Precision: 0.7246
28
+ - F1: 0.7610
29
+
30
+ ## Model description
31
+
32
+ More information needed
33
+
34
+ ## Intended uses & limitations
35
+
36
+ More information needed
37
+
38
+ ## Training and evaluation data
39
+
40
+ More information needed
41
+
42
+ ## Training procedure
43
+
44
+ ### Training hyperparameters
45
+
46
+ The following hyperparameters were used during training:
47
+ - learning_rate: 6.292247538797816e-06
48
+ - train_batch_size: 8
49
+ - eval_batch_size: 16
50
+ - seed: 16
51
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
+ - lr_scheduler_type: linear
53
+ - num_epochs: 4
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Recall | Precision | F1 |
58
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
59
+ | 0.2696 | 1.0 | 654 | 0.3423 | 0.8699 | 0.6090 | 0.7983 | 0.6909 |
60
+ | 0.4222 | 2.0 | 1308 | 0.3537 | 0.8814 | 0.7596 | 0.7476 | 0.7536 |
61
+ | 0.3135 | 3.0 | 1962 | 0.3992 | 0.8699 | 0.8205 | 0.6919 | 0.7507 |
62
+ | 0.2558 | 4.0 | 2616 | 0.4021 | 0.8799 | 0.8013 | 0.7246 | 0.7610 |
63
+
64
+
65
+ ### Framework versions
66
+
67
+ - Transformers 4.46.2
68
+ - Pytorch 2.5.1+cu121
69
+ - Datasets 3.1.0
70
+ - Tokenizers 0.20.3
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3f013b5e59583780ee83d863dc493644833237543dbc19d7d4b2efccf3c015b
3
+ size 541317368
run-0/checkpoint-2616/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
run-0/checkpoint-2616/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc20b909f12366cfaed536bd0051462d09f7d7ba971a1115a72a4830846b2c3
3
+ size 541317368
run-0/checkpoint-2616/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a958f1fb7a130e7304d810babbd1dfea988ff5ad807fba1ed7722fff09fcaaa2
3
+ size 1082696890
run-0/checkpoint-2616/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5720cc9c6b35d5b26684c1cc949c8b6e8ce7e9c35117a034212aa1075dd7544
3
+ size 14244
run-0/checkpoint-2616/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72769ff360c7834055f7cea1cdd36cd37c87f9692499d5e60332b743c1e9765f
3
+ size 1064
run-0/checkpoint-2616/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-0/checkpoint-2616/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-0/checkpoint-2616/tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": false,
47
+ "mask_token": "[MASK]",
48
+ "model_max_length": 512,
49
+ "pad_token": "[PAD]",
50
+ "sep_token": "[SEP]",
51
+ "strip_accents": null,
52
+ "tokenize_chinese_chars": true,
53
+ "tokenizer_class": "DistilBertTokenizer",
54
+ "unk_token": "[UNK]"
55
+ }
run-0/checkpoint-2616/trainer_state.json ADDED
@@ -0,0 +1,1913 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8859984697781178,
3
+ "best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-0/checkpoint-2616",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2616,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01529051987767584,
13
+ "grad_norm": 1.6358972787857056,
14
+ "learning_rate": 9.349717786759033e-06,
15
+ "loss": 0.6796,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03058103975535168,
20
+ "grad_norm": 1.7600407600402832,
21
+ "learning_rate": 9.321037670848728e-06,
22
+ "loss": 0.5945,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.045871559633027525,
27
+ "grad_norm": 2.3243937492370605,
28
+ "learning_rate": 9.292357554938425e-06,
29
+ "loss": 0.5386,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.06116207951070336,
34
+ "grad_norm": 2.1618645191192627,
35
+ "learning_rate": 9.263677439028121e-06,
36
+ "loss": 0.4348,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.0764525993883792,
41
+ "grad_norm": 4.011763572692871,
42
+ "learning_rate": 9.234997323117817e-06,
43
+ "loss": 0.4917,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.09174311926605505,
48
+ "grad_norm": 3.0631511211395264,
49
+ "learning_rate": 9.206317207207513e-06,
50
+ "loss": 0.4419,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.10703363914373089,
55
+ "grad_norm": 5.19169807434082,
56
+ "learning_rate": 9.177637091297209e-06,
57
+ "loss": 0.4885,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.12232415902140673,
62
+ "grad_norm": 3.2671306133270264,
63
+ "learning_rate": 9.148956975386905e-06,
64
+ "loss": 0.3778,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.13761467889908258,
69
+ "grad_norm": 2.0801641941070557,
70
+ "learning_rate": 9.120276859476602e-06,
71
+ "loss": 0.4524,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1529051987767584,
76
+ "grad_norm": 3.2077746391296387,
77
+ "learning_rate": 9.091596743566297e-06,
78
+ "loss": 0.3107,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.16819571865443425,
83
+ "grad_norm": 3.8301804065704346,
84
+ "learning_rate": 9.062916627655994e-06,
85
+ "loss": 0.4904,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.1834862385321101,
90
+ "grad_norm": 2.995657205581665,
91
+ "learning_rate": 9.03423651174569e-06,
92
+ "loss": 0.3564,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.19877675840978593,
97
+ "grad_norm": 5.345709800720215,
98
+ "learning_rate": 9.005556395835386e-06,
99
+ "loss": 0.498,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.21406727828746178,
104
+ "grad_norm": 1.6241793632507324,
105
+ "learning_rate": 8.976876279925083e-06,
106
+ "loss": 0.4324,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.22935779816513763,
111
+ "grad_norm": 6.096868991851807,
112
+ "learning_rate": 8.94819616401478e-06,
113
+ "loss": 0.4474,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.24464831804281345,
118
+ "grad_norm": 4.777408123016357,
119
+ "learning_rate": 8.919516048104475e-06,
120
+ "loss": 0.3605,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.2599388379204893,
125
+ "grad_norm": 4.883125305175781,
126
+ "learning_rate": 8.890835932194172e-06,
127
+ "loss": 0.3694,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.27522935779816515,
132
+ "grad_norm": 2.1104636192321777,
133
+ "learning_rate": 8.862155816283867e-06,
134
+ "loss": 0.3391,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.290519877675841,
139
+ "grad_norm": 3.5193943977355957,
140
+ "learning_rate": 8.833475700373564e-06,
141
+ "loss": 0.4028,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.3058103975535168,
146
+ "grad_norm": 12.695348739624023,
147
+ "learning_rate": 8.80479558446326e-06,
148
+ "loss": 0.5372,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.3211009174311927,
153
+ "grad_norm": 8.019516944885254,
154
+ "learning_rate": 8.776115468552956e-06,
155
+ "loss": 0.3839,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.3363914373088685,
160
+ "grad_norm": 2.723048448562622,
161
+ "learning_rate": 8.747435352642653e-06,
162
+ "loss": 0.3155,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.3516819571865443,
167
+ "grad_norm": 5.75403356552124,
168
+ "learning_rate": 8.718755236732348e-06,
169
+ "loss": 0.4086,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.3669724770642202,
174
+ "grad_norm": 3.325057029724121,
175
+ "learning_rate": 8.690075120822045e-06,
176
+ "loss": 0.293,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.382262996941896,
181
+ "grad_norm": 18.065032958984375,
182
+ "learning_rate": 8.66139500491174e-06,
183
+ "loss": 0.3126,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.39755351681957185,
188
+ "grad_norm": 5.72575569152832,
189
+ "learning_rate": 8.632714889001437e-06,
190
+ "loss": 0.3087,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.41284403669724773,
195
+ "grad_norm": 5.2425055503845215,
196
+ "learning_rate": 8.604034773091134e-06,
197
+ "loss": 0.474,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.42813455657492355,
202
+ "grad_norm": 8.025136947631836,
203
+ "learning_rate": 8.57535465718083e-06,
204
+ "loss": 0.2461,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.4434250764525994,
209
+ "grad_norm": 11.76362133026123,
210
+ "learning_rate": 8.546674541270526e-06,
211
+ "loss": 0.3092,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.45871559633027525,
216
+ "grad_norm": 14.632379531860352,
217
+ "learning_rate": 8.517994425360223e-06,
218
+ "loss": 0.4511,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.4740061162079511,
223
+ "grad_norm": 3.7113358974456787,
224
+ "learning_rate": 8.489314309449918e-06,
225
+ "loss": 0.3188,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.4892966360856269,
230
+ "grad_norm": 10.738109588623047,
231
+ "learning_rate": 8.460634193539615e-06,
232
+ "loss": 0.4563,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5045871559633027,
237
+ "grad_norm": 13.628973960876465,
238
+ "learning_rate": 8.431954077629312e-06,
239
+ "loss": 0.3233,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.5198776758409785,
244
+ "grad_norm": 17.098241806030273,
245
+ "learning_rate": 8.403273961719007e-06,
246
+ "loss": 0.4134,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.5351681957186545,
251
+ "grad_norm": 4.145214080810547,
252
+ "learning_rate": 8.374593845808704e-06,
253
+ "loss": 0.3638,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.5504587155963303,
258
+ "grad_norm": 21.35369300842285,
259
+ "learning_rate": 8.345913729898399e-06,
260
+ "loss": 0.3341,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.5657492354740061,
265
+ "grad_norm": 6.4279937744140625,
266
+ "learning_rate": 8.317233613988096e-06,
267
+ "loss": 0.3332,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.581039755351682,
272
+ "grad_norm": 6.384445667266846,
273
+ "learning_rate": 8.288553498077791e-06,
274
+ "loss": 0.3299,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.5963302752293578,
279
+ "grad_norm": 9.927398681640625,
280
+ "learning_rate": 8.25987338216749e-06,
281
+ "loss": 0.3687,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.6116207951070336,
286
+ "grad_norm": 2.317152261734009,
287
+ "learning_rate": 8.231193266257185e-06,
288
+ "loss": 0.2947,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.6269113149847095,
293
+ "grad_norm": 6.1938982009887695,
294
+ "learning_rate": 8.202513150346881e-06,
295
+ "loss": 0.2495,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.6422018348623854,
300
+ "grad_norm": 9.2775297164917,
301
+ "learning_rate": 8.173833034436577e-06,
302
+ "loss": 0.3011,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.6574923547400612,
307
+ "grad_norm": 12.099929809570312,
308
+ "learning_rate": 8.145152918526273e-06,
309
+ "loss": 0.2866,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.672782874617737,
314
+ "grad_norm": 15.571517944335938,
315
+ "learning_rate": 8.116472802615969e-06,
316
+ "loss": 0.2352,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.6880733944954128,
321
+ "grad_norm": 2.7042486667633057,
322
+ "learning_rate": 8.087792686705666e-06,
323
+ "loss": 0.4442,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.7033639143730887,
328
+ "grad_norm": 24.800390243530273,
329
+ "learning_rate": 8.059112570795362e-06,
330
+ "loss": 0.2647,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.7186544342507645,
335
+ "grad_norm": 1.1386603116989136,
336
+ "learning_rate": 8.030432454885058e-06,
337
+ "loss": 0.339,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.7339449541284404,
