dzungpham commited on
Commit
ceed5bc
·
verified ·
1 Parent(s): 2a2618e

upload model weights after 2400 training steps with frozen pretrained weights, this performs worse than checkpoint 1000 on test set T_T

Browse files
Files changed (27) hide show
  1. codebert-base-cross-entropy/checkpoint-2000/config.json +27 -0
  2. codebert-base-cross-entropy/checkpoint-2000/merges.txt +0 -0
  3. codebert-base-cross-entropy/checkpoint-2000/model.safetensors +3 -0
  4. codebert-base-cross-entropy/checkpoint-2000/optimizer.pt +3 -0
  5. codebert-base-cross-entropy/checkpoint-2000/rng_state.pth +3 -0
  6. codebert-base-cross-entropy/checkpoint-2000/scaler.pt +3 -0
  7. codebert-base-cross-entropy/checkpoint-2000/scheduler.pt +3 -0
  8. codebert-base-cross-entropy/checkpoint-2000/special_tokens_map.json +51 -0
  9. codebert-base-cross-entropy/checkpoint-2000/tokenizer.json +0 -0
  10. codebert-base-cross-entropy/checkpoint-2000/tokenizer_config.json +58 -0
  11. codebert-base-cross-entropy/checkpoint-2000/trainer_state.json +1539 -0
  12. codebert-base-cross-entropy/checkpoint-2000/training_args.bin +3 -0
  13. codebert-base-cross-entropy/checkpoint-2000/vocab.json +0 -0
  14. codebert-base-cross-entropy/checkpoint-2400/config.json +27 -0
  15. codebert-base-cross-entropy/checkpoint-2400/merges.txt +0 -0
  16. codebert-base-cross-entropy/checkpoint-2400/model.safetensors +3 -0
  17. codebert-base-cross-entropy/checkpoint-2400/optimizer.pt +3 -0
  18. codebert-base-cross-entropy/checkpoint-2400/rng_state.pth +3 -0
  19. codebert-base-cross-entropy/checkpoint-2400/scaler.pt +3 -0
  20. codebert-base-cross-entropy/checkpoint-2400/scheduler.pt +3 -0
  21. codebert-base-cross-entropy/checkpoint-2400/special_tokens_map.json +51 -0
  22. codebert-base-cross-entropy/checkpoint-2400/tokenizer.json +0 -0
  23. codebert-base-cross-entropy/checkpoint-2400/tokenizer_config.json +58 -0
  24. codebert-base-cross-entropy/checkpoint-2400/trainer_state.json +1819 -0
  25. codebert-base-cross-entropy/checkpoint-2400/training_args.bin +3 -0
  26. codebert-base-cross-entropy/checkpoint-2400/vocab.json +0 -0
  27. codebert-base-cross-entropy/training.log +324 -0
codebert-base-cross-entropy/checkpoint-2000/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "RobertaForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": null,
8
+ "dtype": "float32",
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "transformers_version": "4.56.0",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 50265
27
+ }
codebert-base-cross-entropy/checkpoint-2000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e787ee99242597f26f290dd5b637c78bacf3c4fe4ac1c5917071082cedf3aa5
3
+ size 498612824
codebert-base-cross-entropy/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f883eedadd2b52c8954fcbb301c7e3cf1b57fd6cc8147b038ec200d2f69ddd3
3
+ size 4741859
codebert-base-cross-entropy/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf2a7a329bb1127b2d8fe243a3acf7994f7043d7dae233d422c0867b9d5cad9d
3
+ size 14645
codebert-base-cross-entropy/checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f8f63d73baf60e5f7e451eaeac214d86c542fa22a61816c566223b9d6dd4bb0
3
+ size 1383
codebert-base-cross-entropy/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dbfea7e105a56a1e972a5a173ec2b16f667b010240d5832ef2b86aff88f86006
3
+ size 1465
codebert-base-cross-entropy/checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
codebert-base-cross-entropy/checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "<mask>",
52
+ "model_max_length": 512,
53
+ "pad_token": "<pad>",
54
+ "sep_token": "</s>",
55
+ "tokenizer_class": "RobertaTokenizer",
56
+ "trim_offsets": true,
57
+ "unk_token": "<unk>"
58
+ }
codebert-base-cross-entropy/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,1539 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2000,
3
+ "best_metric": 0.743516265066894,
4
+ "best_model_checkpoint": "./output_checkpoints/codebert-base-cross-entropy/checkpoint-2000",
5
+ "epoch": 0.2559836170485089,
6
+ "eval_steps": 500,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0012799180852425445,
14
+ "grad_norm": 3.6954128742218018,
15
+ "learning_rate": 3.840409643695328e-08,
16
+ "loss": 0.7763,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.002559836170485089,
21
+ "grad_norm": 3.3660645484924316,
22
+ "learning_rate": 8.10753147002347e-08,
23
+ "loss": 0.7763,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0038397542557276334,
28
+ "grad_norm": 4.491067886352539,
29
+ "learning_rate": 1.2374653296351612e-07,
30
+ "loss": 0.7889,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.005119672340970178,
35
+ "grad_norm": 1.5920521020889282,
36
+ "learning_rate": 1.6641775122679754e-07,
37
+ "loss": 0.7888,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.006399590426212722,
42
+ "grad_norm": 4.410898208618164,
43
+ "learning_rate": 2.0908896949007894e-07,
44
+ "loss": 0.7692,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.007679508511455267,
49
+ "grad_norm": 2.5669264793395996,
50
+ "learning_rate": 2.517601877533604e-07,
51
+ "loss": 0.7735,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.008959426596697812,
56
+ "grad_norm": 6.190986156463623,
57
+ "learning_rate": 2.944314060166418e-07,
58
+ "loss": 0.7679,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.010239344681940356,
63
+ "grad_norm": 5.736413478851318,
64
+ "learning_rate": 3.371026242799232e-07,
65
+ "loss": 0.744,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.011519262767182901,
70
+ "grad_norm": 4.278866767883301,
71
+ "learning_rate": 3.7977384254320464e-07,
72
+ "loss": 0.7542,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.012799180852425445,
77
+ "grad_norm": 5.565455436706543,
78
+ "learning_rate": 4.22445060806486e-07,
79
+ "loss": 0.752,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.01407909893766799,
84
+ "grad_norm": 4.031781196594238,
85
+ "learning_rate": 4.651162790697675e-07,
86
+ "loss": 0.7614,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.015359017022910534,
91
+ "grad_norm": 2.484126329421997,
92
+ "learning_rate": 5.077874973330489e-07,
93
+ "loss": 0.7664,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.016638935108153077,
98
+ "grad_norm": 2.7525155544281006,
99
+ "learning_rate": 5.504587155963304e-07,
100
+ "loss": 0.7521,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.017918853193395624,
105
+ "grad_norm": 1.6374385356903076,
106
+ "learning_rate": 5.931299338596117e-07,
107
+ "loss": 0.7436,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.019198771278638168,
112
+ "grad_norm": 2.370774269104004,
113
+ "learning_rate": 6.358011521228932e-07,
114
+ "loss": 0.7429,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.02047868936388071,
119
+ "grad_norm": 2.8948566913604736,
120
+ "learning_rate": 6.784723703861745e-07,
121
+ "loss": 0.7333,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.021758607449123255,
126
+ "grad_norm": 2.7481186389923096,
127
+ "learning_rate": 7.21143588649456e-07,
128
+ "loss": 0.7294,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.023038525534365802,
133
+ "grad_norm": 2.0159404277801514,
134
+ "learning_rate": 7.638148069127374e-07,
135
+ "loss": 0.731,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.024318443619608346,
140
+ "grad_norm": 3.6898131370544434,
141
+ "learning_rate": 8.064860251760189e-07,
142
+ "loss": 0.7153,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.02559836170485089,
147
+ "grad_norm": 3.2842342853546143,
148
+ "learning_rate": 8.491572434393003e-07,
149
+ "loss": 0.7245,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.02559836170485089,
154
+ "eval_accuracy": 0.47695,
155
+ "eval_loss": 0.7148504853248596,
156
+ "eval_macro_f1": 0.3229290091066048,
157
+ "eval_precision": 0.238475,
158
+ "eval_recall": 0.5,
159
+ "eval_runtime": 1641.324,
160
+ "eval_samples_per_second": 60.926,
161
+ "eval_steps_per_second": 0.476,
162
+ "step": 200
163
+ },
164
+ {
165
+ "epoch": 0.026878279790093433,
166
+ "grad_norm": 2.294558048248291,
167
+ "learning_rate": 8.918284617025817e-07,
168
+ "loss": 0.718,
169
+ "step": 210
170
+ },
171
+ {
172
