Hugging Face's logo

Slava
/
tiny-bert-sst2-distilled

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Slava commited on
Commit
42b93ed
·
verified ·
1 Parent(s): f1bd202

Training in progress, epoch 1

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. logs/events.out.tfevents.1706181411.A-FVFFG2C4Q05P.18450.3 +2 -2
  2. logs/events.out.tfevents.1706189038.A-FVFFG2C4Q05P.18450.4 +3 -0
  3. logs/events.out.tfevents.1706196200.A-FVFFG2C4Q05P.18450.5 +3 -0
  4. model.safetensors +1 -1
  5. run-2/checkpoint-2108/config.json +34 -0
  6. run-2/checkpoint-2108/model.safetensors +3 -0
  7. run-2/checkpoint-2108/optimizer.pt +3 -0
  8. run-2/checkpoint-2108/rng_state.pth +3 -0
  9. run-2/checkpoint-2108/scheduler.pt +3 -0
  10. run-2/checkpoint-2108/special_tokens_map.json +7 -0
  11. run-2/checkpoint-2108/tokenizer.json +0 -0
  12. run-2/checkpoint-2108/tokenizer_config.json +57 -0
  13. run-2/checkpoint-2108/trainer_state.json +86 -0
  14. run-2/checkpoint-2108/training_args.bin +3 -0
  15. run-2/checkpoint-2108/vocab.txt +0 -0
  16. run-2/checkpoint-3689/config.json +34 -0
  17. run-2/checkpoint-3689/model.safetensors +3 -0
  18. run-2/checkpoint-3689/optimizer.pt +3 -0
  19. run-2/checkpoint-3689/rng_state.pth +3 -0
  20. run-2/checkpoint-3689/scheduler.pt +3 -0
  21. run-2/checkpoint-3689/special_tokens_map.json +7 -0
  22. run-2/checkpoint-3689/tokenizer.json +0 -0
  23. run-2/checkpoint-3689/tokenizer_config.json +57 -0
  24. run-2/checkpoint-3689/trainer_state.json +131 -0
  25. run-2/checkpoint-3689/training_args.bin +3 -0
  26. run-2/checkpoint-3689/vocab.txt +0 -0
  27. run-3/checkpoint-2635/config.json +34 -0
  28. run-3/checkpoint-2635/model.safetensors +3 -0
  29. run-3/checkpoint-2635/optimizer.pt +3 -0
  30. run-3/checkpoint-2635/rng_state.pth +3 -0
  31. run-3/checkpoint-2635/scheduler.pt +3 -0
  32. run-3/checkpoint-2635/special_tokens_map.json +7 -0
  33. run-3/checkpoint-2635/tokenizer.json +0 -0
  34. run-3/checkpoint-2635/tokenizer_config.json +57 -0
  35. run-3/checkpoint-2635/trainer_state.json +101 -0
  36. run-3/checkpoint-2635/training_args.bin +3 -0
  37. run-3/checkpoint-2635/vocab.txt +0 -0
  38. run-3/checkpoint-527/config.json +34 -0
  39. run-3/checkpoint-527/model.safetensors +3 -0
  40. run-3/checkpoint-527/optimizer.pt +3 -0
  41. run-3/checkpoint-527/rng_state.pth +3 -0
  42. run-3/checkpoint-527/scheduler.pt +3 -0
  43. run-3/checkpoint-527/special_tokens_map.json +7 -0
  44. run-3/checkpoint-527/tokenizer.json +0 -0
  45. run-3/checkpoint-527/tokenizer_config.json +57 -0
  46. run-3/checkpoint-527/trainer_state.json +41 -0
  47. run-3/checkpoint-527/training_args.bin +3 -0
  48. run-3/checkpoint-527/vocab.txt +0 -0
  49. run-4/checkpoint-527/config.json +34 -0
  50. run-4/checkpoint-527/model.safetensors +3 -0
logs/events.out.tfevents.1706181411.A-FVFFG2C4Q05P.18450.3 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1a846a378b4ba152a38f2ee7d514086d755d7154e548a2ace0cac942fa771be9
3
- size 5396
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba53916c1de06b8b9e3bf5372f7df77dce88d1e1ca976d5a723e5138bb08ace8
3
+ size 8150
logs/events.out.tfevents.1706189038.A-FVFFG2C4Q05P.18450.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed792ecc848af69afb8ff6164744104da56d6f72fb5446eefe50e390800ece89
3
+ size 7190
logs/events.out.tfevents.1706196200.A-FVFFG2C4Q05P.18450.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53b33c54d05ee0e72b3ad04a32dc40448bf1c1c65597a97ff3edeb33a16ee5c9
3
+ size 4915
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:395d28d0adf7ece98dfdd310bc63f02d8bcd77ca2a202581e35924db72b1b4d3
3
  size 17549312
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:deb8aa05f92c3e3e3100697f91476432307ef227d91e417e359f633516f53ade
3
  size 17549312
run-2/checkpoint-2108/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "positive"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 512,