342
+ "grad_norm": 2.4210026264190674,
343
+ "learning_rate": 8.001752338974754e-06,
344
+ "loss": 0.2963,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.7492354740061162,
349
+ "grad_norm": 10.266980171203613,
350
+ "learning_rate": 7.97307222306445e-06,
351
+ "loss": 0.2177,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.764525993883792,
356
+ "grad_norm": 8.048089027404785,
357
+ "learning_rate": 7.944392107154146e-06,
358
+ "loss": 0.3248,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.7798165137614679,
363
+ "grad_norm": 25.07794189453125,
364
+ "learning_rate": 7.915711991243843e-06,
365
+ "loss": 0.3321,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.7951070336391437,
370
+ "grad_norm": 25.960834503173828,
371
+ "learning_rate": 7.88703187533354e-06,
372
+ "loss": 0.4167,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.8103975535168195,
377
+ "grad_norm": 7.0256028175354,
378
+ "learning_rate": 7.858351759423235e-06,
379
+ "loss": 0.2602,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.8256880733944955,
384
+ "grad_norm": 3.3532984256744385,
385
+ "learning_rate": 7.829671643512932e-06,
386
+ "loss": 0.2569,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.8409785932721713,
391
+ "grad_norm": 5.293972969055176,
392
+ "learning_rate": 7.800991527602627e-06,
393
+ "loss": 0.321,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.8562691131498471,
398
+ "grad_norm": 5.642370223999023,
399
+ "learning_rate": 7.772311411692324e-06,
400
+ "loss": 0.2011,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.8715596330275229,
405
+ "grad_norm": 19.925289154052734,
406
+ "learning_rate": 7.743631295782021e-06,
407
+ "loss": 0.3707,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.8868501529051988,
412
+ "grad_norm": 7.974088668823242,
413
+ "learning_rate": 7.714951179871716e-06,
414
+ "loss": 0.3449,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.9021406727828746,
419
+ "grad_norm": 28.756046295166016,
420
+ "learning_rate": 7.686271063961413e-06,
421
+ "loss": 0.5487,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.9174311926605505,
426
+ "grad_norm": 2.515625238418579,
427
+ "learning_rate": 7.657590948051108e-06,
428
+ "loss": 0.3541,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.9327217125382263,
433
+ "grad_norm": 7.654485702514648,
434
+ "learning_rate": 7.628910832140805e-06,
435
+ "loss": 0.3335,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.9480122324159022,
440
+ "grad_norm": 17.905406951904297,
441
+ "learning_rate": 7.600230716230501e-06,
442
+ "loss": 0.3162,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.963302752293578,
447
+ "grad_norm": 16.23491668701172,
448
+ "learning_rate": 7.571550600320198e-06,
449
+ "loss": 0.3626,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.9785932721712538,
454
+ "grad_norm": 0.26539096236228943,
455
+ "learning_rate": 7.542870484409894e-06,
456
+ "loss": 0.2231,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.9938837920489296,
461
+ "grad_norm": 1.7205305099487305,
462
+ "learning_rate": 7.51419036849959e-06,
463
+ "loss": 0.3941,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 1.0,
468
+ "eval_accuracy": 0.8706962509563887,
469
+ "eval_f1": 0.7540029112081513,
470
+ "eval_loss": 0.29941147565841675,
471
+ "eval_precision": 0.6906666666666667,
472
+ "eval_recall": 0.8301282051282052,
473
+ "eval_runtime": 4.8981,
474
+ "eval_samples_per_second": 266.841,
475
+ "eval_steps_per_second": 16.741,
476
+ "step": 654
477
+ },
478
+ {
479
+ "epoch": 1.0091743119266054,
480
+ "grad_norm": 13.617008209228516,
481
+ "learning_rate": 7.485510252589286e-06,
482
+ "loss": 0.3849,
483
+ "step": 660
484
+ },
485
+ {
486
+ "epoch": 1.0244648318042813,
487
+ "grad_norm": 6.546511650085449,
488
+ "learning_rate": 7.456830136678982e-06,
489
+ "loss": 0.2871,
490
+ "step": 670
491
+ },
492
+ {
493
+ "epoch": 1.039755351681957,
494
+ "grad_norm": 5.344148635864258,
495
+ "learning_rate": 7.428150020768678e-06,
496
+ "loss": 0.2358,
497
+ "step": 680
498
+ },
499
+ {
500
+ "epoch": 1.0550458715596331,
501
+ "grad_norm": 7.6064348220825195,
502
+ "learning_rate": 7.399469904858376e-06,
503
+ "loss": 0.2451,
504
+ "step": 690
505
+ },
506
+ {
507
+ "epoch": 1.070336391437309,
508
+ "grad_norm": 21.806922912597656,
509
+ "learning_rate": 7.370789788948072e-06,
510
+ "loss": 0.2107,
511
+ "step": 700
512
+ },
513
+ {
514
+ "epoch": 1.0856269113149848,
515
+ "grad_norm": 10.673025131225586,
516
+ "learning_rate": 7.342109673037768e-06,
517
+ "loss": 0.2411,
518
+ "step": 710
519
+ },
520
+ {
521
+ "epoch": 1.1009174311926606,
522
+ "grad_norm": 14.94886589050293,
523
+ "learning_rate": 7.313429557127464e-06,
524
+ "loss": 0.1456,
525
+ "step": 720
526
+ },
527
+ {
528
+ "epoch": 1.1162079510703364,
529
+ "grad_norm": 12.618221282958984,
530
+ "learning_rate": 7.28474944121716e-06,
531
+ "loss": 0.2997,
532
+ "step": 730
533
+ },
534
+ {
535
+ "epoch": 1.1314984709480123,
536
+ "grad_norm": 0.2970161736011505,
537
+ "learning_rate": 7.256069325306856e-06,
538
+ "loss": 0.2922,
539
+ "step": 740
540
+ },
541
+ {
542
+ "epoch": 1.146788990825688,
543
+ "grad_norm": 3.0866048336029053,
544
+ "learning_rate": 7.227389209396553e-06,
545
+ "loss": 0.1767,
546
+ "step": 750
547
+ },
548
+ {
549
+ "epoch": 1.162079510703364,
550
+ "grad_norm": 12.53414535522461,
551
+ "learning_rate": 7.198709093486249e-06,
552
+ "loss": 0.2948,
553
+ "step": 760
554
+ },
555
+ {
556
+ "epoch": 1.1773700305810397,
557
+ "grad_norm": 24.84918975830078,
558
+ "learning_rate": 7.170028977575945e-06,
559
+ "loss": 0.1902,
560
+ "step": 770
561
+ },
562
+ {
563
+ "epoch": 1.1926605504587156,
564
+ "grad_norm": 1.028529167175293,
565
+ "learning_rate": 7.141348861665641e-06,
566
+ "loss": 0.2246,
567
+ "step": 780
568
+ },
569
+ {
570
+ "epoch": 1.2079510703363914,
571
+ "grad_norm": 20.015581130981445,
572
+ "learning_rate": 7.112668745755337e-06,
573
+ "loss": 0.2278,
574
+ "step": 790
575
+ },
576
+ {
577
+ "epoch": 1.2232415902140672,
578
+ "grad_norm": 12.658394813537598,
579
+ "learning_rate": 7.083988629845033e-06,
580
+ "loss": 0.3021,
581
+ "step": 800
582
+ },
583
+ {
584
+ "epoch": 1.238532110091743,
585
+ "grad_norm": 4.727680683135986,
586
+ "learning_rate": 7.05530851393473e-06,
587
+ "loss": 0.3192,
588
+ "step": 810
589
+ },
590
+ {
591
+ "epoch": 1.2538226299694188,
592
+ "grad_norm": 16.0307559967041,
593
+ "learning_rate": 7.026628398024426e-06,
594
+ "loss": 0.3534,
595
+ "step": 820
596
+ },
597
+ {
598
+ "epoch": 1.2691131498470947,
599
+ "grad_norm": 4.940106391906738,
600
+ "learning_rate": 6.997948282114122e-06,
601
+ "loss": 0.3564,
602
+ "step": 830
603
+ },
604
+ {
605
+ "epoch": 1.2844036697247707,
606
+ "grad_norm": 23.759117126464844,
607
+ "learning_rate": 6.969268166203818e-06,
608
+ "loss": 0.2746,
609
+ "step": 840
610
+ },
611
+ {
612
+ "epoch": 1.2996941896024465,
613
+ "grad_norm": 25.81449317932129,
614
+ "learning_rate": 6.9405880502935144e-06,
615
+ "loss": 0.4967,
616
+ "step": 850
617
+ },
618
+ {
619
+ "epoch": 1.3149847094801224,
620
+ "grad_norm": 4.882550239562988,
621
+ "learning_rate": 6.9119079343832104e-06,
622
+ "loss": 0.2423,
623
+ "step": 860
624
+ },
625
+ {
626
+ "epoch": 1.3302752293577982,
627
+ "grad_norm": 12.674756050109863,
628
+ "learning_rate": 6.883227818472907e-06,
629
+ "loss": 0.3532,
630
+ "step": 870
631
+ },
632
+ {
633
+ "epoch": 1.345565749235474,
634
+ "grad_norm": 13.883800506591797,
635
+ "learning_rate": 6.854547702562603e-06,
636
+ "loss": 0.2396,
637
+ "step": 880
638
+ },
639
+ {
640
+ "epoch": 1.3608562691131498,
641
+ "grad_norm": 7.71323299407959,
642
+ "learning_rate": 6.825867586652299e-06,
643
+ "loss": 0.2591,
644
+ "step": 890
645
+ },
646
+ {
647
+ "epoch": 1.3761467889908257,
648
+ "grad_norm": 2.174020767211914,
649
+ "learning_rate": 6.797187470741995e-06,
650
+ "loss": 0.2171,
651
+ "step": 900
652
+ },
653
+ {
654
+ "epoch": 1.3914373088685015,
655
+ "grad_norm": 2.351623773574829,
656
+ "learning_rate": 6.768507354831691e-06,
657
+ "loss": 0.3057,
658
+ "step": 910
659
+ },
660
+ {
661
+ "epoch": 1.4067278287461773,
662
+ "grad_norm": 26.668006896972656,
663
+ "learning_rate": 6.739827238921387e-06,
664
+ "loss": 0.2594,
665
+ "step": 920
666
+ },
667
+ {
668
+ "epoch": 1.4220183486238533,
669
+ "grad_norm": 13.497000694274902,
670
+ "learning_rate": 6.711147123011084e-06,
671
+ "loss": 0.2886,
672
+ "step": 930
673
+ },
674
+ {
675
+ "epoch": 1.4373088685015292,
676
+ "grad_norm": 9.161819458007812,
677
+ "learning_rate": 6.682467007100781e-06,
678
+ "loss": 0.2983,
679
+ "step": 940
680
+ },
681
+ {
682
+ "epoch": 1.452599388379205,
683
+ "grad_norm": 22.162334442138672,
684
+ "learning_rate": 6.653786891190477e-06,
685
+ "loss": 0.1644,
686
+ "step": 950
687
+ },
688
+ {
689
+ "epoch": 1.4678899082568808,
690
+ "grad_norm": 0.5806663632392883,
691
+ "learning_rate": 6.625106775280173e-06,
692
+ "loss": 0.2643,
693
+ "step": 960
694
+ },
695
+ {
696
+ "epoch": 1.4831804281345566,
697
+ "grad_norm": 0.31279271841049194,
698
+ "learning_rate": 6.596426659369869e-06,
699
+ "loss": 0.2888,
700
+ "step": 970
701
+ },
702
+ {
703
+ "epoch": 1.4984709480122325,
704
+ "grad_norm": 0.543879508972168,
705
+ "learning_rate": 6.567746543459565e-06,
706
+ "loss": 0.2786,
707
+ "step": 980
708
+ },
709
+ {
710
+ "epoch": 1.5137614678899083,
711
+ "grad_norm": 0.9113799333572388,
712
+ "learning_rate": 6.539066427549262e-06,
713
+ "loss": 0.4815,
714
+ "step": 990
715
+ },
716
+ {
717
+ "epoch": 1.529051987767584,
718
+ "grad_norm": 17.983909606933594,
719
+ "learning_rate": 6.510386311638958e-06,
720
+ "loss": 0.2022,
721
+ "step": 1000
722
+ },
723
+ {
724
+ "epoch": 1.54434250764526,
725
+ "grad_norm": 3.3928322792053223,
726
+ "learning_rate": 6.481706195728654e-06,
727
+ "loss": 0.2222,
728
+ "step": 1010
729
+ },
730
+ {
731
+ "epoch": 1.5596330275229358,
732
+ "grad_norm": 27.15311050415039,
733
+ "learning_rate": 6.45302607981835e-06,
734
+ "loss": 0.2879,
735
+ "step": 1020
736
+ },
737
+ {
738
+ "epoch": 1.5749235474006116,
739
+ "grad_norm": 1.4941473007202148,
740
+ "learning_rate": 6.424345963908046e-06,
741
+ "loss": 0.3431,
742
+ "step": 1030
743
+ },
744
+ {
745
+ "epoch": 1.5902140672782874,
746
+ "grad_norm": 26.38034439086914,
747
+ "learning_rate": 6.395665847997742e-06,
748
+ "loss": 0.2812,
749
+ "step": 1040
750
+ },
751
+ {
752
+ "epoch": 1.6055045871559632,
753
+ "grad_norm": 4.923500061035156,
754
+ "learning_rate": 6.366985732087439e-06,
755
+ "loss": 0.2865,
756
+ "step": 1050
757
+ },
758
+ {
759
+ "epoch": 1.620795107033639,
760
+ "grad_norm": 0.2041570544242859,
761
+ "learning_rate": 6.338305616177135e-06,
762
+ "loss": 0.4086,
763
+ "step": 1060
764
+ },
765
+ {
766
+ "epoch": 1.6360856269113149,
767
+ "grad_norm": 1.037766456604004,
768
+ "learning_rate": 6.309625500266832e-06,
769
+ "loss": 0.1451,
770
+ "step": 1070
771
+ },
772
+ {
773
+ "epoch": 1.6513761467889907,
774
+ "grad_norm": 9.758125305175781,
775
+ "learning_rate": 6.280945384356528e-06,
776
+ "loss": 0.3565,
777
+ "step": 1080
778
+ },
779