+ "epoch": 0.02815819787533598,
173
+ "grad_norm": 2.9467175006866455,
174
+ "learning_rate": 9.344996799658632e-07,
175
+ "loss": 0.7076,
176
+ "step": 220
177
+ },
178
+ {
179
+ "epoch": 0.029438115960578524,
180
+ "grad_norm": 2.6889193058013916,
181
+ "learning_rate": 9.771708982291445e-07,
182
+ "loss": 0.7019,
183
+ "step": 230
184
+ },
185
+ {
186
+ "epoch": 0.030718034045821067,
187
+ "grad_norm": 0.7491384744644165,
188
+ "learning_rate": 1.0198421164924258e-06,
189
+ "loss": 0.7046,
190
+ "step": 240
191
+ },
192
+ {
193
+ "epoch": 0.031997952131063614,
194
+ "grad_norm": 2.0363433361053467,
195
+ "learning_rate": 1.0625133347557074e-06,
196
+ "loss": 0.7085,
197
+ "step": 250
198
+ },
199
+ {
200
+ "epoch": 0.033277870216306155,
201
+ "grad_norm": 1.1431337594985962,
202
+ "learning_rate": 1.1051845530189888e-06,
203
+ "loss": 0.7008,
204
+ "step": 260
205
+ },
206
+ {
207
+ "epoch": 0.0345577883015487,
208
+ "grad_norm": 3.7735233306884766,
209
+ "learning_rate": 1.1478557712822702e-06,
210
+ "loss": 0.7015,
211
+ "step": 270
212
+ },
213
+ {
214
+ "epoch": 0.03583770638679125,
215
+ "grad_norm": 0.7889328002929688,
216
+ "learning_rate": 1.1905269895455517e-06,
217
+ "loss": 0.699,
218
+ "step": 280
219
+ },
220
+ {
221
+ "epoch": 0.03711762447203379,
222
+ "grad_norm": 0.8796226382255554,
223
+ "learning_rate": 1.233198207808833e-06,
224
+ "loss": 0.6922,
225
+ "step": 290
226
+ },
227
+ {
228
+ "epoch": 0.038397542557276336,
229
+ "grad_norm": 0.7966652512550354,
230
+ "learning_rate": 1.2758694260721145e-06,
231
+ "loss": 0.6949,
232
+ "step": 300
233
+ },
234
+ {
235
+ "epoch": 0.039677460642518876,
236
+ "grad_norm": 0.7029069066047668,
237
+ "learning_rate": 1.318540644335396e-06,
238
+ "loss": 0.695,
239
+ "step": 310
240
+ },
241
+ {
242
+ "epoch": 0.04095737872776142,
243
+ "grad_norm": 2.04327392578125,
244
+ "learning_rate": 1.3612118625986772e-06,
245
+ "loss": 0.6944,
246
+ "step": 320
247
+ },
248
+ {
249
+ "epoch": 0.04223729681300397,
250
+ "grad_norm": 0.918025553226471,
251
+ "learning_rate": 1.4038830808619588e-06,
252
+ "loss": 0.6863,
253
+ "step": 330
254
+ },
255
+ {
256
+ "epoch": 0.04351721489824651,
257
+ "grad_norm": 0.9528706073760986,
258
+ "learning_rate": 1.4465542991252401e-06,
259
+ "loss": 0.6907,
260
+ "step": 340
261
+ },
262
+ {
263
+ "epoch": 0.04479713298348906,
264
+ "grad_norm": 1.071359634399414,
265
+ "learning_rate": 1.4892255173885215e-06,
266
+ "loss": 0.6873,
267
+ "step": 350
268
+ },
269
+ {
270
+ "epoch": 0.046077051068731605,
271
+ "grad_norm": 1.9875625371932983,
272
+ "learning_rate": 1.531896735651803e-06,
273
+ "loss": 0.6942,
274
+ "step": 360
275
+ },
276
+ {
277
+ "epoch": 0.047356969153974145,
278
+ "grad_norm": 2.0222439765930176,
279
+ "learning_rate": 1.5745679539150842e-06,
280
+ "loss": 0.6898,
281
+ "step": 370
282
+ },
283
+ {
284
+ "epoch": 0.04863688723921669,
285
+ "grad_norm": 1.7827857732772827,
286
+ "learning_rate": 1.6172391721783658e-06,
287
+ "loss": 0.6909,
288
+ "step": 380
289
+ },
290
+ {
291
+ "epoch": 0.04991680532445923,
292
+ "grad_norm": 1.065674901008606,
293
+ "learning_rate": 1.6599103904416472e-06,
294
+ "loss": 0.6877,
295
+ "step": 390
296
+ },
297
+ {
298
+ "epoch": 0.05119672340970178,
299
+ "grad_norm": 0.8346193432807922,
300
+ "learning_rate": 1.7025816087049288e-06,
301
+ "loss": 0.6889,
302
+ "step": 400
303
+ },
304
+ {
305
+ "epoch": 0.05119672340970178,
306
+ "eval_accuracy": 0.54043,
307
+ "eval_loss": 0.6827595233917236,
308
+ "eval_macro_f1": 0.41833840347598905,
309
+ "eval_precision": 0.5839205994464474,
310
+ "eval_recall": 0.5204152767503025,
311
+ "eval_runtime": 1634.6244,
312
+ "eval_samples_per_second": 61.176,
313
+ "eval_steps_per_second": 0.478,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 0.052476641494944326,
318
+ "grad_norm": 0.8074935674667358,
319
+ "learning_rate": 1.7452528269682101e-06,
320
+ "loss": 0.684,
321
+ "step": 410
322
+ },
323
+ {
324
+ "epoch": 0.053756559580186866,
325
+ "grad_norm": 0.8965069651603699,
326
+ "learning_rate": 1.7879240452314913e-06,
327
+ "loss": 0.6922,
328
+ "step": 420
329
+ },
330
+ {
331
+ "epoch": 0.05503647766542941,
332
+ "grad_norm": 2.597215414047241,
333
+ "learning_rate": 1.8305952634947729e-06,
334
+ "loss": 0.6909,
335
+ "step": 430
336
+ },
337
+ {
338
+ "epoch": 0.05631639575067196,
339
+ "grad_norm": 0.8233373165130615,
340
+ "learning_rate": 1.8732664817580542e-06,
341
+ "loss": 0.6831,
342
+ "step": 440
343
+ },
344
+ {
345
+ "epoch": 0.0575963138359145,
346
+ "grad_norm": 1.0624489784240723,
347
+ "learning_rate": 1.915937700021336e-06,
348
+ "loss": 0.6817,
349
+ "step": 450
350
+ },
351
+ {
352
+ "epoch": 0.05887623192115705,
353
+ "grad_norm": 2.0623559951782227,
354
+ "learning_rate": 1.958608918284617e-06,
355
+ "loss": 0.6819,
356
+ "step": 460
357
+ },
358
+ {
359
+ "epoch": 0.06015615000639959,
360
+ "grad_norm": 0.912433922290802,
361
+ "learning_rate": 2.0012801365478988e-06,
362
+ "loss": 0.686,
363
+ "step": 470
364
+ },
365
+ {
366
+ "epoch": 0.061436068091642135,
367
+ "grad_norm": 0.6926271319389343,
368
+ "learning_rate": 2.04395135481118e-06,
369
+ "loss": 0.6834,
370
+ "step": 480
371
+ },
372
+ {
373
+ "epoch": 0.06271598617688467,
374
+ "grad_norm": 1.498634934425354,
375
+ "learning_rate": 2.0866225730744615e-06,
376
+ "loss": 0.6805,
377
+ "step": 490
378
+ },
379
+ {
380
+ "epoch": 0.06399590426212723,
381
+ "grad_norm": 1.1818156242370605,
382
+ "learning_rate": 2.129293791337743e-06,
383
+ "loss": 0.6805,
384
+ "step": 500
385
+ },
386
+ {
387
+ "epoch": 0.06527582234736977,
388
+ "grad_norm": 1.4476051330566406,
389
+ "learning_rate": 2.1719650096010242e-06,
390
+ "loss": 0.6773,
391
+ "step": 510
392
+ },
393
+ {
394
+ "epoch": 0.06655574043261231,
395
+ "grad_norm": 1.1143743991851807,
396
+ "learning_rate": 2.214636227864306e-06,
397
+ "loss": 0.6801,
398
+ "step": 520
399
+ },
400
+ {
401
+ "epoch": 0.06783565851785486,
402
+ "grad_norm": 1.7030247449874878,
403
+ "learning_rate": 2.257307446127587e-06,
404
+ "loss": 0.6794,
405
+ "step": 530
406
+ },
407
+ {
408
+ "epoch": 0.0691155766030974,
409
+ "grad_norm": 0.8592290282249451,
410
+ "learning_rate": 2.2999786643908685e-06,
411
+ "loss": 0.6736,
412
+ "step": 540
413
+ },
414
+ {
415
+ "epoch": 0.07039549468833994,
416
+ "grad_norm": 0.8231940865516663,
417
+ "learning_rate": 2.34264988265415e-06,
418
+ "loss": 0.67,
419
+ "step": 550
420
+ },
421
+ {
422
+ "epoch": 0.0716754127735825,
423
+ "grad_norm": 0.88897705078125,
424
+ "learning_rate": 2.3853211009174317e-06,
425
+ "loss": 0.685,
426
+ "step": 560
427
+ },
428
+ {
429
+ "epoch": 0.07295533085882504,
430
+ "grad_norm": 1.6297541856765747,
431
+ "learning_rate": 2.427992319180713e-06,
432
+ "loss": 0.6797,
433
+ "step": 570
434
+ },
435
+ {
436
+ "epoch": 0.07423524894406758,
437
+ "grad_norm": 2.9297447204589844,
438
+ "learning_rate": 2.470663537443994e-06,
439
+ "loss": 0.6766,
440
+ "step": 580
441
+ },
442
+ {
443
+ "epoch": 0.07551516702931012,
444
+ "grad_norm": 0.9277703762054443,
445
+ "learning_rate": 2.5133347557072756e-06,
446
+ "loss": 0.6824,
447
+ "step": 590
448
+ },
449
+ {
450
+ "epoch": 0.07679508511455267,
451
+ "grad_norm": 0.5883333683013916,
452
+ "learning_rate": 2.556005973970557e-06,
453
+ "loss": 0.6818,
454
+ "step": 600
455
+ },
456
+ {
457
+ "epoch": 0.07679508511455267,
458
+ "eval_accuracy": 0.63909,
459
+ "eval_loss": 0.6700075268745422,
460
+ "eval_macro_f1": 0.621848143053056,
461
+ "eval_precision": 0.6524313071088819,
462
+ "eval_recall": 0.6305863955136095,
463
+ "eval_runtime": 1633.9732,
464
+ "eval_samples_per_second": 61.201,
465
+ "eval_steps_per_second": 0.479,
466
+ "step": 600
467
+ },
468
+ {
469
+ "epoch": 0.07807500319979521,
470
+ "grad_norm": 0.7149448394775391,
471
+ "learning_rate": 2.5986771922338383e-06,
472
+ "loss": 0.6746,
473
+ "step": 610
474
+ },
475
+ {
476
+ "epoch": 0.07935492128503775,
477
+ "grad_norm": 2.0167856216430664,
478
+ "learning_rate": 2.64134841049712e-06,
479
+ "loss": 0.682,
480
+ "step": 620
481
+ },
482
+ {
483
+ "epoch": 0.0806348393702803,
484
+ "grad_norm": 1.303001880645752,
485
+ "learning_rate": 2.6840196287604015e-06,