17
+ "label2id": {
18
+ "negative": "0",
19
+ "positive": "1"
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 2,
25
+ "num_hidden_layers": 2,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.37.1",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 30522
34
+ }
run-2/checkpoint-2108/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:059957aafba48a0874192cdd9adfa8134d4b6e1d4b526ccdc3daf3e95d060417
3
+ size 17549312
run-2/checkpoint-2108/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:800621d9b455b609e2a1673e9ef3bda30026ba964354572696b863407bed623f
3
+ size 35122373
run-2/checkpoint-2108/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2aae3bffb74208d13dfae4d8b66eb00427742bd84752ee081ae0f135d90f179b
3
+ size 13617
run-2/checkpoint-2108/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46f542e15b9cbe13d4728c333b67a5c6dac10a3d94a471014056774cbe2dee00
3
+ size 627
run-2/checkpoint-2108/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-2108/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-2108/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
run-2/checkpoint-2108/trainer_state.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8325688073394495,
3
+ "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-2108",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2108,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 0.00013462097079767568,
14
+ "loss": 1.19,
15
+ "step": 527
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.819954128440367,
20
+ "eval_loss": 1.0689420700073242,
21
+ "eval_runtime": 10.3665,
22
+ "eval_samples_per_second": 84.117,
23
+ "eval_steps_per_second": 0.675,
24
+ "step": 527
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "learning_rate": 0.00011218414233139642,
29
+ "loss": 0.5832,
30
+ "step": 1054
31
+ },
32
+ {
33
+ "epoch": 2.0,
34
+ "eval_accuracy": 0.8279816513761468,
35
+ "eval_loss": 1.1704012155532837,
36
+ "eval_runtime": 6.2773,
37
+ "eval_samples_per_second": 138.912,
38
+ "eval_steps_per_second": 1.115,
39
+ "step": 1054
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "learning_rate": 8.974731386511713e-05,
44
+ "loss": 0.4391,
45
+ "step": 1581
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "eval_accuracy": 0.823394495412844,
50
+ "eval_loss": 1.1875587701797485,
51
+ "eval_runtime": 6.2138,
52
+ "eval_samples_per_second": 140.333,
53
+ "eval_steps_per_second": 1.127,
54
+ "step": 1581
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "learning_rate": 6.731048539883784e-05,
59
+ "loss": 0.3642,
60
+ "step": 2108
61
+ },
62
+ {
63
+ "epoch": 4.0,
64
+ "eval_accuracy": 0.8325688073394495,
65
+ "eval_loss": 1.1609015464782715,
66
+ "eval_runtime": 8.1221,
67
+ "eval_samples_per_second": 107.361,
68
+ "eval_steps_per_second": 0.862,
69
+ "step": 2108
70
+ }
71
+ ],
72
+ "logging_steps": 500,
73
+ "max_steps": 3689,
74
+ "num_input_tokens_seen": 0,
75
+ "num_train_epochs": 7,
76
+ "save_steps": 500,
77
+ "total_flos": 32344049138640.0,
78
+ "train_batch_size": 128,
79
+ "trial_name": null,
80
+ "trial_params": {
81
+ "alpha": 0.6204520709084631,
82
+ "learning_rate": 0.00015705779926395499,
83
+ "num_train_epochs": 7,
84
+ "temperature": 16
85
+ }
86
+ }
run-2/checkpoint-2108/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b4be0222258df26817e91fe31755ef0f0e5b0b1423f1be49de3cd924d2cf509
3
+ size 4283
run-2/checkpoint-2108/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-3689/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "positive"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 512,
17
+ "label2id": {
18
+ "negative": "0",
19
+ "positive": "1"
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 2,
25
+ "num_hidden_layers": 2,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.37.1",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 30522
34
+ }