+ {
780
+ "epoch": 1.6666666666666665,
781
+ "grad_norm": 0.7395672798156738,
782
+ "learning_rate": 6.252265268446224e-06,
783
+ "loss": 0.3725,
784
+ "step": 1090
785
+ },
786
+ {
787
+ "epoch": 1.6819571865443423,
788
+ "grad_norm": 2.0258843898773193,
789
+ "learning_rate": 6.223585152535921e-06,
790
+ "loss": 0.1929,
791
+ "step": 1100
792
+ },
793
+ {
794
+ "epoch": 1.6972477064220184,
795
+ "grad_norm": 6.11789083480835,
796
+ "learning_rate": 6.194905036625617e-06,
797
+ "loss": 0.3753,
798
+ "step": 1110
799
+ },
800
+ {
801
+ "epoch": 1.7125382262996942,
802
+ "grad_norm": 11.606084823608398,
803
+ "learning_rate": 6.166224920715313e-06,
804
+ "loss": 0.227,
805
+ "step": 1120
806
+ },
807
+ {
808
+ "epoch": 1.72782874617737,
809
+ "grad_norm": 17.951644897460938,
810
+ "learning_rate": 6.137544804805009e-06,
811
+ "loss": 0.4915,
812
+ "step": 1130
813
+ },
814
+ {
815
+ "epoch": 1.7431192660550459,
816
+ "grad_norm": 0.44653165340423584,
817
+ "learning_rate": 6.108864688894705e-06,
818
+ "loss": 0.2273,
819
+ "step": 1140
820
+ },
821
+ {
822
+ "epoch": 1.7584097859327217,
823
+ "grad_norm": 17.01247215270996,
824
+ "learning_rate": 6.080184572984401e-06,
825
+ "loss": 0.3209,
826
+ "step": 1150
827
+ },
828
+ {
829
+ "epoch": 1.7737003058103975,
830
+ "grad_norm": 20.699323654174805,
831
+ "learning_rate": 6.0515044570740976e-06,
832
+ "loss": 0.435,
833
+ "step": 1160
834
+ },
835
+ {
836
+ "epoch": 1.7889908256880735,
837
+ "grad_norm": 30.34465789794922,
838
+ "learning_rate": 6.0228243411637936e-06,
839
+ "loss": 0.2472,
840
+ "step": 1170
841
+ },
842
+ {
843
+ "epoch": 1.8042813455657494,
844
+ "grad_norm": 4.261192798614502,
845
+ "learning_rate": 5.99414422525349e-06,
846
+ "loss": 0.285,
847
+ "step": 1180
848
+ },
849
+ {
850
+ "epoch": 1.8195718654434252,
851
+ "grad_norm": 20.993389129638672,
852
+ "learning_rate": 5.965464109343186e-06,
853
+ "loss": 0.3737,
854
+ "step": 1190
855
+ },
856
+ {
857
+ "epoch": 1.834862385321101,
858
+ "grad_norm": 0.8759846091270447,
859
+ "learning_rate": 5.936783993432882e-06,
860
+ "loss": 0.1866,
861
+ "step": 1200
862
+ },
863
+ {
864
+ "epoch": 1.8501529051987768,
865
+ "grad_norm": 29.079099655151367,
866
+ "learning_rate": 5.9081038775225785e-06,
867
+ "loss": 0.4163,
868
+ "step": 1210
869
+ },
870
+ {
871
+ "epoch": 1.8654434250764527,
872
+ "grad_norm": 3.231898546218872,
873
+ "learning_rate": 5.879423761612275e-06,
874
+ "loss": 0.4359,
875
+ "step": 1220
876
+ },
877
+ {
878
+ "epoch": 1.8807339449541285,
879
+ "grad_norm": 17.104846954345703,
880
+ "learning_rate": 5.850743645701971e-06,
881
+ "loss": 0.2606,
882
+ "step": 1230
883
+ },
884
+ {
885
+ "epoch": 1.8960244648318043,
886
+ "grad_norm": 0.409883588552475,
887
+ "learning_rate": 5.822063529791667e-06,
888
+ "loss": 0.1581,
889
+ "step": 1240
890
+ },
891
+ {
892
+ "epoch": 1.9113149847094801,
893
+ "grad_norm": 10.206957817077637,
894
+ "learning_rate": 5.793383413881363e-06,
895
+ "loss": 0.2835,
896
+ "step": 1250
897
+ },
898
+ {
899
+ "epoch": 1.926605504587156,
900
+ "grad_norm": 16.08922576904297,
901
+ "learning_rate": 5.764703297971059e-06,
902
+ "loss": 0.2338,
903
+ "step": 1260
904
+ },
905
+ {
906
+ "epoch": 1.9418960244648318,
907
+ "grad_norm": 3.5700275897979736,
908
+ "learning_rate": 5.736023182060755e-06,
909
+ "loss": 0.4643,
910
+ "step": 1270
911
+ },
912
+ {
913
+ "epoch": 1.9571865443425076,
914
+ "grad_norm": 13.477959632873535,
915
+ "learning_rate": 5.707343066150452e-06,
916
+ "loss": 0.2526,
917
+ "step": 1280
918
+ },
919
+ {
920
+ "epoch": 1.9724770642201834,
921
+ "grad_norm": 8.30550479888916,
922
+ "learning_rate": 5.678662950240148e-06,
923
+ "loss": 0.4456,
924
+ "step": 1290
925
+ },
926
+ {
927
+ "epoch": 1.9877675840978593,
928
+ "grad_norm": 7.599588871002197,
929
+ "learning_rate": 5.649982834329844e-06,
930
+ "loss": 0.259,
931
+ "step": 1300
932
+ },
933
+ {
934
+ "epoch": 2.0,
935
+ "eval_accuracy": 0.881407804131599,
936
+ "eval_f1": 0.7589424572317263,
937
+ "eval_loss": 0.3265691101551056,
938
+ "eval_precision": 0.7371601208459214,
939
+ "eval_recall": 0.782051282051282,
940
+ "eval_runtime": 4.9078,
941
+ "eval_samples_per_second": 266.31,
942
+ "eval_steps_per_second": 16.708,
943
+ "step": 1308
944
+ },
945
+ {
946
+ "epoch": 2.003058103975535,
947
+ "grad_norm": 5.8759026527404785,
948
+ "learning_rate": 5.62130271841954e-06,
949
+ "loss": 0.2964,
950
+ "step": 1310
951
+ },
952
+ {
953
+ "epoch": 2.018348623853211,
954
+ "grad_norm": 5.663475036621094,
955
+ "learning_rate": 5.592622602509236e-06,
956
+ "loss": 0.2355,
957
+ "step": 1320
958
+ },
959
+ {
960
+ "epoch": 2.0336391437308867,
961
+ "grad_norm": 12.831345558166504,
962
+ "learning_rate": 5.563942486598932e-06,
963
+ "loss": 0.2225,
964
+ "step": 1330
965
+ },
966
+ {
967
+ "epoch": 2.0489296636085625,
968
+ "grad_norm": 0.6970747709274292,
969
+ "learning_rate": 5.53526237068863e-06,
970
+ "loss": 0.279,
971
+ "step": 1340
972
+ },
973
+ {
974
+ "epoch": 2.0642201834862384,
975
+ "grad_norm": 20.61145782470703,
976
+ "learning_rate": 5.506582254778326e-06,
977
+ "loss": 0.2825,
978
+ "step": 1350
979
+ },
980
+ {
981
+ "epoch": 2.079510703363914,
982
+ "grad_norm": 4.7170891761779785,
983
+ "learning_rate": 5.477902138868022e-06,
984
+ "loss": 0.2996,
985
+ "step": 1360
986
+ },
987
+ {
988
+ "epoch": 2.09480122324159,
989
+ "grad_norm": 8.943619728088379,
990
+ "learning_rate": 5.449222022957718e-06,
991
+ "loss": 0.1539,
992
+ "step": 1370
993
+ },
994
+ {
995
+ "epoch": 2.1100917431192663,
996
+ "grad_norm": 0.7099681496620178,
997
+ "learning_rate": 5.420541907047414e-06,
998
+ "loss": 0.1958,
999
+ "step": 1380
1000
+ },
1001
+ {
1002
+ "epoch": 2.1253822629969417,
1003
+ "grad_norm": 3.903538227081299,
1004
+ "learning_rate": 5.39186179113711e-06,
1005
+ "loss": 0.3434,
1006
+ "step": 1390
1007
+ },
1008
+ {
1009
+ "epoch": 2.140672782874618,
1010
+ "grad_norm": 4.8174920082092285,
1011
+ "learning_rate": 5.363181675226807e-06,
1012
+ "loss": 0.266,
1013
+ "step": 1400
1014
+ },
1015
+ {
1016
+ "epoch": 2.1559633027522938,
1017
+ "grad_norm": 13.861398696899414,
1018
+ "learning_rate": 5.334501559316503e-06,
1019
+ "loss": 0.226,
1020
+ "step": 1410
1021
+ },
1022
+ {
1023
+ "epoch": 2.1712538226299696,
1024
+ "grad_norm": 0.26603391766548157,
1025
+ "learning_rate": 5.305821443406199e-06,
1026
+ "loss": 0.1668,
1027
+ "step": 1420
1028
+ },
1029
+ {
1030
+ "epoch": 2.1865443425076454,
1031
+ "grad_norm": 8.830183982849121,
1032
+ "learning_rate": 5.277141327495895e-06,
1033
+ "loss": 0.1988,
1034
+ "step": 1430
1035
+ },
1036
+ {
1037
+ "epoch": 2.2018348623853212,
1038
+ "grad_norm": 0.7192776799201965,
1039
+ "learning_rate": 5.248461211585591e-06,
1040
+ "loss": 0.2137,
1041
+ "step": 1440
1042
+ },
1043
+ {
1044
+ "epoch": 2.217125382262997,
1045
+ "grad_norm": 24.46476173400879,
1046
+ "learning_rate": 5.219781095675287e-06,
1047
+ "loss": 0.3146,
1048
+ "step": 1450
1049
+ },
1050
+ {
1051
+ "epoch": 2.232415902140673,
1052
+ "grad_norm": 36.542945861816406,
1053
+ "learning_rate": 5.191100979764985e-06,
1054
+ "loss": 0.3024,
1055
+ "step": 1460
1056
+ },
1057
+ {
1058
+ "epoch": 2.2477064220183487,
1059
+ "grad_norm": 18.665542602539062,
1060
+ "learning_rate": 5.162420863854681e-06,
1061
+ "loss": 0.45,
1062
+ "step": 1470
1063
+ },
1064
+ {
1065
+ "epoch": 2.2629969418960245,
1066
+ "grad_norm": 0.5163823366165161,
1067
+ "learning_rate": 5.133740747944377e-06,
1068
+ "loss": 0.2063,
1069
+ "step": 1480
1070
+ },
1071
+ {
1072
+ "epoch": 2.2782874617737003,
1073
+ "grad_norm": 12.623830795288086,
1074
+ "learning_rate": 5.105060632034073e-06,
1075
+ "loss": 0.3537,
1076
+ "step": 1490
1077
+ },
1078
+ {
1079
+ "epoch": 2.293577981651376,
1080
+ "grad_norm": 6.660862922668457,
1081
+ "learning_rate": 5.076380516123769e-06,
1082
+ "loss": 0.3763,
1083
+ "step": 1500
1084
+ },
1085
+ {
1086
+ "epoch": 2.308868501529052,
1087
+ "grad_norm": 18.302440643310547,
1088
+ "learning_rate": 5.047700400213465e-06,
1089
+ "loss": 0.2796,
1090
+ "step": 1510
1091
+ },
1092
+ {
1093
+ "epoch": 2.324159021406728,
1094
+ "grad_norm": 5.233463764190674,
1095
+ "learning_rate": 5.019020284303162e-06,
1096
+ "loss": 0.2484,
1097
+ "step": 1520
1098
+ },
1099
+ {
1100
+ "epoch": 2.3394495412844036,
1101
+ "grad_norm": 0.2765680253505707,
1102
+ "learning_rate": 4.990340168392858e-06,
1103
+ "loss": 0.197,
1104
+ "step": 1530
1105
+ },
1106
+ {
1107
+ "epoch": 2.3547400611620795,
1108
+ "grad_norm": 7.559242248535156,
1109
+ "learning_rate": 4.961660052482554e-06,
1110
+ "loss": 0.2913,
1111
+ "step": 1540
1112
+ },
1113
+ {
1114
+ "epoch": 2.3700305810397553,
1115
+ "grad_norm": 20.358930587768555,
1116
+ "learning_rate": 4.93297993657225e-06,
1117
+ "loss": 0.2762,
1118
+ "step": 1550
1119
+ },
1120
+ {
1121
+ "epoch": 2.385321100917431,
1122
+ "grad_norm": 23.49895668029785,
1123
+ "learning_rate": 4.904299820661946e-06,
1124
+ "loss": 0.3327,
1125
+ "step": 1560
1126
+ },
1127
+ {
1128
+ "epoch": 2.400611620795107,
1129
+ "grad_norm": 9.241127014160156,
1130
+ "learning_rate": 4.875619704751642e-06,
1131
+ "loss": 0.2981,
1132
+ "step": 1570
1133
+ },
1134
+ {
1135
+ "epoch": 2.4159021406727827,
1136
+ "grad_norm": 9.627830505371094,
1137
+ "learning_rate": 4.846939588841339e-06,
1138
+ "loss": 0.3052,
1139
+ "step": 1580
1140
+ },
1141
+ {
1142
+ "epoch": 2.4311926605504586,
1143
+ "grad_norm": 6.3946075439453125,
1144
+ "learning_rate": 4.818259472931035e-06,
1145
+ "loss": 0.1022,
1146
+ "step": 1590
1147
+ },
1148
+ {
1149
+ "epoch": 2.4464831804281344,
1150
+ "grad_norm": 0.6927328109741211,
1151
+ "learning_rate": 4.789579357020731e-06,
1152
+ "loss": 0.0922,
1153
+ "step": 1600
1154
+ },
1155
+ {
1156
+ "epoch": 2.46177370030581,
1157
+ "grad_norm": 1.498061180114746,
1158
+ "learning_rate": 4.760899241110427e-06,
1159
+ "loss": 0.1883,
1160
+ "step": 1610
1161
+ },
1162
+ {
1163
+ "epoch": 2.477064220183486,
1164
+ "grad_norm": 3.1200311183929443,
1165
+ "learning_rate": 4.732219125200123e-06,
1166
+ "loss": 0.1963,
1167
+ "step": 1620
1168
+ },
1169
+ {
1170
+ "epoch": 2.4923547400611623,
1171
+ "grad_norm": 5.31001615524292,
1172
+ "learning_rate": 4.703539009289819e-06,
1173
+ "loss": 0.3003,
1174
+ "step": 1630
1175
+ },
1176
+ {
1177
+ "epoch": 2.5076452599388377,
1178
+ "grad_norm": 13.968355178833008,
1179
+ "learning_rate": 4.674858893379516e-06,
1180
+ "loss": 0.3723,
1181
+ "step": 1640
1182
+ },
1183
+ {
1184
+ "epoch": 2.522935779816514,
1185
+ "grad_norm": 30.514612197875977,
1186
+ "learning_rate": 4.646178777469212e-06,
1187
+ "loss": 0.2453,
1188
+ "step": 1650
1189
+ },
1190
+ {
1191
+ "epoch": 2.5382262996941893,
1192
+ "grad_norm": 18.336116790771484,
1193
+ "learning_rate": 4.617498661558908e-06,
1194
+ "loss": 0.2514,
1195
+ "step": 1660
1196
+ },
1197
+ {
1198
+ "epoch": 2.5535168195718656,
1199
+ "grad_norm": 3.3345324993133545,
1200
+ "learning_rate": 4.588818545648604e-06,
1201
+ "loss": 0.1784,
1202
+ "step": 1670
1203
+ },
1204
+ {
1205
+ "epoch": 2.5688073394495414,
1206
+ "grad_norm": 0.6828510165214539,
1207
+ "learning_rate": 4.560138429738301e-06,