486
+ "loss": 0.6735,
487
+ "step": 630
488
+ },
489
+ {
490
+ "epoch": 0.08191475745552285,
491
+ "grad_norm": 2.0516321659088135,
492
+ "learning_rate": 2.7266908470236826e-06,
493
+ "loss": 0.6705,
494
+ "step": 640
495
+ },
496
+ {
497
+ "epoch": 0.08319467554076539,
498
+ "grad_norm": 2.242105484008789,
499
+ "learning_rate": 2.7693620652869642e-06,
500
+ "loss": 0.6798,
501
+ "step": 650
502
+ },
503
+ {
504
+ "epoch": 0.08447459362600794,
505
+ "grad_norm": 1.1836538314819336,
506
+ "learning_rate": 2.812033283550246e-06,
507
+ "loss": 0.6684,
508
+ "step": 660
509
+ },
510
+ {
511
+ "epoch": 0.08575451171125048,
512
+ "grad_norm": 0.8640874624252319,
513
+ "learning_rate": 2.8547045018135274e-06,
514
+ "loss": 0.6783,
515
+ "step": 670
516
+ },
517
+ {
518
+ "epoch": 0.08703442979649302,
519
+ "grad_norm": 0.6901177763938904,
520
+ "learning_rate": 2.897375720076808e-06,
521
+ "loss": 0.6724,
522
+ "step": 680
523
+ },
524
+ {
525
+ "epoch": 0.08831434788173557,
526
+ "grad_norm": 1.3443949222564697,
527
+ "learning_rate": 2.9400469383400897e-06,
528
+ "loss": 0.67,
529
+ "step": 690
530
+ },
531
+ {
532
+ "epoch": 0.08959426596697811,
533
+ "grad_norm": 0.7775189280509949,
534
+ "learning_rate": 2.9827181566033713e-06,
535
+ "loss": 0.6733,
536
+ "step": 700
537
+ },
538
+ {
539
+ "epoch": 0.09087418405222066,
540
+ "grad_norm": 1.16916823387146,
541
+ "learning_rate": 3.025389374866653e-06,
542
+ "loss": 0.6713,
543
+ "step": 710
544
+ },
545
+ {
546
+ "epoch": 0.09215410213746321,
547
+ "grad_norm": 1.707152009010315,
548
+ "learning_rate": 3.068060593129934e-06,
549
+ "loss": 0.6659,
550
+ "step": 720
551
+ },
552
+ {
553
+ "epoch": 0.09343402022270575,
554
+ "grad_norm": 0.9780622124671936,
555
+ "learning_rate": 3.1107318113932156e-06,
556
+ "loss": 0.6639,
557
+ "step": 730
558
+ },
559
+ {
560
+ "epoch": 0.09471393830794829,
561
+ "grad_norm": 2.3687374591827393,
562
+ "learning_rate": 3.153403029656497e-06,
563
+ "loss": 0.6666,
564
+ "step": 740
565
+ },
566
+ {
567
+ "epoch": 0.09599385639319083,
568
+ "grad_norm": 0.8506460189819336,
569
+ "learning_rate": 3.1960742479197783e-06,
570
+ "loss": 0.6778,
571
+ "step": 750
572
+ },
573
+ {
574
+ "epoch": 0.09727377447843338,
575
+ "grad_norm": 1.0102107524871826,
576
+ "learning_rate": 3.23874546618306e-06,
577
+ "loss": 0.6635,
578
+ "step": 760
579
+ },
580
+ {
581
+ "epoch": 0.09855369256367592,
582
+ "grad_norm": 1.1480194330215454,
583
+ "learning_rate": 3.2814166844463415e-06,
584
+ "loss": 0.6709,
585
+ "step": 770
586
+ },
587
+ {
588
+ "epoch": 0.09983361064891846,
589
+ "grad_norm": 1.3119179010391235,
590
+ "learning_rate": 3.324087902709623e-06,
591
+ "loss": 0.6564,
592
+ "step": 780
593
+ },
594
+ {
595
+ "epoch": 0.10111352873416102,
596
+ "grad_norm": 0.681377649307251,
597
+ "learning_rate": 3.3667591209729038e-06,
598
+ "loss": 0.6656,
599
+ "step": 790
600
+ },
601
+ {
602
+ "epoch": 0.10239344681940356,
603
+ "grad_norm": 1.022257685661316,
604
+ "learning_rate": 3.4094303392361854e-06,
605
+ "loss": 0.6709,
606
+ "step": 800
607
+ },
608
+ {
609
+ "epoch": 0.10239344681940356,
610
+ "eval_accuracy": 0.66247,
611
+ "eval_loss": 0.6552017331123352,
612
+ "eval_macro_f1": 0.6514270557181998,
613
+ "eval_precision": 0.6718270655389242,
614
+ "eval_recall": 0.6556580720913894,
615
+ "eval_runtime": 1633.5478,
616
+ "eval_samples_per_second": 61.216,
617
+ "eval_steps_per_second": 0.479,
618
+ "step": 800
619
+ },
620
+ {
621
+ "epoch": 0.1036733649046461,
622
+ "grad_norm": 1.0268975496292114,
623
+ "learning_rate": 3.452101557499467e-06,
624
+ "loss": 0.6602,
625
+ "step": 810
626
+ },
627
+ {
628
+ "epoch": 0.10495328298988865,
629
+ "grad_norm": 2.6206603050231934,
630
+ "learning_rate": 3.494772775762748e-06,
631
+ "loss": 0.6705,
632
+ "step": 820
633
+ },
634
+ {
635
+ "epoch": 0.10623320107513119,
636
+ "grad_norm": 1.035157561302185,
637
+ "learning_rate": 3.5374439940260297e-06,
638
+ "loss": 0.6643,
639
+ "step": 830
640
+ },
641
+ {
642
+ "epoch": 0.10751311916037373,
643
+ "grad_norm": 1.736412763595581,
644
+ "learning_rate": 3.5801152122893113e-06,
645
+ "loss": 0.6612,
646
+ "step": 840
647
+ },
648
+ {
649
+ "epoch": 0.10879303724561629,
650
+ "grad_norm": 0.6906964778900146,
651
+ "learning_rate": 3.622786430552593e-06,
652
+ "loss": 0.6576,
653
+ "step": 850
654
+ },
655
+ {
656
+ "epoch": 0.11007295533085883,
657
+ "grad_norm": 0.8148729801177979,
658
+ "learning_rate": 3.665457648815874e-06,
659
+ "loss": 0.6561,
660
+ "step": 860
661
+ },
662
+ {
663
+ "epoch": 0.11135287341610137,
664
+ "grad_norm": 1.0217723846435547,
665
+ "learning_rate": 3.7081288670791556e-06,
666
+ "loss": 0.6581,
667
+ "step": 870
668
+ },
669
+ {
670
+ "epoch": 0.11263279150134392,
671
+ "grad_norm": 1.2743115425109863,
672
+ "learning_rate": 3.750800085342437e-06,
673
+ "loss": 0.6606,
674
+ "step": 880
675
+ },
676
+ {
677
+ "epoch": 0.11391270958658646,
678
+ "grad_norm": 0.6405361294746399,
679
+ "learning_rate": 3.793471303605718e-06,
680
+ "loss": 0.652,
681
+ "step": 890
682
+ },
683
+ {
684
+ "epoch": 0.115192627671829,
685
+ "grad_norm": 0.8357161283493042,
686
+ "learning_rate": 3.836142521869e-06,
687
+ "loss": 0.6558,
688
+ "step": 900
689
+ },
690
+ {
691
+ "epoch": 0.11647254575707154,
692
+ "grad_norm": 1.3505173921585083,
693
+ "learning_rate": 3.8788137401322815e-06,
694
+ "loss": 0.6576,
695
+ "step": 910
696
+ },
697
+ {
698
+ "epoch": 0.1177524638423141,
699
+ "grad_norm": 1.2253319025039673,
700
+ "learning_rate": 3.921484958395563e-06,
701
+ "loss": 0.6548,
702
+ "step": 920
703
+ },
704
+ {
705
+ "epoch": 0.11903238192755664,
706
+ "grad_norm": 0.6269568800926208,
707
+ "learning_rate": 3.964156176658844e-06,
708
+ "loss": 0.6562,
709
+ "step": 930
710
+ },
711
+ {
712
+ "epoch": 0.12031230001279918,
713
+ "grad_norm": 0.9413424134254456,
714
+ "learning_rate": 4.006827394922125e-06,
715
+ "loss": 0.6461,
716
+ "step": 940
717
+ },
718
+ {
719
+ "epoch": 0.12159221809804173,
720
+ "grad_norm": 1.0903513431549072,
721
+ "learning_rate": 4.049498613185407e-06,
722
+ "loss": 0.6457,
723
+ "step": 950
724
+ },
725
+ {
726
+ "epoch": 0.12287213618328427,
727
+ "grad_norm": 2.0023152828216553,
728
+ "learning_rate": 4.092169831448688e-06,
729
+ "loss": 0.6458,
730
+ "step": 960
731
+ },
732
+ {
733
+ "epoch": 0.12415205426852681,
734
+ "grad_norm": 0.9638109803199768,
735
+ "learning_rate": 4.134841049711969e-06,
736
+ "loss": 0.6486,
737
+ "step": 970
738
+ },
739
+ {
740
+ "epoch": 0.12543197235376935,
741
+ "grad_norm": 0.9894546270370483,
742
+ "learning_rate": 4.177512267975251e-06,
743
+ "loss": 0.6453,
744
+ "step": 980
745
+ },
746
+ {
747
+ "epoch": 0.1267118904390119,
748
+ "grad_norm": 1.2289057970046997,
749
+ "learning_rate": 4.220183486238532e-06,
750
+ "loss": 0.6605,
751
+ "step": 990
752
+ },
753
+ {
754
+ "epoch": 0.12799180852425446,
755
+ "grad_norm": 1.119125247001648,
756
+ "learning_rate": 4.262854704501814e-06,
757
+ "loss": 0.6564,
758
+ "step": 1000
759
+ },
760
+ {
761
+ "epoch": 0.12799180852425446,
762
+ "eval_accuracy": 0.69068,
763
+ "eval_loss": 0.6396247148513794,
764
+ "eval_macro_f1": 0.689588440856387,
765
+ "eval_precision": 0.6900076024604327,
766
+ "eval_recall": 0.6894114140312133,
767
+ "eval_runtime": 1634.4321,
768
+ "eval_samples_per_second": 61.183,
769
+ "eval_steps_per_second": 0.478,
770
+ "step": 1000
771
+ },
772
+ {
773
+ "epoch": 0.12927172660949698,
774
+ "grad_norm": 2.02966046333313,
775
+ "learning_rate": 4.3055259227650956e-06,
776
+ "loss": 0.646,
777
+ "step": 1010
778
+ },
779
+ {
780
+ "epoch": 0.13055164469473954,
781
+ "grad_norm": 0.7430800199508667,
782
+ "learning_rate": 4.348197141028377e-06,
783
+ "loss": 0.6453,
784
+ "step": 1020
785
+ },
786
+ {
787
+ "epoch": 0.1318315627799821,
788
+ "grad_norm": 1.031962275505066,
789
+ "learning_rate": 4.390868359291658e-06,
790
+ "loss": 0.6425,
791
+ "step": 1030
792
+ },
793
+ {
794
+ "epoch": 0.13311148086522462,
795
+ "grad_norm": 1.0249106884002686,
796
+ "learning_rate": 4.4335395775549394e-06,
797
+ "loss": 0.6426,
798
+ "step": 1040
799
+ },
800
+ {
801
+ "epoch": 0.13439139895046717,
802
+ "grad_norm": 0.7375448942184448,