run-2/checkpoint-3689/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6343018918e71362934dc01166336ef78fabf6dc5eb527c32df011bc4923cff
3
+ size 17549312
run-2/checkpoint-3689/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73aa6ea179fb648e4fbd8572af5ac4894a38924917ea67eecfcb26ccade89514
3
+ size 35122373
run-2/checkpoint-3689/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1944a0f816099340e1dbf1a93e033a5a3a4bb0c92f096d7559d0edb8afc135de
3
+ size 13617
run-2/checkpoint-3689/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f1a2d7f6656eeeaa378960e23366ac79360db78a37b3d208225a1a1fbe644fa
3
+ size 627
run-2/checkpoint-3689/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-3689/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-3689/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
run-2/checkpoint-3689/trainer_state.json ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8325688073394495,
3
+ "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-2108",
4
+ "epoch": 7.0,
5
+ "eval_steps": 500,
6
+ "global_step": 3689,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 0.00013462097079767568,
14
+ "loss": 1.19,
15
+ "step": 527
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.819954128440367,
20
+ "eval_loss": 1.0689420700073242,
21
+ "eval_runtime": 10.3665,
22
+ "eval_samples_per_second": 84.117,
23
+ "eval_steps_per_second": 0.675,
24
+ "step": 527
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "learning_rate": 0.00011218414233139642,
29
+ "loss": 0.5832,
30
+ "step": 1054
31
+ },
32
+ {
33
+ "epoch": 2.0,
34
+ "eval_accuracy": 0.8279816513761468,
35
+ "eval_loss": 1.1704012155532837,
36
+ "eval_runtime": 6.2773,
37
+ "eval_samples_per_second": 138.912,
38
+ "eval_steps_per_second": 1.115,
39
+ "step": 1054
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "learning_rate": 8.974731386511713e-05,
44
+ "loss": 0.4391,
45
+ "step": 1581
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "eval_accuracy": 0.823394495412844,
50
+ "eval_loss": 1.1875587701797485,
51
+ "eval_runtime": 6.2138,
52
+ "eval_samples_per_second": 140.333,
53
+ "eval_steps_per_second": 1.127,
54
+ "step": 1581
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "learning_rate": 6.731048539883784e-05,
59
+ "loss": 0.3642,
60
+ "step": 2108
61
+ },
62
+ {
63
+ "epoch": 4.0,
64
+ "eval_accuracy": 0.8325688073394495,
65
+ "eval_loss": 1.1609015464782715,
66
+ "eval_runtime": 8.1221,
67
+ "eval_samples_per_second": 107.361,
68
+ "eval_steps_per_second": 0.862,
69
+ "step": 2108
70
+ },
71
+ {
72
+ "epoch": 5.0,
73
+ "learning_rate": 4.4873656932558565e-05,
74
+ "loss": 0.3145,
75
+ "step": 2635
76
+ },
77
+ {
78
+ "epoch": 5.0,
79
+ "eval_accuracy": 0.823394495412844,
80
+ "eval_loss": 1.2187858819961548,
81
+ "eval_runtime": 8.576,
82
+ "eval_samples_per_second": 101.679,
83
+ "eval_steps_per_second": 0.816,
84
+ "step": 2635
85
+ },
86
+ {
87
+ "epoch": 6.0,
88
+ "learning_rate": 2.2436828466279283e-05,
89
+ "loss": 0.2816,
90
+ "step": 3162
91
+ },
92
+ {
93
+ "epoch": 6.0,
94
+ "eval_accuracy": 0.8268348623853211,
95
+ "eval_loss": 1.215812087059021,
96
+ "eval_runtime": 9.7324,
97
+ "eval_samples_per_second": 89.598,
98
+ "eval_steps_per_second": 0.719,
99
+ "step": 3162
100
+ },
101
+ {
102
+ "epoch": 7.0,
103
+ "learning_rate": 0.0,
104
+ "loss": 0.2599,
105
+ "step": 3689
106
+ },
107
+ {
108
+ "epoch": 7.0,
109
+ "eval_accuracy": 0.8256880733944955,
110
+ "eval_loss": 1.249680519104004,
111
+ "eval_runtime": 9.3708,
112
+ "eval_samples_per_second": 93.055,
113
+ "eval_steps_per_second": 0.747,
114
+ "step": 3689
115
+ }
116
+ ],
117
+ "logging_steps": 500,
118
+ "max_steps": 3689,
119
+ "num_input_tokens_seen": 0,
120
+ "num_train_epochs": 7,
121
+ "save_steps": 500,
122
+ "total_flos": 56642674579560.0,
123
+ "train_batch_size": 128,
124
+ "trial_name": null,
125
+ "trial_params": {
126
+ "alpha": 0.6204520709084631,