1208
+ "loss": 0.1926,
1209
+ "step": 1680
1210
+ },
1211
+ {
1212
+ "epoch": 2.5840978593272173,
1213
+ "grad_norm": 0.3491632044315338,
1214
+ "learning_rate": 4.531458313827997e-06,
1215
+ "loss": 0.3459,
1216
+ "step": 1690
1217
+ },
1218
+ {
1219
+ "epoch": 2.599388379204893,
1220
+ "grad_norm": 14.196633338928223,
1221
+ "learning_rate": 4.502778197917693e-06,
1222
+ "loss": 0.2798,
1223
+ "step": 1700
1224
+ },
1225
+ {
1226
+ "epoch": 2.614678899082569,
1227
+ "grad_norm": 3.067326307296753,
1228
+ "learning_rate": 4.47409808200739e-06,
1229
+ "loss": 0.2548,
1230
+ "step": 1710
1231
+ },
1232
+ {
1233
+ "epoch": 2.6299694189602447,
1234
+ "grad_norm": 10.058695793151855,
1235
+ "learning_rate": 4.445417966097086e-06,
1236
+ "loss": 0.3088,
1237
+ "step": 1720
1238
+ },
1239
+ {
1240
+ "epoch": 2.6452599388379205,
1241
+ "grad_norm": 24.748844146728516,
1242
+ "learning_rate": 4.416737850186782e-06,
1243
+ "loss": 0.2524,
1244
+ "step": 1730
1245
+ },
1246
+ {
1247
+ "epoch": 2.6605504587155964,
1248
+ "grad_norm": 9.57737922668457,
1249
+ "learning_rate": 4.388057734276478e-06,
1250
+ "loss": 0.1089,
1251
+ "step": 1740
1252
+ },
1253
+ {
1254
+ "epoch": 2.675840978593272,
1255
+ "grad_norm": 4.255125999450684,
1256
+ "learning_rate": 4.359377618366174e-06,
1257
+ "loss": 0.2619,
1258
+ "step": 1750
1259
+ },
1260
+ {
1261
+ "epoch": 2.691131498470948,
1262
+ "grad_norm": 12.66395378112793,
1263
+ "learning_rate": 4.33069750245587e-06,
1264
+ "loss": 0.113,
1265
+ "step": 1760
1266
+ },
1267
+ {
1268
+ "epoch": 2.706422018348624,
1269
+ "grad_norm": 27.81719970703125,
1270
+ "learning_rate": 4.302017386545567e-06,
1271
+ "loss": 0.1587,
1272
+ "step": 1770
1273
+ },
1274
+ {
1275
+ "epoch": 2.7217125382262997,
1276
+ "grad_norm": 9.098254203796387,
1277
+ "learning_rate": 4.273337270635263e-06,
1278
+ "loss": 0.1325,
1279
+ "step": 1780
1280
+ },
1281
+ {
1282
+ "epoch": 2.7370030581039755,
1283
+ "grad_norm": 7.441411018371582,
1284
+ "learning_rate": 4.244657154724959e-06,
1285
+ "loss": 0.4097,
1286
+ "step": 1790
1287
+ },
1288
+ {
1289
+ "epoch": 2.7522935779816513,
1290
+ "grad_norm": 28.30503273010254,
1291
+ "learning_rate": 4.215977038814656e-06,
1292
+ "loss": 0.2826,
1293
+ "step": 1800
1294
+ },
1295
+ {
1296
+ "epoch": 2.767584097859327,
1297
+ "grad_norm": 47.75388717651367,
1298
+ "learning_rate": 4.187296922904352e-06,
1299
+ "loss": 0.2992,
1300
+ "step": 1810
1301
+ },
1302
+ {
1303
+ "epoch": 2.782874617737003,
1304
+ "grad_norm": 5.235270023345947,
1305
+ "learning_rate": 4.158616806994048e-06,
1306
+ "loss": 0.1981,
1307
+ "step": 1820
1308
+ },
1309
+ {
1310
+ "epoch": 2.7981651376146788,
1311
+ "grad_norm": 0.08484458923339844,
1312
+ "learning_rate": 4.129936691083745e-06,
1313
+ "loss": 0.3315,
1314
+ "step": 1830
1315
+ },
1316
+ {
1317
+ "epoch": 2.8134556574923546,
1318
+ "grad_norm": 17.93990707397461,
1319
+ "learning_rate": 4.101256575173441e-06,
1320
+ "loss": 0.3326,
1321
+ "step": 1840
1322
+ },
1323
+ {
1324
+ "epoch": 2.8287461773700304,
1325
+ "grad_norm": 0.8848928213119507,
1326
+ "learning_rate": 4.072576459263137e-06,
1327
+ "loss": 0.1852,
1328
+ "step": 1850
1329
+ },
1330
+ {
1331
+ "epoch": 2.8440366972477067,
1332
+ "grad_norm": 1.0666664838790894,
1333
+ "learning_rate": 4.043896343352833e-06,
1334
+ "loss": 0.1235,
1335
+ "step": 1860
1336
+ },
1337
+ {
1338
+ "epoch": 2.859327217125382,
1339
+ "grad_norm": 29.053184509277344,
1340
+ "learning_rate": 4.015216227442529e-06,
1341
+ "loss": 0.309,
1342
+ "step": 1870
1343
+ },
1344
+ {
1345
+ "epoch": 2.8746177370030583,
1346
+ "grad_norm": 30.479965209960938,
1347
+ "learning_rate": 3.986536111532225e-06,
1348
+ "loss": 0.2986,
1349
+ "step": 1880
1350
+ },
1351
+ {
1352
+ "epoch": 2.8899082568807337,
1353
+ "grad_norm": 23.085195541381836,
1354
+ "learning_rate": 3.957855995621922e-06,
1355
+ "loss": 0.1596,
1356
+ "step": 1890
1357
+ },
1358
+ {
1359
+ "epoch": 2.90519877675841,
1360
+ "grad_norm": 1.5042694807052612,
1361
+ "learning_rate": 3.929175879711618e-06,
1362
+ "loss": 0.2213,
1363
+ "step": 1900
1364
+ },
1365
+ {
1366
+ "epoch": 2.9204892966360854,
1367
+ "grad_norm": 22.314062118530273,
1368
+ "learning_rate": 3.900495763801314e-06,
1369
+ "loss": 0.1854,
1370
+ "step": 1910
1371
+ },
1372
+ {
1373
+ "epoch": 2.9357798165137616,
1374
+ "grad_norm": 3.0091068744659424,
1375
+ "learning_rate": 3.8718156478910105e-06,
1376
+ "loss": 0.1231,
1377
+ "step": 1920
1378
+ },
1379
+ {
1380
+ "epoch": 2.9510703363914375,
1381
+ "grad_norm": 0.4003874361515045,
1382
+ "learning_rate": 3.8431355319807065e-06,
1383
+ "loss": 0.1013,
1384
+ "step": 1930
1385
+ },
1386
+ {
1387
+ "epoch": 2.9663608562691133,
1388
+ "grad_norm": 36.91960906982422,
1389
+ "learning_rate": 3.8144554160704025e-06,
1390
+ "loss": 0.1974,
1391
+ "step": 1940
1392
+ },
1393
+ {
1394
+ "epoch": 2.981651376146789,
1395
+ "grad_norm": 10.349553108215332,
1396
+ "learning_rate": 3.785775300160099e-06,
1397
+ "loss": 0.3974,
1398
+ "step": 1950
1399
+ },
1400
+ {
1401
+ "epoch": 2.996941896024465,
1402
+ "grad_norm": 4.602114677429199,
1403
+ "learning_rate": 3.757095184249795e-06,
1404
+ "loss": 0.1764,
1405
+ "step": 1960
1406
+ },
1407
+ {
1408
+ "epoch": 3.0,
1409
+ "eval_accuracy": 0.8837031369548585,
1410
+ "eval_f1": 0.7661538461538462,
1411
+ "eval_loss": 0.416372686624527,
1412
+ "eval_precision": 0.7366863905325444,
1413
+ "eval_recall": 0.7980769230769231,
1414
+ "eval_runtime": 4.9368,
1415
+ "eval_samples_per_second": 264.749,
1416
+ "eval_steps_per_second": 16.61,
1417
+ "step": 1962
1418
+ },
1419
+ {
1420
+ "epoch": 3.0122324159021407,
1421
+ "grad_norm": 0.9072074890136719,
1422
+ "learning_rate": 3.728415068339491e-06,
1423
+ "loss": 0.0342,
1424
+ "step": 1970
1425
+ },
1426
+ {
1427
+ "epoch": 3.0275229357798166,
1428
+ "grad_norm": 11.377812385559082,
1429
+ "learning_rate": 3.699734952429188e-06,
1430
+ "loss": 0.1661,
1431
+ "step": 1980
1432
+ },
1433
+ {
1434
+ "epoch": 3.0428134556574924,
1435
+ "grad_norm": 17.510143280029297,
1436
+ "learning_rate": 3.671054836518884e-06,
1437
+ "loss": 0.1267,
1438
+ "step": 1990
1439
+ },
1440
+ {
1441
+ "epoch": 3.058103975535168,
1442
+ "grad_norm": 0.12311755865812302,
1443
+ "learning_rate": 3.64237472060858e-06,
1444
+ "loss": 0.272,
1445
+ "step": 2000
1446
+ },
1447
+ {
1448
+ "epoch": 3.073394495412844,
1449
+ "grad_norm": 7.080625534057617,
1450
+ "learning_rate": 3.6136946046982763e-06,
1451
+ "loss": 0.3724,
1452
+ "step": 2010
1453
+ },
1454
+ {
1455
+ "epoch": 3.08868501529052,
1456
+ "grad_norm": 15.893166542053223,
1457
+ "learning_rate": 3.5850144887879723e-06,
1458
+ "loss": 0.211,
1459
+ "step": 2020
1460
+ },
1461
+ {
1462
+ "epoch": 3.1039755351681957,
1463
+ "grad_norm": 70.26548767089844,
1464
+ "learning_rate": 3.5563343728776683e-06,
1465
+ "loss": 0.1047,
1466
+ "step": 2030
1467
+ },
1468
+ {
1469
+ "epoch": 3.1192660550458715,
1470
+ "grad_norm": 1.035912036895752,
1471
+ "learning_rate": 3.527654256967365e-06,
1472
+ "loss": 0.0586,
1473
+ "step": 2040
1474
+ },
1475
+ {
1476
+ "epoch": 3.1345565749235473,
1477
+ "grad_norm": 21.898548126220703,
1478
+ "learning_rate": 3.498974141057061e-06,
1479
+ "loss": 0.2289,
1480
+ "step": 2050
1481
+ },
1482
+ {
1483
+ "epoch": 3.149847094801223,
1484
+ "grad_norm": 9.897711753845215,
1485
+ "learning_rate": 3.4702940251467572e-06,
1486
+ "loss": 0.2056,
1487
+ "step": 2060
1488
+ },
1489
+ {
1490
+ "epoch": 3.165137614678899,
1491
+ "grad_norm": 0.49520930647850037,
1492
+ "learning_rate": 3.4416139092364537e-06,
1493
+ "loss": 0.2293,
1494
+ "step": 2070
1495
+ },
1496
+ {
1497
+ "epoch": 3.180428134556575,
1498
+ "grad_norm": 0.14918214082717896,
1499
+ "learning_rate": 3.4129337933261497e-06,
1500
+ "loss": 0.1414,
1501
+ "step": 2080
1502
+ },
1503
+ {
1504
+ "epoch": 3.1957186544342506,
1505
+ "grad_norm": 0.4814736247062683,
1506
+ "learning_rate": 3.3842536774158457e-06,
1507
+ "loss": 0.2611,
1508
+ "step": 2090
1509
+ },
1510
+ {
1511
+ "epoch": 3.2110091743119265,
1512
+ "grad_norm": 0.19969391822814941,
1513
+ "learning_rate": 3.355573561505542e-06,
1514
+ "loss": 0.1482,
1515
+ "step": 2100
1516
+ },
1517
+ {
1518
+ "epoch": 3.2262996941896023,
1519
+ "grad_norm": 0.16247302293777466,
1520
+ "learning_rate": 3.3268934455952385e-06,
1521
+ "loss": 0.4237,
1522
+ "step": 2110
1523
+ },
1524
+ {
1525
+ "epoch": 3.241590214067278,
1526
+ "grad_norm": 17.90416717529297,
1527
+ "learning_rate": 3.2982133296849346e-06,
1528
+ "loss": 0.127,
1529
+ "step": 2120
1530
+ },
1531
+ {
1532
+ "epoch": 3.2568807339449544,
1533
+ "grad_norm": 34.14628219604492,
1534
+ "learning_rate": 3.269533213774631e-06,
1535
+ "loss": 0.1481,
1536
+ "step": 2130
1537
+ },
1538
+ {
1539
+ "epoch": 3.2721712538226297,
1540
+ "grad_norm": 0.18314845860004425,
1541
+ "learning_rate": 3.240853097864327e-06,
1542
+ "loss": 0.1436,
1543
+ "step": 2140
1544
+ },
1545
+ {
1546
+ "epoch": 3.287461773700306,
1547
+ "grad_norm": 0.13257867097854614,
1548
+ "learning_rate": 3.212172981954023e-06,
1549
+ "loss": 0.1203,
1550
+ "step": 2150
1551
+ },
1552
+ {
1553
+ "epoch": 3.302752293577982,
1554
+ "grad_norm": 42.67006301879883,
1555
+ "learning_rate": 3.1834928660437194e-06,
1556
+ "loss": 0.2017,
1557
+ "step": 2160
1558
+ },
1559
+ {
1560
+ "epoch": 3.3180428134556577,
1561
+ "grad_norm": 0.3060612082481384,
1562
+ "learning_rate": 3.154812750133416e-06,
1563
+ "loss": 0.1398,
1564
+ "step": 2170
1565
+ },
1566
+ {
1567
+ "epoch": 3.3333333333333335,
1568
+ "grad_norm": 6.72953987121582,
1569
+ "learning_rate": 3.126132634223112e-06,
1570
+ "loss": 0.1634,
1571
+ "step": 2180
1572
+ },
1573
+ {
1574
+ "epoch": 3.3486238532110093,
1575
+ "grad_norm": 0.4008305072784424,
1576
+ "learning_rate": 3.0974525183128083e-06,
1577
+ "loss": 0.0838,
1578
+ "step": 2190
1579
+ },
1580
+ {
1581
+ "epoch": 3.363914373088685,
1582
+ "grad_norm": 0.2475535124540329,
1583
+ "learning_rate": 3.0687724024025043e-06,
1584
+ "loss": 0.2601,
1585
+ "step": 2200
1586
+ },
1587
+ {
1588
+ "epoch": 3.379204892966361,
1589
+ "grad_norm": 0.6882356405258179,
1590
+ "learning_rate": 3.0400922864922004e-06,
1591
+ "loss": 0.2079,
1592
+ "step": 2210
1593
+ },
1594
+ {
1595
+ "epoch": 3.3944954128440368,
1596
+ "grad_norm": 0.11873319745063782,
1597
+ "learning_rate": 3.0114121705818968e-06,
1598
+ "loss": 0.1479,
1599
+ "step": 2220
1600
+ },
1601
+ {
1602
+ "epoch": 3.4097859327217126,
1603
+ "grad_norm": 2.4591195583343506,
1604
+ "learning_rate": 2.982732054671593e-06,
1605
+ "loss": 0.0835,
1606
+ "step": 2230
1607
+ },
1608
+ {
1609
+ "epoch": 3.4250764525993884,
1610
+ "grad_norm": 33.77700424194336,
1611
+ "learning_rate": 2.9540519387612892e-06,
1612
+ "loss": 0.325,
1613
+ "step": 2240
1614
+ },
1615
+ {
1616
+ "epoch": 3.4403669724770642,
1617
+ "grad_norm": 0.19437803328037262,
1618
+ "learning_rate": 2.9253718228509857e-06,
1619
+ "loss": 0.147,
1620
+ "step": 2250
1621
+ },
1622
+ {
1623
+ "epoch": 3.45565749235474,
1624
+ "grad_norm": 5.968377113342285,
1625
+ "learning_rate": 2.8966917069406817e-06,
1626
+ "loss": 0.3554,
1627
+ "step": 2260