803
+ "learning_rate": 4.476210795818221e-06,
804
+ "loss": 0.6401,
805
+ "step": 1050
806
+ },
807
+ {
808
+ "epoch": 0.13567131703570973,
809
+ "grad_norm": 1.295983076095581,
810
+ "learning_rate": 4.518882014081503e-06,
811
+ "loss": 0.6449,
812
+ "step": 1060
813
+ },
814
+ {
815
+ "epoch": 0.13695123512095225,
816
+ "grad_norm": 1.6978998184204102,
817
+ "learning_rate": 4.561553232344783e-06,
818
+ "loss": 0.6484,
819
+ "step": 1070
820
+ },
821
+ {
822
+ "epoch": 0.1382311532061948,
823
+ "grad_norm": 0.7747512459754944,
824
+ "learning_rate": 4.604224450608065e-06,
825
+ "loss": 0.6526,
826
+ "step": 1080
827
+ },
828
+ {
829
+ "epoch": 0.13951107129143736,
830
+ "grad_norm": 0.7895913124084473,
831
+ "learning_rate": 4.6468956688713465e-06,
832
+ "loss": 0.6311,
833
+ "step": 1090
834
+ },
835
+ {
836
+ "epoch": 0.1407909893766799,
837
+ "grad_norm": 0.6975072026252747,
838
+ "learning_rate": 4.689566887134628e-06,
839
+ "loss": 0.6454,
840
+ "step": 1100
841
+ },
842
+ {
843
+ "epoch": 0.14207090746192244,
844
+ "grad_norm": 0.8316811919212341,
845
+ "learning_rate": 4.73223810539791e-06,
846
+ "loss": 0.6447,
847
+ "step": 1110
848
+ },
849
+ {
850
+ "epoch": 0.143350825547165,
851
+ "grad_norm": 0.6887441277503967,
852
+ "learning_rate": 4.774909323661191e-06,
853
+ "loss": 0.6325,
854
+ "step": 1120
855
+ },
856
+ {
857
+ "epoch": 0.14463074363240752,
858
+ "grad_norm": 1.0218420028686523,
859
+ "learning_rate": 4.817580541924473e-06,
860
+ "loss": 0.6308,
861
+ "step": 1130
862
+ },
863
+ {
864
+ "epoch": 0.14591066171765008,
865
+ "grad_norm": 0.7663633823394775,
866
+ "learning_rate": 4.8602517601877535e-06,
867
+ "loss": 0.6227,
868
+ "step": 1140
869
+ },
870
+ {
871
+ "epoch": 0.1471905798028926,
872
+ "grad_norm": 1.5115224123001099,
873
+ "learning_rate": 4.902922978451035e-06,
874
+ "loss": 0.6315,
875
+ "step": 1150
876
+ },
877
+ {
878
+ "epoch": 0.14847049788813516,
879
+ "grad_norm": 0.8011078238487244,
880
+ "learning_rate": 4.945594196714317e-06,
881
+ "loss": 0.6428,
882
+ "step": 1160
883
+ },
884
+ {
885
+ "epoch": 0.1497504159733777,
886
+ "grad_norm": 1.3246116638183594,
887
+ "learning_rate": 4.9882654149775974e-06,
888
+ "loss": 0.6397,
889
+ "step": 1170
890
+ },
891
+ {
892
+ "epoch": 0.15103033405862024,
893
+ "grad_norm": 0.979498565196991,
894
+ "learning_rate": 5.030936633240879e-06,
895
+ "loss": 0.6308,
896
+ "step": 1180
897
+ },
898
+ {
899
+ "epoch": 0.1523102521438628,
900
+ "grad_norm": 0.8756135106086731,
901
+ "learning_rate": 5.073607851504161e-06,
902
+ "loss": 0.6373,
903
+ "step": 1190
904
+ },
905
+ {
906
+ "epoch": 0.15359017022910534,
907
+ "grad_norm": 0.845967710018158,
908
+ "learning_rate": 5.116279069767442e-06,
909
+ "loss": 0.6322,
910
+ "step": 1200
911
+ },
912
+ {
913
+ "epoch": 0.15359017022910534,
914
+ "eval_accuracy": 0.69931,
915
+ "eval_loss": 0.6224722862243652,
916
+ "eval_macro_f1": 0.6985627236986651,
917
+ "eval_precision": 0.6986454338713279,
918
+ "eval_recall": 0.6984991383538209,
919
+ "eval_runtime": 692.1526,
920
+ "eval_samples_per_second": 144.477,
921
+ "eval_steps_per_second": 2.258,
922
+ "step": 1200
923
+ },
924
+ {
925
+ "epoch": 0.15487008831434787,
926
+ "grad_norm": 2.0670483112335205,
927
+ "learning_rate": 1.032010243277849e-05,
928
+ "loss": 0.6386,
929
+ "step": 1210
930
+ },
931
+ {
932
+ "epoch": 0.15615000639959042,
933
+ "grad_norm": 2.482168436050415,
934
+ "learning_rate": 1.0405463081519421e-05,
935
+ "loss": 0.6281,
936
+ "step": 1220
937
+ },
938
+ {
939
+ "epoch": 0.15742992448483298,
940
+ "grad_norm": 1.4530706405639648,
941
+ "learning_rate": 1.049082373026035e-05,
942
+ "loss": 0.6277,
943
+ "step": 1230
944
+ },
945
+ {
946
+ "epoch": 0.1587098425700755,
947
+ "grad_norm": 2.9233577251434326,
948
+ "learning_rate": 1.0576184379001282e-05,
949
+ "loss": 0.6363,
950
+ "step": 1240
951
+ },
952
+ {
953
+ "epoch": 0.15998976065531806,
954
+ "grad_norm": 0.6975642442703247,
955
+ "learning_rate": 1.0661545027742212e-05,
956
+ "loss": 0.6344,
957
+ "step": 1250
958
+ },
959
+ {
960
+ "epoch": 0.1612696787405606,
961
+ "grad_norm": 2.3634731769561768,
962
+ "learning_rate": 1.0746905676483142e-05,
963
+ "loss": 0.6217,
964
+ "step": 1260
965
+ },
966
+ {
967
+ "epoch": 0.16254959682580314,
968
+ "grad_norm": 0.8812585473060608,
969
+ "learning_rate": 1.0832266325224072e-05,
970
+ "loss": 0.6311,
971
+ "step": 1270
972
+ },
973
+ {
974
+ "epoch": 0.1638295149110457,
975
+ "grad_norm": 1.8986798524856567,
976
+ "learning_rate": 1.0917626973965003e-05,
977
+ "loss": 0.6304,
978
+ "step": 1280
979
+ },
980
+ {
981
+ "epoch": 0.16510943299628825,
982
+ "grad_norm": 0.8174984455108643,
983
+ "learning_rate": 1.1002987622705935e-05,
984
+ "loss": 0.6176,
985
+ "step": 1290
986
+ },
987
+ {
988
+ "epoch": 0.16638935108153077,
989
+ "grad_norm": 1.0268642902374268,
990
+ "learning_rate": 1.1088348271446865e-05,
991
+ "loss": 0.6233,
992
+ "step": 1300
993
+ },
994
+ {
995
+ "epoch": 0.16766926916677333,
996
+ "grad_norm": 1.047821044921875,
997
+ "learning_rate": 1.1173708920187793e-05,
998
+ "loss": 0.6127,
999
+ "step": 1310
1000
+ },
1001
+ {
1002
+ "epoch": 0.16894918725201588,
1003
+ "grad_norm": 0.9875057935714722,
1004
+ "learning_rate": 1.1259069568928724e-05,
1005
+ "loss": 0.6211,
1006
+ "step": 1320
1007
+ },
1008
+ {
1009
+ "epoch": 0.1702291053372584,
1010
+ "grad_norm": 1.0082708597183228,
1011
+ "learning_rate": 1.1344430217669656e-05,
1012
+ "loss": 0.6202,
1013
+ "step": 1330
1014
+ },
1015
+ {
1016
+ "epoch": 0.17150902342250096,
1017
+ "grad_norm": 0.7229795455932617,
1018
+ "learning_rate": 1.1429790866410586e-05,
1019
+ "loss": 0.6071,
1020
+ "step": 1340
1021
+ },
1022
+ {
1023
+ "epoch": 0.17278894150774352,
1024
+ "grad_norm": 0.7687681317329407,
1025
+ "learning_rate": 1.1515151515151517e-05,
1026
+ "loss": 0.6218,
1027
+ "step": 1350
1028
+ },
1029
+ {
1030
+ "epoch": 0.17406885959298604,
1031
+ "grad_norm": 1.1643378734588623,
1032
+ "learning_rate": 1.1600512163892445e-05,
1033
+ "loss": 0.6208,
1034
+ "step": 1360
1035
+ },
1036
+ {
1037
+ "epoch": 0.1753487776782286,
1038
+ "grad_norm": 0.7338509559631348,
1039
+ "learning_rate": 1.1685872812633377e-05,
1040
+ "loss": 0.6105,
1041
+ "step": 1370
1042
+ },
1043
+ {
1044
+ "epoch": 0.17662869576347115,
1045
+ "grad_norm": 0.7444656491279602,
1046
+ "learning_rate": 1.1771233461374307e-05,
1047
+ "loss": 0.6183,
1048
+ "step": 1380
1049
+ },
1050
+ {
1051
+ "epoch": 0.17790861384871368,
1052
+ "grad_norm": 1.1902283430099487,
1053
+ "learning_rate": 1.1856594110115238e-05,
1054
+ "loss": 0.6209,
1055
+ "step": 1390
1056
+ },
1057
+ {
1058
+ "epoch": 0.17918853193395623,
1059
+ "grad_norm": 1.0872228145599365,
1060
+ "learning_rate": 1.194195475885617e-05,
1061
+ "loss": 0.6172,
1062
+ "step": 1400
1063
+ },
1064
+ {
1065
+ "epoch": 0.18046845001919878,
1066
+ "grad_norm": 1.8439041376113892,
1067
+ "learning_rate": 1.2027315407597098e-05,
1068
+ "loss": 0.6086,
1069
+ "step": 1410
1070
+ },
1071
+ {
1072
+ "epoch": 0.1817483681044413,
1073
+ "grad_norm": 0.8699899911880493,
1074
+ "learning_rate": 1.211267605633803e-05,
1075
+ "loss": 0.6247,
1076
+ "step": 1420
1077
+ },
1078
+ {
1079
+ "epoch": 0.18302828618968386,
1080
+ "grad_norm": 1.1468650102615356,
1081
+ "learning_rate": 1.2198036705078959e-05,
1082
+ "loss": 0.6114,
1083
+ "step": 1430
1084
+ },
1085
+ {
1086
+ "epoch": 0.18430820427492642,
1087
+ "grad_norm": 1.0204615592956543,
1088
+ "learning_rate": 1.228339735381989e-05,
1089
+ "loss": 0.6015,
1090
+ "step": 1440
1091
+ },
1092
+ {
1093
+ "epoch": 0.18558812236016894,
1094
+ "grad_norm": 2.660583257675171,
1095
+ "learning_rate": 1.236875800256082e-05,
1096
+ "loss": 0.6002,
1097
+ "step": 1450
1098
+ },
1099
+ {
1100
+ "epoch": 0.1868680404454115,
1101
+ "grad_norm": 1.1916371583938599,
1102
+ "learning_rate": 1.2454118651301752e-05,
1103
+ "loss": 0.6076,
1104
+ "step": 1460
1105
+ },
1106
+ {
1107
+ "epoch": 0.18814795853065402,
1108
+ "grad_norm": 0.8097987174987793,
1109
+ "learning_rate": 1.253947930004268e-05,
1110
+ "loss": 0.6126,
1111
+ "step": 1470
1112
+ },
1113
+ {
1114
+ "epoch": 0.18942787661589658,
1115