127
+ "learning_rate": 0.00015705779926395499,
128
+ "num_train_epochs": 7,
129
+ "temperature": 16
130
+ }
131
+ }
run-2/checkpoint-3689/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b4be0222258df26817e91fe31755ef0f0e5b0b1423f1be49de3cd924d2cf509
3
+ size 4283
run-2/checkpoint-3689/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-2635/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "positive"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 512,
17
+ "label2id": {
18
+ "negative": "0",
19
+ "positive": "1"
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 2,
25
+ "num_hidden_layers": 2,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.37.1",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 30522
34
+ }
run-3/checkpoint-2635/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f06a2e462faf95a7a6dfe7b2df811a907b6b9c5ea691a5d849433067ca082183
3
+ size 17549312
run-3/checkpoint-2635/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8dc7d7662f62738e0ccd547c15022f154f503da4201760caabfa5920412fade0
3
+ size 35122373
run-3/checkpoint-2635/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb154c82007a8758ab9f9ae31d1389cae06c1f2182ea0779f330182347f52f96
3
+ size 13617
run-3/checkpoint-2635/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f8e46e076b308dd2c6d09606b687c832bccc97292bc86eb495c55b072d3f5aa
3
+ size 627
run-3/checkpoint-2635/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-2635/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-2635/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
run-3/checkpoint-2635/trainer_state.json ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8256880733944955,
3
+ "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-527",
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2635,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 0.0002646598360783595,
14
+ "loss": 0.4864,
15
+ "step": 527
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.8256880733944955,
20
+ "eval_loss": 0.5460890531539917,
21
+ "eval_runtime": 6.1659,
22
+ "eval_samples_per_second": 141.424,
23
+ "eval_steps_per_second": 1.135,
24
+ "step": 527
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "learning_rate": 0.00019849487705876962,
29
+ "loss": 0.2623,
30
+ "step": 1054
31
+ },
32
+ {
33
+ "epoch": 2.0,
34
+ "eval_accuracy": 0.8188073394495413,
35
+ "eval_loss": 0.7069737315177917,
36
+ "eval_runtime": 7.2395,
37
+ "eval_samples_per_second": 120.451,
38
+ "eval_steps_per_second": 0.967,
39
+ "step": 1054
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "learning_rate": 0.00013232991803917974,
44
+ "loss": 0.1902,
45
+ "step": 1581
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "eval_accuracy": 0.823394495412844,
50
+ "eval_loss": 0.7423160076141357,
51
+ "eval_runtime": 7.2214,
52
+ "eval_samples_per_second": 120.753,
53
+ "eval_steps_per_second": 0.969,
54
+ "step": 1581
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "learning_rate": 6.616495901958987e-05,
59
+ "loss": 0.1515,
60
+ "step": 2108
61
+ },
62
+ {
63
+ "epoch": 4.0,
64
+ "eval_accuracy": 0.819954128440367,
65
+ "eval_loss": 0.7632457613945007,
66
+ "eval_runtime": 7.0986,
67
+ "eval_samples_per_second": 122.841,
68
+ "eval_steps_per_second": 0.986,
69
+ "step": 2108
70
+ },
71
+ {
72
+ "epoch": 5.0,
73
+ "learning_rate": 0.0,
74
+ "loss": 0.1242,
75
+ "step": 2635
76
+ },
77
+ {
78
+ "epoch": 5.0,
79
+ "eval_accuracy": 0.8222477064220184,
80
+ "eval_loss": 0.8318789601325989,
81
+ "eval_runtime": 7.1337,
82
+ "eval_samples_per_second": 122.236,
83
+ "eval_steps_per_second": 0.981,
84
+ "step": 2635
85
+ }
86
+ ],
87
+ "logging_steps": 500,
88
+ "max_steps": 2635,
89
+ "num_input_tokens_seen": 0,
90
+ "num_train_epochs": 5,
91
+ "save_steps": 500,
92
+ "total_flos": 40436714110320.0,
93
+ "train_batch_size": 128,
94
+ "trial_name": null,
95
+ "trial_params": {
96