1628
+ },
1629
+ {
1630
+ "epoch": 3.470948012232416,
1631
+ "grad_norm": 12.378327369689941,
1632
+ "learning_rate": 2.8680115910303777e-06,
1633
+ "loss": 0.2007,
1634
+ "step": 2270
1635
+ },
1636
+ {
1637
+ "epoch": 3.4862385321100917,
1638
+ "grad_norm": 0.21296323835849762,
1639
+ "learning_rate": 2.839331475120074e-06,
1640
+ "loss": 0.2296,
1641
+ "step": 2280
1642
+ },
1643
+ {
1644
+ "epoch": 3.5015290519877675,
1645
+ "grad_norm": 13.099735260009766,
1646
+ "learning_rate": 2.81065135920977e-06,
1647
+ "loss": 0.1264,
1648
+ "step": 2290
1649
+ },
1650
+ {
1651
+ "epoch": 3.5168195718654434,
1652
+ "grad_norm": 0.31538063287734985,
1653
+ "learning_rate": 2.781971243299466e-06,
1654
+ "loss": 0.4666,
1655
+ "step": 2300
1656
+ },
1657
+ {
1658
+ "epoch": 3.532110091743119,
1659
+ "grad_norm": 12.299838066101074,
1660
+ "learning_rate": 2.753291127389163e-06,
1661
+ "loss": 0.2505,
1662
+ "step": 2310
1663
+ },
1664
+ {
1665
+ "epoch": 3.547400611620795,
1666
+ "grad_norm": 16.644168853759766,
1667
+ "learning_rate": 2.724611011478859e-06,
1668
+ "loss": 0.4427,
1669
+ "step": 2320
1670
+ },
1671
+ {
1672
+ "epoch": 3.562691131498471,
1673
+ "grad_norm": 39.950279235839844,
1674
+ "learning_rate": 2.695930895568555e-06,
1675
+ "loss": 0.3459,
1676
+ "step": 2330
1677
+ },
1678
+ {
1679
+ "epoch": 3.5779816513761467,
1680
+ "grad_norm": 5.1495256423950195,
1681
+ "learning_rate": 2.6672507796582515e-06,
1682
+ "loss": 0.3096,
1683
+ "step": 2340
1684
+ },
1685
+ {
1686
+ "epoch": 3.5932721712538225,
1687
+ "grad_norm": 35.178226470947266,
1688
+ "learning_rate": 2.6385706637479475e-06,
1689
+ "loss": 0.2467,
1690
+ "step": 2350
1691
+ },
1692
+ {
1693
+ "epoch": 3.6085626911314987,
1694
+ "grad_norm": 6.108614921569824,
1695
+ "learning_rate": 2.6098905478376435e-06,
1696
+ "loss": 0.0156,
1697
+ "step": 2360
1698
+ },
1699
+ {
1700
+ "epoch": 3.623853211009174,
1701
+ "grad_norm": 0.34197351336479187,
1702
+ "learning_rate": 2.5812104319273403e-06,
1703
+ "loss": 0.1566,
1704
+ "step": 2370
1705
+ },
1706
+ {
1707
+ "epoch": 3.6391437308868504,
1708
+ "grad_norm": 5.665993690490723,
1709
+ "learning_rate": 2.5525303160170364e-06,
1710
+ "loss": 0.268,
1711
+ "step": 2380
1712
+ },
1713
+ {
1714
+ "epoch": 3.6544342507645258,
1715
+ "grad_norm": 3.0416011810302734,
1716
+ "learning_rate": 2.5238502001067324e-06,
1717
+ "loss": 0.1028,
1718
+ "step": 2390
1719
+ },
1720
+ {
1721
+ "epoch": 3.669724770642202,
1722
+ "grad_norm": 5.079551696777344,
1723
+ "learning_rate": 2.495170084196429e-06,
1724
+ "loss": 0.2225,
1725
+ "step": 2400
1726
+ },
1727
+ {
1728
+ "epoch": 3.6850152905198774,
1729
+ "grad_norm": 0.19564682245254517,
1730
+ "learning_rate": 2.466489968286125e-06,
1731
+ "loss": 0.1261,
1732
+ "step": 2410
1733
+ },
1734
+ {
1735
+ "epoch": 3.7003058103975537,
1736
+ "grad_norm": 15.542204856872559,
1737
+ "learning_rate": 2.437809852375821e-06,
1738
+ "loss": 0.1688,
1739
+ "step": 2420
1740
+ },
1741
+ {
1742
+ "epoch": 3.7155963302752295,
1743
+ "grad_norm": 0.636468231678009,
1744
+ "learning_rate": 2.4091297364655177e-06,
1745
+ "loss": 0.3542,
1746
+ "step": 2430
1747
+ },
1748
+ {
1749
+ "epoch": 3.7308868501529053,
1750
+ "grad_norm": 0.26796048879623413,
1751
+ "learning_rate": 2.3804496205552137e-06,
1752
+ "loss": 0.0689,
1753
+ "step": 2440
1754
+ },
1755
+ {
1756
+ "epoch": 3.746177370030581,
1757
+ "grad_norm": 9.50084400177002,
1758
+ "learning_rate": 2.3517695046449097e-06,
1759
+ "loss": 0.3902,
1760
+ "step": 2450
1761
+ },
1762
+ {
1763
+ "epoch": 3.761467889908257,
1764
+ "grad_norm": 1.7848141193389893,
1765
+ "learning_rate": 2.323089388734606e-06,
1766
+ "loss": 0.0613,
1767
+ "step": 2460
1768
+ },
1769
+ {
1770
+ "epoch": 3.776758409785933,
1771
+ "grad_norm": 20.41553497314453,
1772
+ "learning_rate": 2.294409272824302e-06,
1773
+ "loss": 0.2541,
1774
+ "step": 2470
1775
+ },
1776
+ {
1777
+ "epoch": 3.7920489296636086,
1778
+ "grad_norm": 1.6424323320388794,
1779
+ "learning_rate": 2.2657291569139986e-06,
1780
+ "loss": 0.2198,
1781
+ "step": 2480
1782
+ },
1783
+ {
1784
+ "epoch": 3.8073394495412844,
1785
+ "grad_norm": 0.24343428015708923,
1786
+ "learning_rate": 2.237049041003695e-06,
1787
+ "loss": 0.2029,
1788
+ "step": 2490
1789
+ },
1790
+ {
1791
+ "epoch": 3.8226299694189603,
1792
+ "grad_norm": 44.71451950073242,
1793
+ "learning_rate": 2.208368925093391e-06,
1794
+ "loss": 0.3573,
1795
+ "step": 2500
1796
+ },
1797
+ {
1798
+ "epoch": 3.837920489296636,
1799
+ "grad_norm": 30.519176483154297,
1800
+ "learning_rate": 2.179688809183087e-06,
1801
+ "loss": 0.3304,
1802
+ "step": 2510
1803
+ },
1804
+ {
1805
+ "epoch": 3.853211009174312,
1806
+ "grad_norm": 25.292348861694336,
1807
+ "learning_rate": 2.1510086932727835e-06,
1808
+ "loss": 0.1079,
1809
+ "step": 2520
1810
+ },
1811
+ {
1812
+ "epoch": 3.8685015290519877,
1813
+ "grad_norm": 35.308006286621094,
1814
+ "learning_rate": 2.1223285773624795e-06,
1815
+ "loss": 0.1774,
1816
+ "step": 2530
1817
+ },
1818
+ {
1819
+ "epoch": 3.8837920489296636,
1820
+ "grad_norm": 0.16474059224128723,
1821
+ "learning_rate": 2.093648461452176e-06,
1822
+ "loss": 0.0328,
1823
+ "step": 2540
1824
+ },
1825
+ {
1826
+ "epoch": 3.8990825688073394,
1827
+ "grad_norm": 0.09989167749881744,
1828
+ "learning_rate": 2.0649683455418724e-06,
1829
+ "loss": 0.1969,
1830
+ "step": 2550
1831
+ },
1832
+ {
1833
+ "epoch": 3.914373088685015,
1834
+ "grad_norm": 18.91350746154785,
1835
+ "learning_rate": 2.0362882296315684e-06,
1836
+ "loss": 0.1445,
1837
+ "step": 2560
1838
+ },
1839
+ {
1840
+ "epoch": 3.929663608562691,
1841
+ "grad_norm": 0.08901531994342804,
1842
+ "learning_rate": 2.0076081137212644e-06,
1843
+ "loss": 0.3773,
1844
+ "step": 2570
1845
+ },
1846
+ {
1847
+ "epoch": 3.944954128440367,
1848
+ "grad_norm": 46.0722770690918,
1849
+ "learning_rate": 1.978927997810961e-06,
1850
+ "loss": 0.3817,
1851
+ "step": 2580
1852
+ },
1853
+ {
1854
+ "epoch": 3.9602446483180427,
1855
+ "grad_norm": 7.898420810699463,
1856
+ "learning_rate": 1.950247881900657e-06,
1857
+ "loss": 0.1688,
1858
+ "step": 2590
1859
+ },
1860
+ {
1861
+ "epoch": 3.9755351681957185,
1862
+ "grad_norm": 51.515220642089844,
1863
+ "learning_rate": 1.9215677659903533e-06,
1864
+ "loss": 0.1521,
1865
+ "step": 2600
1866
+ },
1867
+ {
1868
+ "epoch": 3.9908256880733948,
1869
+ "grad_norm": 0.9271498918533325,
1870
+ "learning_rate": 1.8928876500800495e-06,
1871
+ "loss": 0.1549,
1872
+ "step": 2610
1873
+ },
1874
+ {
1875
+ "epoch": 4.0,
1876
+ "eval_accuracy": 0.8859984697781178,
1877
+ "eval_f1": 0.7585089141004863,
1878
+ "eval_loss": 0.44139671325683594,
1879
+ "eval_precision": 0.7672131147540984,
1880
+ "eval_recall": 0.75,
1881
+ "eval_runtime": 4.9043,
1882
+ "eval_samples_per_second": 266.502,
1883
+ "eval_steps_per_second": 16.72,
1884
+ "step": 2616
1885
+ }
1886
+ ],
1887
+ "logging_steps": 10,
1888
+ "max_steps": 3270,
1889
+ "num_input_tokens_seen": 0,
1890
+ "num_train_epochs": 5,
1891
+ "save_steps": 500,
1892
+ "stateful_callbacks": {
1893
+ "TrainerControl": {
1894
+ "args": {
1895
+ "should_epoch_stop": false,
1896
+ "should_evaluate": false,
1897
+ "should_log": false,
1898
+ "should_save": true,
1899
+ "should_training_stop": false
1900
+ },
1901
+ "attributes": {}
1902
+ }
1903
+ },
1904
+ "total_flos": 2764329675153408.0,
1905
+ "train_batch_size": 8,
1906
+ "trial_name": null,
1907
+ "trial_params": {
1908
+ "learning_rate": 9.378397902669336e-06,
1909
+ "num_train_epochs": 5,
1910
+ "per_device_train_batch_size": 8,
1911
+ "seed": 28
1912
+ }
1913
+ }
run-0/checkpoint-2616/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40f24b9f959faf9f78234785503a7c7dfa403799a1c001c4311a87daac4c4bbd
3
+ size 5304
run-0/checkpoint-2616/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-1/checkpoint-654/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
run-1/checkpoint-654/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e405c55984cc27cd3e03d775338134281503bb3531f22447f2059c544e59f974
3
+ size 541317368
run-1/checkpoint-654/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4abcb6fa6739bd040bfeaa4c5f049fc4925654fa44c189bee6a289c643b29dfa
3
+ size 1082696890
run-1/checkpoint-654/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38a2e74768590541c8a0877c5c1da8de71a1eebc687a1582315c96c87c8f2c4a
3
+ size 14244
run-1/checkpoint-654/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d1c3c746a4e9df76413db7c5ea09b20c2d96f0564f1eef2b90a35c7161b5e07
3
+ size 1064
run-1/checkpoint-654/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-1/checkpoint-654/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-1/checkpoint-654/tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": false,
47
+ "mask_token": "[MASK]",
48
+ "model_max_length": 512,
49
+ "pad_token": "[PAD]",
50
+ "sep_token": "[SEP]",
51
+ "strip_accents": null,
52
+ "tokenize_chinese_chars": true,
53
+ "tokenizer_class": "DistilBertTokenizer",
54
+ "unk_token": "[UNK]"
55
+ }
run-1/checkpoint-654/trainer_state.json ADDED
@@ -0,0 +1,505 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8783473603672532,
3
+ "best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-1/checkpoint-654",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 654,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01529051987767584,
13
+ "grad_norm": 2.427048683166504,
14
+ "learning_rate": 7.600520205160823e-05,
15
+ "loss": 0.5308,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03058103975535168,
20
+ "grad_norm": 0.8249368071556091,
21
+ "learning_rate": 7.482499705080687e-05,
22
+ "loss": 0.4091,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.045871559633027525,
27
+ "grad_norm": 3.725421905517578,
28
+ "learning_rate": 7.36447920500055e-05,
29
+ "loss": 0.6422,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.06116207951070336,
34
+ "grad_norm": 2.656184196472168,
35
+ "learning_rate": 7.246458704920412e-05,
36
+ "loss": 0.3677,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.0764525993883792,
41
+ "grad_norm": 3.110931396484375,
42
+ "learning_rate": 7.128438204840275e-05,
43
+ "loss": 0.5322,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.09174311926605505,
48
+ "grad_norm": 6.087479114532471,
49
+ "learning_rate": 7.010417704760138e-05,
50
+ "loss": 0.4355,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.10703363914373089,
55
+ "grad_norm": 0.9337313771247864,
56
+ "learning_rate": 6.892397204680002e-05,
57
+ "loss": 0.408,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.12232415902140673,
62
+ "grad_norm": 6.667540073394775,
63
+ "learning_rate": 6.774376704599865e-05,
64
+ "loss": 0.3401,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.13761467889908258,
69
+ "grad_norm": 3.2563517093658447,
70
+ "learning_rate": 6.656356204519727e-05,
71
+ "loss": 0.2753,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.1529051987767584,
76
+ "grad_norm": 0.436352014541626,
77
+ "learning_rate": 6.53833570443959e-05,
78
+ "loss": 0.1758,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.16819571865443425,
83
+ "grad_norm": 2.314504623413086,
84