+ "grad_norm": 1.978747010231018,
1116
+ "learning_rate": 1.2624839948783612e-05,
1117
+ "loss": 0.5962,
1118
+ "step": 1480
1119
+ },
1120
+ {
1121
+ "epoch": 0.19070779470113913,
1122
+ "grad_norm": 1.0975425243377686,
1123
+ "learning_rate": 1.2710200597524543e-05,
1124
+ "loss": 0.587,
1125
+ "step": 1490
1126
+ },
1127
+ {
1128
+ "epoch": 0.19198771278638166,
1129
+ "grad_norm": 2.150351047515869,
1130
+ "learning_rate": 1.2795561246265473e-05,
1131
+ "loss": 0.5962,
1132
+ "step": 1500
1133
+ },
1134
+ {
1135
+ "epoch": 0.19198771278638166,
1136
+ "eval_accuracy": 0.72526,
1137
+ "eval_loss": 0.5823243856430054,
1138
+ "eval_macro_f1": 0.7252418839884757,
1139
+ "eval_precision": 0.7275813809678697,
1140
+ "eval_recall": 0.7271797416587706,
1141
+ "eval_runtime": 708.4119,
1142
+ "eval_samples_per_second": 141.161,
1143
+ "eval_steps_per_second": 1.104,
1144
+ "step": 1500
1145
+ },
1146
+ {
1147
+ "epoch": 0.1932676308716242,
1148
+ "grad_norm": 1.8355706930160522,
1149
+ "learning_rate": 1.2880921895006405e-05,
1150
+ "loss": 0.6133,
1151
+ "step": 1510
1152
+ },
1153
+ {
1154
+ "epoch": 0.19454754895686677,
1155
+ "grad_norm": 1.4349753856658936,
1156
+ "learning_rate": 1.2966282543747333e-05,
1157
+ "loss": 0.6062,
1158
+ "step": 1520
1159
+ },
1160
+ {
1161
+ "epoch": 0.1958274670421093,
1162
+ "grad_norm": 1.8322596549987793,
1163
+ "learning_rate": 1.3051643192488264e-05,
1164
+ "loss": 0.5982,
1165
+ "step": 1530
1166
+ },
1167
+ {
1168
+ "epoch": 0.19710738512735185,
1169
+ "grad_norm": 1.509338617324829,
1170
+ "learning_rate": 1.3137003841229194e-05,
1171
+ "loss": 0.5961,
1172
+ "step": 1540
1173
+ },
1174
+ {
1175
+ "epoch": 0.1983873032125944,
1176
+ "grad_norm": 0.9844911098480225,
1177
+ "learning_rate": 1.3222364489970126e-05,
1178
+ "loss": 0.583,
1179
+ "step": 1550
1180
+ },
1181
+ {
1182
+ "epoch": 0.19966722129783693,
1183
+ "grad_norm": 1.035900354385376,
1184
+ "learning_rate": 1.3307725138711055e-05,
1185
+ "loss": 0.598,
1186
+ "step": 1560
1187
+ },
1188
+ {
1189
+ "epoch": 0.20094713938307948,
1190
+ "grad_norm": 0.7891924977302551,
1191
+ "learning_rate": 1.3393085787451985e-05,
1192
+ "loss": 0.6064,
1193
+ "step": 1570
1194
+ },
1195
+ {
1196
+ "epoch": 0.20222705746832204,
1197
+ "grad_norm": 1.4326112270355225,
1198
+ "learning_rate": 1.3478446436192915e-05,
1199
+ "loss": 0.5954,
1200
+ "step": 1580
1201
+ },
1202
+ {
1203
+ "epoch": 0.20350697555356456,
1204
+ "grad_norm": 0.9054266214370728,
1205
+ "learning_rate": 1.3563807084933847e-05,
1206
+ "loss": 0.59,
1207
+ "step": 1590
1208
+ },
1209
+ {
1210
+ "epoch": 0.20478689363880712,
1211
+ "grad_norm": 1.0216469764709473,
1212
+ "learning_rate": 1.3649167733674778e-05,
1213
+ "loss": 0.5845,
1214
+ "step": 1600
1215
+ },
1216
+ {
1217
+ "epoch": 0.20606681172404967,
1218
+ "grad_norm": 1.0591177940368652,
1219
+ "learning_rate": 1.3734528382415708e-05,
1220
+ "loss": 0.599,
1221
+ "step": 1610
1222
+ },
1223
+ {
1224
+ "epoch": 0.2073467298092922,
1225
+ "grad_norm": 0.7826931476593018,
1226
+ "learning_rate": 1.3819889031156638e-05,
1227
+ "loss": 0.5898,
1228
+ "step": 1620
1229
+ },
1230
+ {
1231
+ "epoch": 0.20862664789453475,
1232
+ "grad_norm": 0.755007803440094,
1233
+ "learning_rate": 1.3905249679897568e-05,
1234
+ "loss": 0.5769,
1235
+ "step": 1630
1236
+ },
1237
+ {
1238
+ "epoch": 0.2099065659797773,
1239
+ "grad_norm": 1.3526090383529663,
1240
+ "learning_rate": 1.3990610328638499e-05,
1241
+ "loss": 0.5819,
1242
+ "step": 1640
1243
+ },
1244
+ {
1245
+ "epoch": 0.21118648406501983,
1246
+ "grad_norm": 0.7078200578689575,
1247
+ "learning_rate": 1.4075970977379429e-05,
1248
+ "loss": 0.5884,
1249
+ "step": 1650
1250
+ },
1251
+ {
1252
+ "epoch": 0.21246640215026238,
1253
+ "grad_norm": 0.80736243724823,
1254
+ "learning_rate": 1.416133162612036e-05,
1255
+ "loss": 0.5991,
1256
+ "step": 1660
1257
+ },
1258
+ {
1259
+ "epoch": 0.21374632023550494,
1260
+ "grad_norm": 0.6882274150848389,
1261
+ "learning_rate": 1.4246692274861289e-05,
1262
+ "loss": 0.5825,
1263
+ "step": 1670
1264
+ },
1265
+ {
1266
+ "epoch": 0.21502623832074746,
1267
+ "grad_norm": 0.7223654985427856,
1268
+ "learning_rate": 1.433205292360222e-05,
1269
+ "loss": 0.5689,
1270
+ "step": 1680
1271
+ },
1272
+ {
1273
+ "epoch": 0.21630615640599002,
1274
+ "grad_norm": 0.6446322798728943,
1275
+ "learning_rate": 1.4417413572343152e-05,
1276
+ "loss": 0.5773,
1277
+ "step": 1690
1278
+ },
1279
+ {
1280
+ "epoch": 0.21758607449123257,
1281
+ "grad_norm": 1.505007266998291,
1282
+ "learning_rate": 1.4502774221084081e-05,
1283
+ "loss": 0.5677,
1284
+ "step": 1700
1285
+ },
1286
+ {
1287
+ "epoch": 0.2188659925764751,
1288
+ "grad_norm": 1.1939284801483154,
1289
+ "learning_rate": 1.4588134869825013e-05,
1290
+ "loss": 0.5805,
1291
+ "step": 1710
1292
+ },
1293
+ {
1294
+ "epoch": 0.22014591066171765,
1295
+ "grad_norm": 2.848264217376709,
1296
+ "learning_rate": 1.4673495518565943e-05,
1297
+ "loss": 0.5712,
1298
+ "step": 1720
1299
+ },
1300
+ {
1301
+ "epoch": 0.2214258287469602,
1302
+ "grad_norm": 1.2117749452590942,
1303
+ "learning_rate": 1.4758856167306873e-05,
1304
+ "loss": 0.582,
1305
+ "step": 1730
1306
+ },
1307
+ {
1308
+ "epoch": 0.22270574683220273,
1309
+ "grad_norm": 2.2297441959381104,
1310
+ "learning_rate": 1.4844216816047802e-05,
1311
+ "loss": 0.5751,
1312
+ "step": 1740
1313
+ },
1314
+ {
1315
+ "epoch": 0.2239856649174453,
1316
+ "grad_norm": 1.211978554725647,
1317
+ "learning_rate": 1.4929577464788734e-05,
1318
+ "loss": 0.5607,
1319
+ "step": 1750
1320
+ },
1321
+ {
1322
+ "epoch": 0.22526558300268784,
1323
+ "grad_norm": 0.8025451898574829,
1324
+ "learning_rate": 1.5014938113529664e-05,
1325
+ "loss": 0.5803,
1326
+ "step": 1760
1327
+ },
1328
+ {
1329
+ "epoch": 0.22654550108793037,
1330
+ "grad_norm": 1.2300900220870972,
1331
+ "learning_rate": 1.5100298762270595e-05,
1332
+ "loss": 0.5827,
1333
+ "step": 1770
1334
+ },
1335
+ {
1336
+ "epoch": 0.22782541917317292,
1337
+ "grad_norm": 1.0382864475250244,
1338
+ "learning_rate": 1.5185659411011523e-05,
1339
+ "loss": 0.5916,
1340
+ "step": 1780
1341
+ },
1342
+ {
1343
+ "epoch": 0.22910533725841545,
1344
+ "grad_norm": 1.4876656532287598,
1345
+ "learning_rate": 1.5271020059752453e-05,
1346
+ "loss": 0.5536,
1347
+ "step": 1790
1348
+ },
1349
+ {
1350
+ "epoch": 0.230385255343658,
1351
+ "grad_norm": 1.1409252882003784,
1352
+ "learning_rate": 1.5356380708493386e-05,
1353
+ "loss": 0.5484,
1354
+ "step": 1800
1355
+ },
1356
+ {
1357
+ "epoch": 0.23166517342890056,
1358
+ "grad_norm": 0.7180280089378357,
1359
+ "learning_rate": 1.5441741357234316e-05,
1360
+ "loss": 0.5712,
1361
+ "step": 1810
1362
+ },
1363
+ {
1364
+ "epoch": 0.23294509151414308,
1365
+ "grad_norm": 1.5890289545059204,
1366
+ "learning_rate": 1.5527102005975246e-05,
1367
+ "loss": 0.5758,
1368
+ "step": 1820
1369
+ },
1370
+ {
1371
+ "epoch": 0.23422500959938564,
1372
+ "grad_norm": 0.6635769605636597,
1373
+ "learning_rate": 1.5612462654716176e-05,
1374
+ "loss": 0.5543,
1375
+ "step": 1830
1376
+ },
1377
+ {
1378
+ "epoch": 0.2355049276846282,
1379
+ "grad_norm": 0.8864602446556091,
1380
+ "learning_rate": 1.5697823303457106e-05,
1381
+ "loss": 0.5536,
1382
+ "step": 1840
1383
+ },
1384
+ {
1385
+ "epoch": 0.23678484576987072,
1386
+ "grad_norm": 0.8372675776481628,
1387
+ "learning_rate": 1.578318395219804e-05,
1388
+ "loss": 0.5743,
1389
+ "step": 1850
1390
+ },
1391
+ {
1392
+ "epoch": 0.23806476385511327,
1393
+ "grad_norm": 0.7532811164855957,
1394
+ "learning_rate": 1.586854460093897e-05,
1395
+ "loss": 0.5593,
1396
+ "step": 1860
1397
+ },
1398
+ {
1399
+ "epoch": 0.23934468194035582,
1400
+ "grad_norm": 0.7231799960136414,
1401
+ "learning_rate": 1.59539052496799e-05,
1402
+ "loss": 0.5831,
1403
+ "step": 1870
1404
+ },
1405
+ {
1406
+ "epoch": 0.24062460002559835,
1407
+ "grad_norm": 1.124228596687317,
1408
+ "learning_rate": 1.603926589842083e-05,
1409
+ "loss": 0.5936,
1410
+ "step": 1880
1411
+ },
1412
+ {
1413
+ "epoch": 0.2419045181108409,
1414
+ "grad_norm": 1.0107067823410034,
1415
+ "learning_rate": 1.612462654716176e-05,
1416
+ "loss": 0.5598,
1417
+ "step": 1890
1418
+ },
1419
+ {
1420