+ "alpha": 0.9299267464979704,
97
+ "learning_rate": 0.00033082479509794936,
98
+ "num_train_epochs": 5,
99
+ "temperature": 11
100
+ }
101
+ }
run-3/checkpoint-2635/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db4a3bf2450a26eff9bfb630b2c96a3526f6ad8689cd760892210e4dc034be42
3
+ size 4283
run-3/checkpoint-2635/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-527/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "positive"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 512,
17
+ "label2id": {
18
+ "negative": "0",
19
+ "positive": "1"
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 2,
25
+ "num_hidden_layers": 2,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.37.1",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 30522
34
+ }
run-3/checkpoint-527/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fea5a3a721533db0cf8666635e2ae9961df8bbe1b0aba043f68650fcd34fb6af
3
+ size 17549312
run-3/checkpoint-527/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:007778dbd50ca906a55463f7fb2ea0b2484e05afc4f460fdf01931038f920f55
3
+ size 35122373
run-3/checkpoint-527/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f943a8d8f93fea7f44cad3b9bfb2d2b97d65b141eb704b9cba669cf42e776b14
3
+ size 13617
run-3/checkpoint-527/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d000fd306f2a2387119376b2a7ce6a310cafb331ef764c119e50e54e29df6617
3
+ size 627
run-3/checkpoint-527/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-527/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-527/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "never_split": null,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "strip_accents": null,
54
+ "tokenize_chinese_chars": true,
55
+ "tokenizer_class": "BertTokenizer",
56
+ "unk_token": "[UNK]"
57
+ }
run-3/checkpoint-527/trainer_state.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8256880733944955,
3
+ "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-527",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 527,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 0.0002646598360783595,
14
+ "loss": 0.4864,
15
+ "step": 527
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.8256880733944955,
20
+ "eval_loss": 0.5460890531539917,
21
+ "eval_runtime": 6.1659,
22
+ "eval_samples_per_second": 141.424,
23
+ "eval_steps_per_second": 1.135,
24
+ "step": 527
25
+ }
26
+ ],
27
+ "logging_steps": 500,
28
+ "max_steps": 2635,
29
+ "num_input_tokens_seen": 0,
30
+ "num_train_epochs": 5,
31
+ "save_steps": 500,
32
+ "total_flos": 8069051778960.0,
33
+ "train_batch_size": 128,
34
+ "trial_name": null,
35
+ "trial_params": {
36
+ "alpha": 0.9299267464979704,
37
+ "learning_rate": 0.00033082479509794936,
38
+ "num_train_epochs": 5,
39
+ "temperature": 11
40
+ }
41
+ }
run-3/checkpoint-527/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db4a3bf2450a26eff9bfb630b2c96a3526f6ad8689cd760892210e4dc034be42
3
+ size 4283
run-3/checkpoint-527/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-527/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/bert_uncased_L-2_H-128_A-2",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 128,
11
+ "id2label": {
12
+ "0": "negative",
13
+ "1": "positive"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 512,
17
+ "label2id": {
18
+ "negative": "0",
19
+ "positive": "1"
20
+ },
21
+ "layer_norm_eps": 1e-12,
22
+ "max_position_embeddings": 512,
23
+ "model_type": "bert",
24
+ "num_attention_heads": 2,
25
+ "num_hidden_layers": 2,
26
+ "pad_token_id": 0,
27
+ "position_embedding_type": "absolute",
28
+ "problem_type": "single_label_classification",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.37.1",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 30522
34
+ }
run-4/checkpoint-527/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:deb8aa05f92c3e3e3100697f91476432307ef227d91e417e359f633516f53ade
3
+ size 17549312