+ "learning_rate": 6.420315204359453e-05,
85
+ "loss": 0.4508,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.1834862385321101,
90
+ "grad_norm": 5.20156192779541,
91
+ "learning_rate": 6.302294704279317e-05,
92
+ "loss": 0.3001,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.19877675840978593,
97
+ "grad_norm": 0.4600387513637543,
98
+ "learning_rate": 6.184274204199179e-05,
99
+ "loss": 0.4469,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.21406727828746178,
104
+ "grad_norm": 7.823936462402344,
105
+ "learning_rate": 6.0662537041190425e-05,
106
+ "loss": 0.535,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.22935779816513763,
111
+ "grad_norm": 8.010571479797363,
112
+ "learning_rate": 5.948233204038906e-05,
113
+ "loss": 0.3132,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.24464831804281345,
118
+ "grad_norm": 0.605796754360199,
119
+ "learning_rate": 5.8302127039587676e-05,
120
+ "loss": 0.4912,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.2599388379204893,
125
+ "grad_norm": 1.5800836086273193,
126
+ "learning_rate": 5.712192203878631e-05,
127
+ "loss": 0.4953,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.27522935779816515,
132
+ "grad_norm": 5.56485652923584,
133
+ "learning_rate": 5.594171703798494e-05,
134
+ "loss": 0.4611,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.290519877675841,
139
+ "grad_norm": 1.0299450159072876,
140
+ "learning_rate": 5.476151203718357e-05,
141
+ "loss": 0.3979,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.3058103975535168,
146
+ "grad_norm": 3.140592098236084,
147
+ "learning_rate": 5.35813070363822e-05,
148
+ "loss": 0.6532,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.3211009174311927,
153
+ "grad_norm": 7.104299545288086,
154
+ "learning_rate": 5.240110203558083e-05,
155
+ "loss": 0.4224,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.3363914373088685,
160
+ "grad_norm": 0.5627728700637817,
161
+ "learning_rate": 5.1220897034779465e-05,
162
+ "loss": 0.2206,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.3516819571865443,
167
+ "grad_norm": 5.487335681915283,
168
+ "learning_rate": 5.0040692033978085e-05,
169
+ "loss": 0.4927,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.3669724770642202,
174
+ "grad_norm": 2.892604112625122,
175
+ "learning_rate": 4.886048703317672e-05,
176
+ "loss": 0.6768,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.382262996941896,
181
+ "grad_norm": 7.0879807472229,
182
+ "learning_rate": 4.768028203237535e-05,
183
+ "loss": 0.3987,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.39755351681957185,
188
+ "grad_norm": 28.431095123291016,
189
+ "learning_rate": 4.650007703157398e-05,
190
+ "loss": 0.3762,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.41284403669724773,
195
+ "grad_norm": 30.579540252685547,
196
+ "learning_rate": 4.5319872030772615e-05,
197
+ "loss": 0.3611,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.42813455657492355,
202
+ "grad_norm": 0.6805054545402527,
203
+ "learning_rate": 4.413966702997124e-05,
204
+ "loss": 0.4075,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.4434250764525994,
209
+ "grad_norm": 0.8125110864639282,
210
+ "learning_rate": 4.295946202916987e-05,
211
+ "loss": 0.3955,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.45871559633027525,
216
+ "grad_norm": 0.7836199998855591,
217
+ "learning_rate": 4.17792570283685e-05,
218
+ "loss": 0.2571,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.4740061162079511,
223
+ "grad_norm": 1.1527467966079712,
224
+ "learning_rate": 4.059905202756713e-05,
225
+ "loss": 0.4111,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.4892966360856269,
230
+ "grad_norm": 2.705764055252075,
231
+ "learning_rate": 3.941884702676576e-05,
232
+ "loss": 0.3557,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5045871559633027,
237
+ "grad_norm": 2.6506361961364746,
238
+ "learning_rate": 3.823864202596439e-05,
239
+ "loss": 0.39,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.5198776758409785,
244
+ "grad_norm": 1.7446061372756958,
245
+ "learning_rate": 3.7058437025163016e-05,
246
+ "loss": 0.3624,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.5351681957186545,
251
+ "grad_norm": 2.0617918968200684,
252
+ "learning_rate": 3.587823202436165e-05,
253
+ "loss": 0.4086,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.5504587155963303,
258
+ "grad_norm": 35.361595153808594,
259
+ "learning_rate": 3.469802702356028e-05,
260
+ "loss": 0.2898,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.5657492354740061,
265
+ "grad_norm": 2.274883508682251,
266
+ "learning_rate": 3.351782202275891e-05,
267
+ "loss": 0.3872,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.581039755351682,
272
+ "grad_norm": 4.199743747711182,
273
+ "learning_rate": 3.233761702195754e-05,
274
+ "loss": 0.3484,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.5963302752293578,
279
+ "grad_norm": 4.481986045837402,
280
+ "learning_rate": 3.115741202115617e-05,
281
+ "loss": 0.3719,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.6116207951070336,
286
+ "grad_norm": 1.4046356678009033,
287
+ "learning_rate": 2.99772070203548e-05,
288
+ "loss": 0.2779,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.6269113149847095,
293
+ "grad_norm": 4.377758979797363,
294
+ "learning_rate": 2.8797002019553428e-05,
295
+ "loss": 0.3411,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.6422018348623854,
300
+ "grad_norm": 7.996399402618408,
301
+ "learning_rate": 2.761679701875206e-05,
302
+ "loss": 0.4509,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.6574923547400612,
307
+ "grad_norm": 1.9654350280761719,
308
+ "learning_rate": 2.6436592017950687e-05,
309
+ "loss": 0.256,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.672782874617737,
314
+ "grad_norm": 6.0152435302734375,
315
+ "learning_rate": 2.525638701714932e-05,
316
+ "loss": 0.3963,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.6880733944954128,
321
+ "grad_norm": 1.1808817386627197,
322
+ "learning_rate": 2.4076182016347952e-05,
323
+ "loss": 0.2257,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.7033639143730887,
328
+ "grad_norm": 2.4409139156341553,
329
+ "learning_rate": 2.2895977015546578e-05,
330
+ "loss": 0.476,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.7186544342507645,
335
+ "grad_norm": 2.775444507598877,
336
+ "learning_rate": 2.171577201474521e-05,
337
+ "loss": 0.2066,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.7339449541284404,
342
+ "grad_norm": 1.8437581062316895,
343
+ "learning_rate": 2.053556701394384e-05,
344
+ "loss": 0.2967,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.7492354740061162,
349
+ "grad_norm": 0.8174537420272827,
350
+ "learning_rate": 1.935536201314247e-05,
351
+ "loss": 0.2105,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.764525993883792,
356
+ "grad_norm": 6.47490119934082,
357
+ "learning_rate": 1.8175157012341098e-05,
358
+ "loss": 0.2868,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.7798165137614679,
363
+ "grad_norm": 0.36116883158683777,
364
+ "learning_rate": 1.699495201153973e-05,
365
+ "loss": 0.2572,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.7951070336391437,
370
+ "grad_norm": 3.4261679649353027,
371
+ "learning_rate": 1.581474701073836e-05,
372
+ "loss": 0.4464,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.8103975535168195,
377
+ "grad_norm": 5.803319931030273,
378
+ "learning_rate": 1.4634542009936987e-05,
379
+ "loss": 0.3471,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.8256880733944955,
384
+ "grad_norm": 13.390414237976074,
385
+ "learning_rate": 1.345433700913562e-05,
386
+ "loss": 0.305,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.8409785932721713,
391
+ "grad_norm": 8.899558067321777,
392
+ "learning_rate": 1.227413200833425e-05,
393
+ "loss": 0.3923,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.8562691131498471,
398
+ "grad_norm": 7.016339302062988,
399
+ "learning_rate": 1.1093927007532877e-05,
400
+ "loss": 0.2968,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.8715596330275229,
405
+ "grad_norm": 4.5831427574157715,
406
+ "learning_rate": 9.91372200673151e-06,
407
+ "loss": 0.2957,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.8868501529051988,
412
+ "grad_norm": 1.5068073272705078,
413
+ "learning_rate": 8.733517005930139e-06,
414
+ "loss": 0.2239,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.9021406727828746,
419
+ "grad_norm": 11.313027381896973,
420
+ "learning_rate": 7.553312005128768e-06,
421
+ "loss": 0.368,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.9174311926605505,
426
+ "grad_norm": 6.738920211791992,
427
+ "learning_rate": 6.373107004327399e-06,
428
+ "loss": 0.4739,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.9327217125382263,
433
+ "grad_norm": 4.635502815246582,
434
+ "learning_rate": 5.192902003526027e-06,
435
+ "loss": 0.3667,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.9480122324159022,
440
+ "grad_norm": 2.101200580596924,
441
+ "learning_rate": 4.012697002724658e-06,
442
+ "loss": 0.4852,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.963302752293578,
447
+ "grad_norm": 0.7536319494247437,
448
+ "learning_rate": 2.8324920019232884e-06,
449
+ "loss": 0.1926,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.9785932721712538,
454
+ "grad_norm": 2.7094335556030273,
455
+ "learning_rate": 1.652287001121918e-06,
456
+ "loss": 0.3674,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.9938837920489296,
461
+ "grad_norm": 30.233768463134766,
462
+ "learning_rate": 4.72082000320548e-07,
463
+ "loss": 0.3524,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 1.0,
468
+ "eval_accuracy": 0.8783473603672532,
469
+ "eval_f1": 0.7527216174183515,
470
+ "eval_loss": 0.3004890978336334,
471
+ "eval_precision": 0.7311178247734139,
472
+ "eval_recall": 0.7756410256410257,
473
+ "eval_runtime": 4.9159,
474
+ "eval_samples_per_second": 265.874,
475
+ "eval_steps_per_second": 16.681,
476
+ "step": 654
477
+ }
478
+ ],
479
+ "logging_steps": 10,
480
+ "max_steps": 654,
481
+ "num_input_tokens_seen": 0,
482
+ "num_train_epochs": 1,
483
+ "save_steps": 500,
484
+ "stateful_callbacks": {
485
+ "TrainerControl": {
486
+ "args": {
487
+ "should_epoch_stop": false,
488
+ "should_evaluate": false,
489
+ "should_log": false,
490
+ "should_save": true,
491
+ "should_training_stop": true
492
+ },
493
+ "attributes": {}
494
+ }
495
+ },
496
+ "total_flos": 688830473011200.0,
497
+ "train_batch_size": 8,
498
+ "trial_name": null,
499
+ "trial_params": {
500
+ "learning_rate": 7.71854070524096e-05,
501
+ "num_train_epochs": 1,
502
+ "per_device_train_batch_size": 8,
503
+ "seed": 4
504
+ }
505
+ }
run-1/checkpoint-654/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:076f4e50b7e42e2dabdab3aac7da71a30ce1f51982e20a1ad3ab423d6f31bb1b
3
+ size 5304
run-1/checkpoint-654/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-492/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
run-2/checkpoint-492/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4288c1f01672e804e837dc6923e20bb8a780538ced5e9f71e36fb446b2ebfc57