+ "epoch": 0.24318443619608346,
1421
+ "grad_norm": 1.6117154359817505,
1422
+ "learning_rate": 1.620998719590269e-05,
1423
+ "loss": 0.5605,
1424
+ "step": 1900
1425
+ },
1426
+ {
1427
+ "epoch": 0.24446435428132599,
1428
+ "grad_norm": 1.5564496517181396,
1429
+ "learning_rate": 1.629534784464362e-05,
1430
+ "loss": 0.5666,
1431
+ "step": 1910
1432
+ },
1433
+ {
1434
+ "epoch": 0.24574427236656854,
1435
+ "grad_norm": 0.8275917768478394,
1436
+ "learning_rate": 1.638070849338455e-05,
1437
+ "loss": 0.5597,
1438
+ "step": 1920
1439
+ },
1440
+ {
1441
+ "epoch": 0.2470241904518111,
1442
+ "grad_norm": 0.8956324458122253,
1443
+ "learning_rate": 1.646606914212548e-05,
1444
+ "loss": 0.5453,
1445
+ "step": 1930
1446
+ },
1447
+ {
1448
+ "epoch": 0.24830410853705362,
1449
+ "grad_norm": 1.9644063711166382,
1450
+ "learning_rate": 1.655142979086641e-05,
1451
+ "loss": 0.5732,
1452
+ "step": 1940
1453
+ },
1454
+ {
1455
+ "epoch": 0.24958402662229617,
1456
+ "grad_norm": 0.7495090365409851,
1457
+ "learning_rate": 1.663679043960734e-05,
1458
+ "loss": 0.5549,
1459
+ "step": 1950
1460
+ },
1461
+ {
1462
+ "epoch": 0.2508639447075387,
1463
+ "grad_norm": 0.9346863627433777,
1464
+ "learning_rate": 1.6722151088348274e-05,
1465
+ "loss": 0.5803,
1466
+ "step": 1960
1467
+ },
1468
+ {
1469
+ "epoch": 0.2521438627927813,
1470
+ "grad_norm": 0.5847868919372559,
1471
+ "learning_rate": 1.6807511737089204e-05,
1472
+ "loss": 0.5705,
1473
+ "step": 1970
1474
+ },
1475
+ {
1476
+ "epoch": 0.2534237808780238,
1477
+ "grad_norm": 1.5892306566238403,
1478
+ "learning_rate": 1.6892872385830134e-05,
1479
+ "loss": 0.5371,
1480
+ "step": 1980
1481
+ },
1482
+ {
1483
+ "epoch": 0.25470369896326633,
1484
+ "grad_norm": 0.8782947063446045,
1485
+ "learning_rate": 1.6978233034571063e-05,
1486
+ "loss": 0.565,
1487
+ "step": 1990
1488
+ },
1489
+ {
1490
+ "epoch": 0.2559836170485089,
1491
+ "grad_norm": 1.0143548250198364,
1492
+ "learning_rate": 1.7063593683311993e-05,
1493
+ "loss": 0.557,
1494
+ "step": 2000
1495
+ },
1496
+ {
1497
+ "epoch": 0.2559836170485089,
1498
+ "eval_accuracy": 0.74427,
1499
+ "eval_loss": 0.5308485627174377,
1500
+ "eval_macro_f1": 0.743516265066894,
1501
+ "eval_precision": 0.7538937627554241,
1502
+ "eval_recall": 0.7483595020974525,
1503
+ "eval_runtime": 708.2272,
1504
+ "eval_samples_per_second": 141.198,
1505
+ "eval_steps_per_second": 1.104,
1506
+ "step": 2000
1507
+ }
1508
+ ],
1509
+ "logging_steps": 10,
1510
+ "max_steps": 23439,
1511
+ "num_input_tokens_seen": 0,
1512
+ "num_train_epochs": 3,
1513
+ "save_steps": 200,
1514
+ "stateful_callbacks": {
1515
+ "EarlyStoppingCallback": {
1516
+ "args": {
1517
+ "early_stopping_patience": 3,
1518
+ "early_stopping_threshold": 0.0
1519
+ },
1520
+ "attributes": {
1521
+ "early_stopping_patience_counter": 0
1522
+ }
1523
+ },
1524
+ "TrainerControl": {
1525
+ "args": {
1526
+ "should_epoch_stop": false,
1527
+ "should_evaluate": false,
1528
+ "should_log": false,
1529
+ "should_save": true,
1530
+ "should_training_stop": false
1531
+ },
1532
+ "attributes": {}
1533
+ }
1534
+ },
1535
+ "total_flos": 3.367821508608e+16,
1536
+ "train_batch_size": 64,
1537
+ "trial_name": null,
1538
+ "trial_params": null
1539
+ }
codebert-base-cross-entropy/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66cbef846d2329e000727fbcf0db86e3ad5252ae9de8caa12e196b1fac3c9577
3
+ size 5841
codebert-base-cross-entropy/checkpoint-2000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/checkpoint-2400/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "RobertaForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": null,
8
+ "dtype": "float32",
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "transformers_version": "4.56.0",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 50265
27
+ }
codebert-base-cross-entropy/checkpoint-2400/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/checkpoint-2400/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a938d5016a359936007454fcb9fc6ee62fbf9e6d6d3cd161afb8837cb13dd09
3
+ size 498612824
codebert-base-cross-entropy/checkpoint-2400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ce6dc1a5c937704f311162e0086e31ccb26673b31d2c6150a54b302f97f6ab8
3
+ size 4741859
codebert-base-cross-entropy/checkpoint-2400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fd9e4028a7acfc43907a0422d42bcd5dace812dedd6d7d186883f451858cdf9
3
+ size 14645
codebert-base-cross-entropy/checkpoint-2400/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54e30e6a2fc2de666ad3bd0097b38b5bd243339a48d5061ef4ab8fe7480885c6
3
+ size 1383
codebert-base-cross-entropy/checkpoint-2400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31de19de734a55e97e8568f4c3138a9a71ff7211dda32f22d92a5d16c71d9fe6
3
+ size 1465
codebert-base-cross-entropy/checkpoint-2400/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
codebert-base-cross-entropy/checkpoint-2400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/checkpoint-2400/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "<mask>",
52
+ "model_max_length": 512,
53
+ "pad_token": "<pad>",
54
+ "sep_token": "</s>",
55
+ "tokenizer_class": "RobertaTokenizer",
56
+ "trim_offsets": true,
57
+ "unk_token": "<unk>"
58
+ }
codebert-base-cross-entropy/checkpoint-2400/trainer_state.json ADDED
@@ -0,0 +1,1819 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2000,
3
+ "best_metric": 0.743516265066894,
4
+ "best_model_checkpoint": "./output_checkpoints/codebert-base-cross-entropy/checkpoint-2000",
5
+ "epoch": 0.3071803404582107,
6
+ "eval_steps": 500,
7
+ "global_step": 2400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0012799180852425445,
14
+ "grad_norm": 3.6954128742218018,
15
+ "learning_rate": 3.840409643695328e-08,
16
+ "loss": 0.7763,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.002559836170485089,
21
+ "grad_norm": 3.3660645484924316,
22
+ "learning_rate": 8.10753147002347e-08,
23
+ "loss": 0.7763,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0038397542557276334,
28
+ "grad_norm": 4.491067886352539,
29
+ "learning_rate": 1.2374653296351612e-07,
30
+ "loss": 0.7889,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.005119672340970178,
35
+ "grad_norm": 1.5920521020889282,
36
+ "learning_rate": 1.6641775122679754e-07,
37
+ "loss": 0.7888,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.006399590426212722,
42
+ "grad_norm": 4.410898208618164,
43
+ "learning_rate": 2.0908896949007894e-07,
44
+ "loss": 0.7692,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.007679508511455267,
49
+ "grad_norm": 2.5669264793395996,
50
+ "learning_rate": 2.517601877533604e-07,
51
+ "loss": 0.7735,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.008959426596697812,
56
+ "grad_norm": 6.190986156463623,
57
+ "learning_rate": 2.944314060166418e-07,
58
+ "loss": 0.7679,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.010239344681940356,
63
+ "grad_norm": 5.736413478851318,
64
+ "learning_rate": 3.371026242799232e-07,
65
+ "loss": 0.744,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.011519262767182901,
70
+ "grad_norm": 4.278866767883301,
71
+ "learning_rate": 3.7977384254320464e-07,
72
+ "loss": 0.7542,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.012799180852425445,
77
+ "grad_norm": 5.565455436706543,
78
+ "learning_rate": 4.22445060806486e-07,
79
+ "loss": 0.752,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.01407909893766799,
84
+ "grad_norm": 4.031781196594238,
85
+ "learning_rate": 4.651162790697675e-07,
86
+ "loss": 0.7614,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.015359017022910534,
91
+ "grad_norm": 2.484126329421997,
92
+ "learning_rate": 5.077874973330489e-07,
93
+ "loss": 0.7664,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.016638935108153077,
98
+ "grad_norm": 2.7525155544281006,
99
+ "learning_rate": 5.504587155963304e-07,
100
+ "loss": 0.7521,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.017918853193395624,
105
+ "grad_norm": 1.6374385356903076,
106
+ "learning_rate": 5.931299338596117e-07,
107
+ "loss": 0.7436,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.019198771278638168,
112
+ "grad_norm": 2.370774269104004,
113
+ "learning_rate": 6.358011521228932e-07,
114
+ "loss": 0.7429,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.02047868936388071,
119
+ "grad_norm": 2.8948566913604736,
120
+ "learning_rate": 6.784723703861745e-07,
121
+ "loss": 0.7333,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.021758607449123255,
126
+ "grad_norm": 2.7481186389923096,
127
+ "learning_rate": 7.21143588649456e-07,
128
+ "loss": 0.7294,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.023038525534365802,
133
+ "grad_norm": 2.0159404277801514,
134
+ "learning_rate": 7.638148069127374e-07,
135
+ "loss": 0.731,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.024318443619608346,
140
+ "grad_norm": 3.6898131370544434,
141
+ "learning_rate": 8.064860251760189e-07,
142
+ "loss": 0.7153,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.02559836170485089,
147
+ "grad_norm": 3.2842342853546143,
148
+ "learning_rate": 8.491572434393003e-07,
149
+ "loss": 0.7245,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.02559836170485089,
154
+ "eval_accuracy": 0.47695,
155
+ "eval_loss": 0.7148504853248596,
156
+ "eval_macro_f1": 0.3229290091066048,
157
+ "eval_precision": 0.238475,
158
+ "eval_recall": 0.5,
159
+ "eval_runtime": 1641.324,
160
+ "eval_samples_per_second": 60.926,
161
+ "eval_steps_per_second": 0.476,
162
+ "step": 200
163
+ },
164
+ {
165
+ "epoch": 0.026878279790093433,
166
+ "grad_norm": 2.294558048248291,
167
+ "learning_rate": 8.918284617025817e-07,
168
+ "loss": 0.718,
169
+ "step": 210
170
+ },
171
+ {
172
+ "epoch": 0.02815819787533598,
173
+ "grad_norm": 2.9467175006866455,
174
+ "learning_rate": 9.344996799658632e-07,
175
+ "loss": 0.7076,
176
+ "step": 220
177
+ },
178
+ {
179
+ "epoch": 0.029438115960578524,
180
+ "grad_norm": 2.6889193058013916,
181
+ "learning_rate": 9.771708982291445e-07,
182
+ "loss": 0.7019,
183
+ "step": 230
184
+ },
185
+ {
186
+ "epoch": 0.030718034045821067,
187
+ "grad_norm": 0.7491384744644165,
188
+ "learning_rate": 1.0198421164924258e-06,
189
+ "loss": 0.7046,
190
+ "step": 240
191
+ },
192
+ {
193
+ "epoch": 0.031997952131063614,
194
+ "grad_norm": 2.0363433361053467,
195
+ "learning_rate": 1.0625133347557074e-06,
196
+ "loss": 0.7085,
197
+ "step": 250
198
+ },
199
+ {
200
+ "epoch": 0.033277870216306155,
201
+ "grad_norm": 1.1431337594985962,
202
+ "learning_rate": 1.1051845530189888e-06,
203
+ "loss": 0.7008,
204
+ "step": 260
205
+ },
206
+ {
207
+ "epoch": 0.0345577883015487,
208
+ "grad_norm": 3.7735233306884766,
209
+ "learning_rate": 1.1478557712822702e-06,
210
+ "loss": 0.7015,
211
+ "step": 270
212
+ },
213
+ {
214
+ "epoch": 0.03583770638679125,
215
+ "grad_norm": 0.7889328002929688,
216
+ "learning_rate": 1.1905269895455517e-06,
217
+ "loss": 0.699,
218
+ "step": 280
219
+ },
220
+ {
221
+ "epoch": 0.03711762447203379,
222
+ "grad_norm": 0.8796226382255554,
223
+ "learning_rate": 1.233198207808833e-06,
224
+ "loss": 0.6922,
225
+ "step": 290
226
+ },
227
+ {
228
+ "epoch": 0.038397542557276336,
229
+ "grad_norm": 0.7966652512550354,
230
+ "learning_rate": 1.2758694260721145e-06,
231
+ "loss": 0.6949,
232
+ "step": 300
233
+ },
234
+ {
235
+ "epoch": 0.039677460642518876,
236
+ "grad_norm": 0.7029069066047668,
237
+ "learning_rate": 1.318540644335396e-06,
238
+ "loss": 0.695,
239
+ "step": 310
240
+ },
241
+ {
242
+ "epoch": 0.04095737872776142,
243
+ "grad_norm": 2.04327392578125,
244
+ "learning_rate": 1.3612118625986772e-06,
245
+ "loss": 0.6944,
246
+ "step": 320
247
+ },
248
+ {
249
+ "epoch": 0.04223729681300397,
250
+ "grad_norm": 0.918025553226471,
251
+ "learning_rate": 1.4038830808619588e-06,
252
+ "loss": 0.6863,
253
+ "step": 330
254
+ },
255
+ {
256
+ "epoch": 0.04351721489824651,
257
+ "grad_norm": 0.9528706073760986,
258
+ "learning_rate": 1.4465542991252401e-06,
259
+ "loss": 0.6907,
260
+ "step": 340
261
+ },
262
+ {
263
+ "epoch": 0.04479713298348906,
264
+ "grad_norm": 1.071359634399414,
265
+ "learning_rate": 1.4892255173885215e-06,
266
+ "loss": 0.6873,
267
+ "step": 350
268
+ },
269
+ {
270
+ "epoch": 0.046077051068731605,
271
+ "grad_norm": 1.9875625371932983,
272
+ "learning_rate": 1.531896735651803e-06,
273
+ "loss": 0.6942,
274
+ "step": 360
275
+ },
276
+ {
277
+ "epoch": 0.047356969153974145,
278
+ "grad_norm": 2.0222439765930176,
279
+ "learning_rate": 1.5745679539150842e-06,
280
+ "loss": 0.6898,
281
+ "step": 370
282
+ },
283
+ {
284
+ "epoch": 0.04863688723921669,
285
+ "grad_norm": 1.7827857732772827,
286
+ "learning_rate": 1.6172391721783658e-06,
287
+ "loss": 0.6909,
288
+ "step": 380
289
+ },
290
+ {
291
+ "epoch": 0.04991680532445923,
292
+ "grad_norm": 1.065674901008606,
293
+ "learning_rate": 1.6599103904416472e-06,
294
+ "loss": 0.6877,
295
+ "step": 390
296
+ },
297
+ {
298
+ "epoch": 0.05119672340970178,
299
+ "grad_norm": 0.8346193432807922,
300
+ "learning_rate": 1.7025816087049288e-06,
301
+ "loss": 0.6889,
302
+ "step": 400
303
+ },
304
+ {
305
+ "epoch": 0.05119672340970178,
306
+ "eval_accuracy": 0.54043,
307
+ "eval_loss": 0.6827595233917236,
308
+ "eval_macro_f1": 0.41833840347598905,
309
+ "eval_precision": 0.5839205994464474,
310
+ "eval_recall": 0.5204152767503025,
311
+ "eval_runtime": 1634.6244,
312
+ "eval_samples_per_second": 61.176,
313
+ "eval_steps_per_second": 0.478,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 0.052476641494944326,
318
+ "grad_norm": 0.8074935674667358,
319
+ "learning_rate": 1.7452528269682101e-06,
320
+ "loss": 0.684,
321
+ "step": 410
322
+ },
323
+ {
324
+ "epoch": 0.053756559580186866,
325
+ "grad_norm": 0.8965069651603699,
326
+ "learning_rate": 1.7879240452314913e-06,
327
+ "loss": 0.6922,
328
+ "step": 420
329
+ },
330
+ {
331
+ "epoch": 0.05503647766542941,
332
+ "grad_norm": 2.597215414047241,
333
+ "learning_rate": 1.8305952634947729e-06,
334
+ "loss": 0.6909,
335
+ "step": 430
336
+ },
337
+ {
338
+ "epoch": 0.05631639575067196,
339
+ "grad_norm": 0.8233373165130615,
340
+ "learning_rate": 1.8732664817580542e-06,
341
+ "loss": 0.6831,
342
+ "step": 440
343
+ },
344
+ {
345
+ "epoch": 0.0575963138359145,
346
+ "grad_norm": 1.0624489784240723,
347
+ "learning_rate": 1.915937700021336e-06,
348
+ "loss": 0.6817,
349
+ "step": 450
350
+ },
351
+ {
352
+ "epoch": 0.05887623192115705,
353
+ "grad_norm": 2.0623559951782227,
354
+ "learning_rate": 1.958608918284617e-06,
355
+ "loss": 0.6819,
356
+ "step": 460
357
+ },
358
+ {
359
+ "epoch": 0.06015615000639959,
360
+ "grad_norm": 0.912433922290802,
361
+ "learning_rate": 2.0012801365478988e-06,
362
+ "loss": 0.686,
363
+ "step": 470
364
+ },
365
+ {
366
+ "epoch": 0.061436068091642135,
367
+ "grad_norm": 0.6926271319389343,
368
+ "learning_rate": 2.04395135481118e-06,
369
+ "loss": 0.6834,
370
+ "step": 480
371
+ },
372
+ {
373
+ "epoch": 0.06271598617688467,
374
+ "grad_norm": 1.498634934425354,
375
+ "learning_rate": 2.0866225730744615e-06,
376
+ "loss": 0.6805,
377
+ "step": 490
378
+ },
379
+ {
380
+ "epoch": 0.06399590426212723,
381
+ "grad_norm": 1.1818156242370605,
382
+ "learning_rate": 2.129293791337743e-06,
383
+ "loss": 0.6805,
384
+ "step": 500
385
+ },
386
+ {
387
+ "epoch": 0.06527582234736977,
388
+ "grad_norm": 1.4476051330566406,
389
+ "learning_rate": 2.1719650096010242e-06,
390
+ "loss": 0.6773,
391
+ "step": 510
392
+ },
393
+ {
394
+ "epoch": 0.06655574043261231,
395
+ "grad_norm": 1.1143743991851807,
396
+ "learning_rate": 2.214636227864306e-06,
397
+ "loss": 0.6801,
398
+ "step": 520
399
+ },
400
+ {
401
+ "epoch": 0.06783565851785486,
402
+ "grad_norm": 1.7030247449874878,
403
+ "learning_rate": 2.257307446127587e-06,
404
+ "loss": 0.6794,
405
+ "step": 530
406
+ },
407
+ {
408
+ "epoch": 0.0691155766030974,
409
+ "grad_norm": 0.8592290282249451,
410
+ "learning_rate": 2.2999786643908685e-06,
411
+ "loss": 0.6736,
412
+ "step": 540
413
+ },
414
+ {
415
+ "epoch": 0.07039549468833994,
416
+ "grad_norm": 0.8231940865516663,
417
+ "learning_rate": 2.34264988265415e-06,
418
+ "loss": 0.67,
419
+ "step": 550
420
+ },
421
+ {
422
+ "epoch": 0.0716754127735825,
423
+ "grad_norm": 0.88897705078125,
424
+ "learning_rate": 2.3853211009174317e-06,
425
+ "loss": 0.685,
426
+ "step": 560
427
+ },
428
+ {
429
+ "epoch": 0.07295533085882504,
430