3
+ size 541317368
run-2/checkpoint-492/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee80cbb82698e2781ac0155ce7534eb528232b5474a565ae6aa34e588043ce5a
3
+ size 1082696890
run-2/checkpoint-492/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4232552436e4651f6371a0dd4079cf82889ecfe75ebea70442299843daf08cb9
3
+ size 14308
run-2/checkpoint-492/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:048f8be56a6b40d0f9c2d9ee6761b79f64b68cbde8bd277cb478882c33e977e5
3
+ size 1064
run-2/checkpoint-492/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-2/checkpoint-492/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-492/tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": false,
47
+ "mask_token": "[MASK]",
48
+ "model_max_length": 512,
49
+ "pad_token": "[PAD]",
50
+ "sep_token": "[SEP]",
51
+ "strip_accents": null,
52
+ "tokenize_chinese_chars": true,
53
+ "tokenizer_class": "DistilBertTokenizer",
54
+ "unk_token": "[UNK]"
55
+ }
run-2/checkpoint-492/trainer_state.json ADDED
@@ -0,0 +1,417 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.791124713083397,
3
+ "best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-2/checkpoint-492",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 492,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06097560975609756,
13
+ "grad_norm": 3.2577075958251953,
14
+ "learning_rate": 1.7097812500543722e-06,
15
+ "loss": 0.7201,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.12195121951219512,
20
+ "grad_norm": 1.7486741542816162,
21
+ "learning_rate": 1.6743086100117504e-06,
22
+ "loss": 0.6884,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.18292682926829268,
27
+ "grad_norm": 2.3384933471679688,
28
+ "learning_rate": 1.6388359699691285e-06,
29
+ "loss": 0.6482,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.24390243902439024,
34
+ "grad_norm": 2.0673463344573975,
35
+ "learning_rate": 1.6033633299265067e-06,
36
+ "loss": 0.6254,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.3048780487804878,
41
+ "grad_norm": 1.9012755155563354,
42
+ "learning_rate": 1.5678906898838849e-06,
43
+ "loss": 0.6049,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.36585365853658536,
48
+ "grad_norm": 1.1691405773162842,
49
+ "learning_rate": 1.532418049841263e-06,
50
+ "loss": 0.5601,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.4268292682926829,
55
+ "grad_norm": 2.259779691696167,
56
+ "learning_rate": 1.4969454097986412e-06,
57
+ "loss": 0.5567,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.4878048780487805,
62
+ "grad_norm": 1.7687610387802124,
63
+ "learning_rate": 1.4614727697560194e-06,
64
+ "loss": 0.5499,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.5487804878048781,
69
+ "grad_norm": 1.2938015460968018,
70
+ "learning_rate": 1.4260001297133975e-06,
71
+ "loss": 0.5247,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.6097560975609756,
76
+ "grad_norm": 1.534727692604065,
77
+ "learning_rate": 1.3905274896707757e-06,
78
+ "loss": 0.5265,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.6707317073170732,
83
+ "grad_norm": 1.6018732786178589,
84
+ "learning_rate": 1.3550548496281539e-06,
85
+ "loss": 0.4982,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.7317073170731707,
90
+ "grad_norm": 1.9590264558792114,
91
+ "learning_rate": 1.319582209585532e-06,
92
+ "loss": 0.4593,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.7926829268292683,
97
+ "grad_norm": 1.6383683681488037,
98
+ "learning_rate": 1.2841095695429102e-06,
99
+ "loss": 0.5132,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.8536585365853658,
104
+ "grad_norm": 1.6192731857299805,
105
+ "learning_rate": 1.2486369295002884e-06,
106
+ "loss": 0.4783,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.9146341463414634,
111
+ "grad_norm": 1.8902620077133179,
112
+ "learning_rate": 1.2131642894576666e-06,
113
+ "loss": 0.4747,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.975609756097561,
118
+ "grad_norm": 2.1500866413116455,
119
+ "learning_rate": 1.1776916494150447e-06,
120
+ "loss": 0.5078,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 1.0,
125
+ "eval_accuracy": 0.7697016067329763,
126
+ "eval_f1": 0.0960960960960961,
127
+ "eval_loss": 0.4581395089626312,
128
+ "eval_precision": 0.7619047619047619,
129
+ "eval_recall": 0.05128205128205128,
130
+ "eval_runtime": 4.8871,
131
+ "eval_samples_per_second": 267.439,
132
+ "eval_steps_per_second": 16.779,
133
+ "step": 164
134
+ },
135
+ {
136
+ "epoch": 1.0365853658536586,
137
+ "grad_norm": 2.7179384231567383,
138
+ "learning_rate": 1.1422190093724229e-06,
139
+ "loss": 0.4381,
140
+ "step": 170
141
+ },
142
+ {
143
+ "epoch": 1.0975609756097562,
144
+ "grad_norm": 1.8275049924850464,
145
+ "learning_rate": 1.106746369329801e-06,
146
+ "loss": 0.446,
147
+ "step": 180
148
+ },
149
+ {
150
+ "epoch": 1.1585365853658536,
151
+ "grad_norm": 1.6741889715194702,
152
+ "learning_rate": 1.0712737292871792e-06,
153
+ "loss": 0.4451,
154
+ "step": 190
155
+ },
156
+ {
157
+ "epoch": 1.2195121951219512,
158
+ "grad_norm": 1.8023988008499146,
159
+ "learning_rate": 1.0358010892445574e-06,
160
+ "loss": 0.4538,
161
+ "step": 200
162
+ },
163
+ {
164
+ "epoch": 1.2804878048780488,
165
+ "grad_norm": 2.3258135318756104,
166
+ "learning_rate": 1.0003284492019356e-06,
167
+ "loss": 0.4531,
168
+ "step": 210
169
+ },
170
+ {
171
+ "epoch": 1.3414634146341464,
172
+ "grad_norm": 2.654583215713501,
173
+ "learning_rate": 9.648558091593137e-07,
174
+ "loss": 0.426,
175
+ "step": 220
176
+ },
177
+ {
178
+ "epoch": 1.4024390243902438,
179
+ "grad_norm": 2.7538747787475586,
180
+ "learning_rate": 9.293831691166919e-07,
181
+ "loss": 0.4056,
182
+ "step": 230
183
+ },
184
+ {
185
+ "epoch": 1.4634146341463414,
186
+ "grad_norm": 1.887471079826355,
187
+ "learning_rate": 8.939105290740702e-07,
188
+ "loss": 0.448,
189
+ "step": 240
190
+ },
191
+ {
192
+ "epoch": 1.524390243902439,
193
+ "grad_norm": 1.7449885606765747,
194
+ "learning_rate": 8.584378890314483e-07,
195
+ "loss": 0.4132,
196
+ "step": 250
197
+ },
198
+ {
199
+ "epoch": 1.5853658536585367,
200
+ "grad_norm": 1.4998096227645874,
201
+ "learning_rate": 8.229652489888265e-07,
202
+ "loss": 0.4164,
203
+ "step": 260
204
+ },
205
+ {
206
+ "epoch": 1.6463414634146343,
207
+ "grad_norm": 4.520744800567627,
208
+ "learning_rate": 7.874926089462047e-07,
209
+ "loss": 0.4662,
210
+ "step": 270
211
+ },
212
+ {
213
+ "epoch": 1.7073170731707317,
214
+ "grad_norm": 1.9681720733642578,
215
+ "learning_rate": 7.520199689035828e-07,
216
+ "loss": 0.3977,
217
+ "step": 280
218
+ },
219
+ {
220
+ "epoch": 1.7682926829268293,
221
+ "grad_norm": 3.49233341217041,
222
+ "learning_rate": 7.16547328860961e-07,
223
+ "loss": 0.4133,
224
+ "step": 290
225
+ },
226
+ {
227
+ "epoch": 1.8292682926829267,
228
+ "grad_norm": 2.177455425262451,
229
+ "learning_rate": 6.810746888183392e-07,
230
+ "loss": 0.4403,
231
+ "step": 300
232
+ },
233
+ {
234
+ "epoch": 1.8902439024390243,
235
+ "grad_norm": 3.338192939758301,
236
+ "learning_rate": 6.456020487757174e-07,
237
+ "loss": 0.4154,
238
+ "step": 310
239
+ },
240
+ {
241
+ "epoch": 1.951219512195122,
242
+ "grad_norm": 2.005662441253662,
243
+ "learning_rate": 6.101294087330955e-07,
244
+ "loss": 0.4137,
245
+ "step": 320
246
+ },
247
+ {
248
+ "epoch": 2.0,
249
+ "eval_accuracy": 0.7888293802601377,
250
+ "eval_f1": 0.31,
251
+ "eval_loss": 0.399868905544281,
252
+ "eval_precision": 0.7045454545454546,
253
+ "eval_recall": 0.1987179487179487,
254
+ "eval_runtime": 4.9013,
255
+ "eval_samples_per_second": 266.666,
256
+ "eval_steps_per_second": 16.73,
257
+ "step": 328
258
+ },
259
+ {
260
+ "epoch": 2.0121951219512195,
261
+ "grad_norm": 1.8187142610549927,
262
+ "learning_rate": 5.746567686904737e-07,
263
+ "loss": 0.4231,
264
+ "step": 330
265
+ },
266
+ {
267
+ "epoch": 2.073170731707317,
268
+ "grad_norm": 2.4677720069885254,
269
+ "learning_rate": 5.391841286478519e-07,
270
+ "loss": 0.3832,
271
+ "step": 340
272
+ },
273
+ {
274
+ "epoch": 2.1341463414634148,
275
+ "grad_norm": 2.2957379817962646,
276
+ "learning_rate": 5.0371148860523e-07,
277
+ "loss": 0.3671,
278
+ "step": 350
279
+ },
280
+ {
281
+ "epoch": 2.1951219512195124,
282
+ "grad_norm": 1.6115466356277466,
283
+ "learning_rate": 4.682388485626082e-07,
284
+ "loss": 0.3776,
285
+ "step": 360
286
+ },
287
+ {
288
+ "epoch": 2.2560975609756095,
289
+ "grad_norm": 2.2633934020996094,
290
+ "learning_rate": 4.327662085199863e-07,
291
+ "loss": 0.4199,
292
+ "step": 370
293
+ },
294
+ {
295
+ "epoch": 2.317073170731707,
296
+ "grad_norm": 3.485372304916382,
297
+ "learning_rate": 3.972935684773645e-07,
298
+ "loss": 0.3784,
299
+ "step": 380
300
+ },
301
+ {
302
+ "epoch": 2.3780487804878048,
303
+ "grad_norm": 1.9504722356796265,
304
+ "learning_rate": 3.6182092843474265e-07,
305
+ "loss": 0.3538,
306
+ "step": 390
307
+ },
308
+ {
309
+ "epoch": 2.4390243902439024,
310
+ "grad_norm": 2.1430771350860596,
311
+ "learning_rate": 3.263482883921208e-07,
312
+ "loss": 0.4277,
313
+ "step": 400
314
+ },
315
+ {
316
+ "epoch": 2.5,
317
+ "grad_norm": 1.9325039386749268,
318
+ "learning_rate": 2.90875648349499e-07,
319
+ "loss": 0.4083,
320
+ "step": 410
321
+ },
322
+ {
323
+ "epoch": 2.5609756097560976,
324
+ "grad_norm": 1.617710828781128,
325
+ "learning_rate": 2.5540300830687715e-07,
326
+ "loss": 0.4297,
327
+ "step": 420
328
+ },
329
+ {
330
+ "epoch": 2.6219512195121952,
331
+ "grad_norm": 3.280532121658325,
332
+ "learning_rate": 2.1993036826425532e-07,
333
+ "loss": 0.4056,
334
+ "step": 430
335
+ },
336
+ {
337
+ "epoch": 2.682926829268293,
338
+ "grad_norm": 2.3135297298431396,
339
+ "learning_rate": 1.8445772822163352e-07,
340
+ "loss": 0.438,
341
+ "step": 440
342
+ },
343
+ {
344
+ "epoch": 2.7439024390243905,
345
+ "grad_norm": 2.195956230163574,
346
+ "learning_rate": 1.489850881790117e-07,
347
+ "loss": 0.4187,
348
+ "step": 450
349
+ },
350
+ {
351
+ "epoch": 2.8048780487804876,
352
+ "grad_norm": 2.371246099472046,
353
+ "learning_rate": 1.1351244813638987e-07,
354
+ "loss": 0.3459,
355
+ "step": 460
356
+ },
357
+ {
358
+ "epoch": 2.8658536585365852,
359
+ "grad_norm": 3.243722915649414,
360
+ "learning_rate": 7.803980809376802e-08,
361
+ "loss": 0.3754,
362
+ "step": 470
363
+ },
364
+ {
365
+ "epoch": 2.926829268292683,
366
+ "grad_norm": 2.472179412841797,
367
+ "learning_rate": 4.25671680511462e-08,
368
+ "loss": 0.4097,
369
+ "step": 480
370
+ },
371
+ {
372
+ "epoch": 2.9878048780487805,
373
+ "grad_norm": 2.2248072624206543,
374
+ "learning_rate": 7.094528008524367e-09,
375
+ "loss": 0.3784,
376
+ "step": 490
377
+ },
378
+ {
379
+ "epoch": 3.0,
380
+ "eval_accuracy": 0.791124713083397,
381
+ "eval_f1": 0.3357664233576642,
382
+ "eval_loss": 0.386940062046051,
383
+ "eval_precision": 0.696969696969697,
384
+ "eval_recall": 0.22115384615384615,
385
+ "eval_runtime": 4.9814,
386
+ "eval_samples_per_second": 262.374,
387
+ "eval_steps_per_second": 16.461,
388
+ "step": 492
389
+ }
390
+ ],
391
+ "logging_steps": 10,
392
+ "max_steps": 492,
393
+ "num_input_tokens_seen": 0,
394
+ "num_train_epochs": 3,
395
+ "save_steps": 500,
396
+ "stateful_callbacks": {
397
+ "TrainerControl": {
398
+ "args": {
399
+ "should_epoch_stop": false,
400
+ "should_evaluate": false,
401
+ "should_log": false,
402
+ "should_save": true,
403
+ "should_training_stop": true
404
+ },
405
+ "attributes": {}
406
+ }
407
+ },
408
+ "total_flos": 2071790114979840.0,
409
+ "train_batch_size": 32,
410
+ "trial_name": null,
411
+ "trial_params": {
412
+ "learning_rate": 1.745253890096994e-06,
413
+ "num_train_epochs": 3,
414
+ "per_device_train_batch_size": 32,
415
+ "seed": 32
416
+ }
417
+ }
run-2/checkpoint-492/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f480cd110cd7aed1ecf9813a881be3f8605f38e33ceb066267c06832fac0b9e1
3
+ size 5304
run-2/checkpoint-492/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-328/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
run-3/checkpoint-328/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7d72d4db36fdf4eee8a19cc74341ad3d3a9c63ce6bcb32175a46289dfa33e54
3
+ size 541317368
run-3/checkpoint-328/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ae012ac8ab8cb26d222eaee096dd279cf6e3d78a7e3fe1b58685739642d7343
3
+ size 1082696890
run-3/checkpoint-328/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab8cf7c093dc91f12ea4d3e9b8f2616776606719d84c6266df137b0642fa1b65
3
+ size 14244
run-3/checkpoint-328/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afd2b4a0ff87ce2b1e4da58af48f5b67e7d33011949b0ef45dba2421ea1449dc
3
+ size 1064
run-3/checkpoint-328/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-3/checkpoint-328/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-328/tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": false,
47
+ "mask_token": "[MASK]",
48
+ "model_max_length": 512,
49
+ "pad_token": "[PAD]",
50
+ "sep_token": "[SEP]",
51
+ "strip_accents": null,
52
+ "tokenize_chinese_chars": true,
53
+ "tokenizer_class": "DistilBertTokenizer",
54
+ "unk_token": "[UNK]"
55
+ }
run-3/checkpoint-328/trainer_state.json ADDED
@@ -0,0 +1,310 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8722264728385616,
3
+ "best_model_checkpoint": "distilbert-base-multilingual-cased-aoe-hyper/run-3/checkpoint-328",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 328,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12195121951219512,
13
+ "grad_norm": 1.4140416383743286,
14
+ "learning_rate": 9.720486892282597e-06,
15
+ "loss": 0.5947,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.24390243902439024,
20
+ "grad_norm": 0.7245877981185913,
21
+ "learning_rate": 9.477474719975531e-06,
22
+ "loss": 0.5154,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.36585365853658536,
27
+ "grad_norm": 0.9261638522148132,
28
+ "learning_rate": 9.234462547668467e-06,
29
+ "loss": 0.479,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.4878048780487805,
34
+ "grad_norm": 1.6241005659103394,
35
+ "learning_rate": 8.991450375361402e-06,
36
+ "loss": 0.4377,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.6097560975609756,
41
+ "grad_norm": 1.7590631246566772,
42
+ "learning_rate": 8.748438203054338e-06,
43
+ "loss": 0.4112,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.7317073170731707,
48
+ "grad_norm": 2.0623185634613037,
49
+ "learning_rate": 8.505426030747272e-06,
50
+ "loss": 0.3754,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.8536585365853658,
55
+ "grad_norm": 1.2114999294281006,
56
+ "learning_rate": 8.262413858440207e-06,
57
+ "loss": 0.3731,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.975609756097561,
62
+ "grad_norm": 1.9598636627197266,
63
+ "learning_rate": 8.019401686133143e-06,
64
+ "loss": 0.3791,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 1.0,
69
+ "eval_accuracy": 0.8607498087222647,
70
+ "eval_f1": 0.7225609756097561,
71
+ "eval_loss": 0.33913305401802063,
72
+ "eval_precision": 0.688953488372093,
73
+ "eval_recall": 0.7596153846153846,
74
+ "eval_runtime": 4.8898,
75
+ "eval_samples_per_second": 267.29,
76
+ "eval_steps_per_second": 16.77,
77
+ "step": 82
78
+ },
79
+ {
80
+ "epoch": 1.0975609756097562,
81
+ "grad_norm": 1.7537786960601807,
82
+ "learning_rate": 7.776389513826078e-06,
83
+ "loss": 0.3125,
84
+ "step": 90
85
+ },
86
+ {
87
+ "epoch": 1.2195121951219512,
88
+ "grad_norm": 4.089900493621826,
89
+ "learning_rate": 7.533377341519012e-06,
90
+ "loss": 0.332,
91
+ "step": 100
92
+ },
93
+ {
94
+ "epoch": 1.3414634146341464,
95
+ "grad_norm": 1.488580346107483,
96
+ "learning_rate": 7.290365169211947e-06,
97
+ "loss": 0.286,
98
+ "step": 110
99
+ },
100
+ {
101
+ "epoch": 1.4634146341463414,
102
+ "grad_norm": 1.6245107650756836,
103
+ "learning_rate": 7.0473529969048826e-06,
104
+ "loss": 0.2952,
105
+ "step": 120
106
+ },
107
+ {
108
+ "epoch": 1.5853658536585367,
109
+ "grad_norm": 2.5228676795959473,
110
+ "learning_rate": 6.804340824597818e-06,
111
+ "loss": 0.2956,
112
+ "step": 130
113
+ },
114
+ {
115
+ "epoch": 1.7073170731707317,
116
+ "grad_norm": 5.439332008361816,
117
+ "learning_rate": 6.561328652290753e-06,
118
+ "loss": 0.2699,
119
+ "step": 140
120
+ },
121
+ {
122
+ "epoch": 1.8292682926829267,
123
+ "grad_norm": 6.190291404724121,
124
+ "learning_rate": 6.318316479983688e-06,
125
+ "loss": 0.2851,
126
+ "step": 150
127
+ },
128
+ {
129
+ "epoch": 1.951219512195122,
130
+ "grad_norm": 2.851015567779541,
131
+ "learning_rate": 6.075304307676622e-06,
132
+ "loss": 0.2729,
133
+ "step": 160
134
+ },
135
+ {
136
+ "epoch": 2.0,
137
+ "eval_accuracy": 0.8684009181331293,
138
+ "eval_f1": 0.7329192546583851,
139
+ "eval_loss": 0.29830601811408997,
140
+ "eval_precision": 0.7108433734939759,
141
+ "eval_recall": 0.7564102564102564,
142
+ "eval_runtime": 4.8964,
143
+ "eval_samples_per_second": 266.931,
144
+ "eval_steps_per_second": 16.747,
145
+ "step": 164
146
+ },
147
+ {
148
+ "epoch": 2.073170731707317,
149
+ "grad_norm": 1.8465652465820312,
150
+ "learning_rate": 5.832292135369558e-06,
151
+ "loss": 0.2559,
152
+ "step": 170
153
+ },
154
+ {
155
+ "epoch": 2.1951219512195124,
156
+ "grad_norm": 1.781294822692871,
157
+ "learning_rate": 5.5892799630624935e-06,
158
+ "loss": 0.2523,
159
+ "step": 180
160
+ },
161
+ {
162
+ "epoch": 2.317073170731707,
163
+ "grad_norm": 4.9645256996154785,
164
+ "learning_rate": 5.346267790755428e-06,
165
+ "loss": 0.2737,
166
+ "step": 190
167
+ },
168
+ {
169
+ "epoch": 2.4390243902439024,
170
+ "grad_norm": 2.3662397861480713,
171
+ "learning_rate": 5.103255618448363e-06,
172
+ "loss": 0.2607,
173
+ "step": 200
174
+ },
175
+ {
176
+ "epoch": 2.5609756097560976,
177
+ "grad_norm": 3.443523645401001,
178
+ "learning_rate": 4.8602434461412986e-06,
179
+ "loss": 0.2301,
180
+ "step": 210
181
+ },
182
+ {
183
+ "epoch": 2.682926829268293,
184
+ "grad_norm": 4.988992691040039,
185
+ "learning_rate": 4.617231273834233e-06,
186
+ "loss": 0.2451,
187
+ "step": 220
188
+ },
189
+ {
190
+ "epoch": 2.8048780487804876,
191
+ "grad_norm": 6.013027667999268,
192
+ "learning_rate": 4.374219101527169e-06,
193
+ "loss": 0.2386,
194
+ "step": 230
195
+ },
196
+ {
197
+ "epoch": 2.926829268292683,
198
+ "grad_norm": 4.056751728057861,
199
+ "learning_rate": 4.131206929220104e-06,
200
+ "loss": 0.236,
201
+ "step": 240
202
+ },
203
+ {
204
+ "epoch": 3.0,
205
+ "eval_accuracy": 0.8661055853098699,
206
+ "eval_f1": 0.7422680412371134,
207
+ "eval_loss": 0.30293241143226624,
208
+ "eval_precision": 0.6866485013623979,
209
+ "eval_recall": 0.8076923076923077,
210
+ "eval_runtime": 4.9202,
211
+ "eval_samples_per_second": 265.637,
212
+ "eval_steps_per_second": 16.666,
213
+ "step": 246
214
+ },
215
+ {
216
+ "epoch": 3.048780487804878,
217
+ "grad_norm": 5.498621940612793,
218
+ "learning_rate": 3.888194756913039e-06,
219
+ "loss": 0.2761,
220
+ "step": 250
221
+ },
222
+ {
223
+ "epoch": 3.1707317073170733,
224
+ "grad_norm": 2.0462112426757812,
225
+ "learning_rate": 3.6451825846059735e-06,
226
+ "loss": 0.2074,
227
+ "step": 260
228
+ },
229
+ {
230
+ "epoch": 3.292682926829268,
231
+ "grad_norm": 4.663333892822266,
232
+ "learning_rate": 3.402170412298909e-06,
233
+ "loss": 0.2357,
234
+ "step": 270
235
+ },
236
+ {
237
+ "epoch": 3.4146341463414633,
238
+ "grad_norm": 4.013532638549805,
239
+ "learning_rate": 3.159158239991844e-06,
240
+ "loss": 0.2385,
241
+ "step": 280
242
+ },
243
+ {
244
+ "epoch": 3.5365853658536586,
245
+ "grad_norm": 4.440213680267334,
246
+ "learning_rate": 2.916146067684779e-06,
247
+ "loss": 0.226,
248
+ "step": 290
249
+ },
250
+ {
251
+ "epoch": 3.658536585365854,
252
+ "grad_norm": 4.662286758422852,
253
+ "learning_rate": 2.673133895377714e-06,
254
+ "loss": 0.2129,
255
+ "step": 300
256
+ },
257
+ {
258
+ "epoch": 3.7804878048780486,
259
+ "grad_norm": 4.345000267028809,
260
+ "learning_rate": 2.4301217230706493e-06,
261
+ "loss": 0.2366,
262
+ "step": 310
263
+ },
264
+ {
265
+ "epoch": 3.902439024390244,
266
+ "grad_norm": 2.3199422359466553,
267
+ "learning_rate": 2.1871095507635844e-06,
268
+ "loss": 0.2462,
269
+ "step": 320
270
+ },
271
+ {
272
+ "epoch": 4.0,
273
+ "eval_accuracy": 0.8722264728385616,
274
+ "eval_f1": 0.7481146304675717,
275
+ "eval_loss": 0.3029697835445404,
276
+ "eval_precision": 0.7065527065527065,
277
+ "eval_recall": 0.7948717948717948,
278
+ "eval_runtime": 4.8827,
279
+ "eval_samples_per_second": 267.68,
280
+ "eval_steps_per_second": 16.794,
281
+ "step": 328
282
+ }
283
+ ],
284
+ "logging_steps": 10,
285
+ "max_steps": 410,
286
+ "num_input_tokens_seen": 0,
287
+ "num_train_epochs": 5,
288
+ "save_steps": 500,
289
+ "stateful_callbacks": {
290
+ "TrainerControl": {
291
+ "args": {
292
+ "should_epoch_stop": false,
293
+ "should_evaluate": false,
294
+ "should_log": false,
295
+ "should_save": true,
296
+ "should_training_stop": false
297
+ },
298
+ "attributes": {}
299
+ }
300
+ },
301
+ "total_flos": 2704984280555520.0,
302
+ "train_batch_size": 64,
303
+ "trial_name": null,
304
+ "trial_params": {
305
+ "learning_rate": 9.963499064589662e-06,
306
+ "num_train_epochs": 5,
307
+ "per_device_train_batch_size": 64,
308
+ "seed": 9
309
+ }
310
+ }
run-3/checkpoint-328/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ba0abad364ab2cb40de43a35618f46f727f31cf408908221b491914804b79c4
3
+ size 5304
run-3/checkpoint-328/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-1308/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-multilingual-cased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "initializer_range": 0.02,
12
+ "max_position_embeddings": 512,
13
+ "model_type": "distilbert",
14
+ "n_heads": 12,
15
+ "n_layers": 6,
16
+ "output_past": true,
17
+ "pad_token_id": 0,
18
+ "problem_type": "single_label_classification",
19
+ "qa_dropout": 0.1,
20
+ "seq_classif_dropout": 0.2,
21
+ "sinusoidal_pos_embds": false,
22
+ "tie_weights_": true,
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.46.2",
25
+ "vocab_size": 119547
26
+ }
run-4/checkpoint-1308/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce10587e7c563cda180ab83457698a1bc0b6aecbac909c89e4d1514e9a925734
3
+ size 541317368
run-4/checkpoint-1308/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:443a8c1523e68cdf482c192fd0e444a65b55c1027d1aac986b0a5aefd4724d47
3
+ size 1082696890