+ "grad_norm": 1.6297541856765747,
431
+ "learning_rate": 2.427992319180713e-06,
432
+ "loss": 0.6797,
433
+ "step": 570
434
+ },
435
+ {
436
+ "epoch": 0.07423524894406758,
437
+ "grad_norm": 2.9297447204589844,
438
+ "learning_rate": 2.470663537443994e-06,
439
+ "loss": 0.6766,
440
+ "step": 580
441
+ },
442
+ {
443
+ "epoch": 0.07551516702931012,
444
+ "grad_norm": 0.9277703762054443,
445
+ "learning_rate": 2.5133347557072756e-06,
446
+ "loss": 0.6824,
447
+ "step": 590
448
+ },
449
+ {
450
+ "epoch": 0.07679508511455267,
451
+ "grad_norm": 0.5883333683013916,
452
+ "learning_rate": 2.556005973970557e-06,
453
+ "loss": 0.6818,
454
+ "step": 600
455
+ },
456
+ {
457
+ "epoch": 0.07679508511455267,
458
+ "eval_accuracy": 0.63909,
459
+ "eval_loss": 0.6700075268745422,
460
+ "eval_macro_f1": 0.621848143053056,
461
+ "eval_precision": 0.6524313071088819,
462
+ "eval_recall": 0.6305863955136095,
463
+ "eval_runtime": 1633.9732,
464
+ "eval_samples_per_second": 61.201,
465
+ "eval_steps_per_second": 0.479,
466
+ "step": 600
467
+ },
468
+ {
469
+ "epoch": 0.07807500319979521,
470
+ "grad_norm": 0.7149448394775391,
471
+ "learning_rate": 2.5986771922338383e-06,
472
+ "loss": 0.6746,
473
+ "step": 610
474
+ },
475
+ {
476
+ "epoch": 0.07935492128503775,
477
+ "grad_norm": 2.0167856216430664,
478
+ "learning_rate": 2.64134841049712e-06,
479
+ "loss": 0.682,
480
+ "step": 620
481
+ },
482
+ {
483
+ "epoch": 0.0806348393702803,
484
+ "grad_norm": 1.303001880645752,
485
+ "learning_rate": 2.6840196287604015e-06,
486
+ "loss": 0.6735,
487
+ "step": 630
488
+ },
489
+ {
490
+ "epoch": 0.08191475745552285,
491
+ "grad_norm": 2.0516321659088135,
492
+ "learning_rate": 2.7266908470236826e-06,
493
+ "loss": 0.6705,
494
+ "step": 640
495
+ },
496
+ {
497
+ "epoch": 0.08319467554076539,
498
+ "grad_norm": 2.242105484008789,
499
+ "learning_rate": 2.7693620652869642e-06,
500
+ "loss": 0.6798,
501
+ "step": 650
502
+ },
503
+ {
504
+ "epoch": 0.08447459362600794,
505
+ "grad_norm": 1.1836538314819336,
506
+ "learning_rate": 2.812033283550246e-06,
507
+ "loss": 0.6684,
508
+ "step": 660
509
+ },
510
+ {
511
+ "epoch": 0.08575451171125048,
512
+ "grad_norm": 0.8640874624252319,
513
+ "learning_rate": 2.8547045018135274e-06,
514
+ "loss": 0.6783,
515
+ "step": 670
516
+ },
517
+ {
518
+ "epoch": 0.08703442979649302,
519
+ "grad_norm": 0.6901177763938904,
520
+ "learning_rate": 2.897375720076808e-06,
521
+ "loss": 0.6724,
522
+ "step": 680
523
+ },
524
+ {
525
+ "epoch": 0.08831434788173557,
526
+ "grad_norm": 1.3443949222564697,
527
+ "learning_rate": 2.9400469383400897e-06,
528
+ "loss": 0.67,
529
+ "step": 690
530
+ },
531
+ {
532
+ "epoch": 0.08959426596697811,
533
+ "grad_norm": 0.7775189280509949,
534
+ "learning_rate": 2.9827181566033713e-06,
535
+ "loss": 0.6733,
536
+ "step": 700
537
+ },
538
+ {
539
+ "epoch": 0.09087418405222066,
540
+ "grad_norm": 1.16916823387146,
541
+ "learning_rate": 3.025389374866653e-06,
542
+ "loss": 0.6713,
543
+ "step": 710
544
+ },
545
+ {
546
+ "epoch": 0.09215410213746321,
547
+ "grad_norm": 1.707152009010315,
548
+ "learning_rate": 3.068060593129934e-06,
549
+ "loss": 0.6659,
550
+ "step": 720
551
+ },
552
+ {
553
+ "epoch": 0.09343402022270575,
554
+ "grad_norm": 0.9780622124671936,
555
+ "learning_rate": 3.1107318113932156e-06,
556
+ "loss": 0.6639,
557
+ "step": 730
558
+ },
559
+ {
560
+ "epoch": 0.09471393830794829,
561
+ "grad_norm": 2.3687374591827393,
562
+ "learning_rate": 3.153403029656497e-06,
563
+ "loss": 0.6666,
564
+ "step": 740
565
+ },
566
+ {
567
+ "epoch": 0.09599385639319083,
568
+ "grad_norm": 0.8506460189819336,
569
+ "learning_rate": 3.1960742479197783e-06,
570
+ "loss": 0.6778,
571
+ "step": 750
572
+ },
573
+ {
574
+ "epoch": 0.09727377447843338,
575
+ "grad_norm": 1.0102107524871826,
576
+ "learning_rate": 3.23874546618306e-06,
577
+ "loss": 0.6635,
578
+ "step": 760
579
+ },
580
+ {
581
+ "epoch": 0.09855369256367592,
582
+ "grad_norm": 1.1480194330215454,
583
+ "learning_rate": 3.2814166844463415e-06,
584
+ "loss": 0.6709,
585
+ "step": 770
586
+ },
587
+ {
588
+ "epoch": 0.09983361064891846,
589
+ "grad_norm": 1.3119179010391235,
590
+ "learning_rate": 3.324087902709623e-06,
591
+ "loss": 0.6564,
592
+ "step": 780
593
+ },
594
+ {
595
+ "epoch": 0.10111352873416102,
596
+ "grad_norm": 0.681377649307251,
597
+ "learning_rate": 3.3667591209729038e-06,
598
+ "loss": 0.6656,
599
+ "step": 790
600
+ },
601
+ {
602
+ "epoch": 0.10239344681940356,
603
+ "grad_norm": 1.022257685661316,
604
+ "learning_rate": 3.4094303392361854e-06,
605
+ "loss": 0.6709,
606
+ "step": 800
607
+ },
608
+ {
609
+ "epoch": 0.10239344681940356,
610
+ "eval_accuracy": 0.66247,
611
+ "eval_loss": 0.6552017331123352,
612
+ "eval_macro_f1": 0.6514270557181998,
613
+ "eval_precision": 0.6718270655389242,
614
+ "eval_recall": 0.6556580720913894,
615
+ "eval_runtime": 1633.5478,
616
+ "eval_samples_per_second": 61.216,
617
+ "eval_steps_per_second": 0.479,
618
+ "step": 800
619
+ },
620
+ {
621
+ "epoch": 0.1036733649046461,
622
+ "grad_norm": 1.0268975496292114,
623
+ "learning_rate": 3.452101557499467e-06,
624
+ "loss": 0.6602,
625
+ "step": 810
626
+ },
627
+ {
628
+ "epoch": 0.10495328298988865,
629
+ "grad_norm": 2.6206603050231934,
630
+ "learning_rate": 3.494772775762748e-06,
631
+ "loss": 0.6705,
632
+ "step": 820
633
+ },
634
+ {
635
+ "epoch": 0.10623320107513119,
636
+ "grad_norm": 1.035157561302185,
637
+ "learning_rate": 3.5374439940260297e-06,
638
+ "loss": 0.6643,
639
+ "step": 830
640
+ },
641
+ {
642
+ "epoch": 0.10751311916037373,
643
+ "grad_norm": 1.736412763595581,
644
+ "learning_rate": 3.5801152122893113e-06,
645
+ "loss": 0.6612,
646
+ "step": 840
647
+ },
648
+ {
649
+ "epoch": 0.10879303724561629,
650
+ "grad_norm": 0.6906964778900146,
651
+ "learning_rate": 3.622786430552593e-06,
652
+ "loss": 0.6576,
653
+ "step": 850
654
+ },
655
+ {
656
+ "epoch": 0.11007295533085883,
657
+ "grad_norm": 0.8148729801177979,
658
+ "learning_rate": 3.665457648815874e-06,
659
+ "loss": 0.6561,
660
+ "step": 860
661
+ },
662
+ {
663
+ "epoch": 0.11135287341610137,
664
+ "grad_norm": 1.0217723846435547,
665
+ "learning_rate": 3.7081288670791556e-06,
666
+ "loss": 0.6581,
667
+ "step": 870
668
+ },
669
+ {
670
+ "epoch": 0.11263279150134392,
671
+ "grad_norm": 1.2743115425109863,
672
+ "learning_rate": 3.750800085342437e-06,
673
+ "loss": 0.6606,
674
+ "step": 880
675
+ },
676
+ {
677
+ "epoch": 0.11391270958658646,
678
+ "grad_norm": 0.6405361294746399,
679
+ "learning_rate": 3.793471303605718e-06,
680
+ "loss": 0.652,
681
+ "step": 890
682
+ },
683
+ {
684
+ "epoch": 0.115192627671829,
685
+ "grad_norm": 0.8357161283493042,
686
+ "learning_rate": 3.836142521869e-06,
687
+ "loss": 0.6558,
688
+ "step": 900
689
+ },
690
+ {
691
+ "epoch": 0.11647254575707154,
692
+ "grad_norm": 1.3505173921585083,
693
+ "learning_rate": 3.8788137401322815e-06,
694
+ "loss": 0.6576,
695
+ "step": 910
696
+ },
697
+ {
698
+ "epoch": 0.1177524638423141,
699
+ "grad_norm": 1.2253319025039673,
700
+ "learning_rate": 3.921484958395563e-06,
701
+ "loss": 0.6548,
702
+ "step": 920
703
+ },
704
+ {
705
+ "epoch": 0.11903238192755664,
706
+ "grad_norm": 0.6269568800926208,
707
+ "learning_rate": 3.964156176658844e-06,
708
+ "loss": 0.6562,
709
+ "step": 930
710
+ },
711
+ {
712
+ "epoch": 0.12031230001279918,
713
+ "grad_norm": 0.9413424134254456,
714
+ "learning_rate": 4.006827394922125e-06,
715
+ "loss": 0.6461,
716
+ "step": 940
717
+ },
718
+ {
719
+ "epoch": 0.12159221809804173,
720
+ "grad_norm": 1.0903513431549072,
721
+ "learning_rate": 4.049498613185407e-06,
722
+ "loss": 0.6457,
723
+ "step": 950
724
+ },
725
+ {
726
+ "epoch": 0.12287213618328427,
727
+ "grad_norm": 2.0023152828216553,
728
+ "learning_rate": 4.092169831448688e-06,
729
+ "loss": 0.6458,
730
+ "step": 960
731
+ },
732
+ {
733
+ "epoch": 0.12415205426852681,
734
+ "grad_norm": 0.9638109803199768,
735
+ "learning_rate": 4.134841049711969e-06,
736
+ "loss": 0.6486,
737
+ "step": 970
738
+ },
739
+ {
740
+ "epoch": 0.12543197235376935,
741
+ "grad_norm": 0.9894546270370483,
742
+ "learning_rate": 4.177512267975251e-06,
743
+ "loss": 0.6453,
744
+ "step": 980
745
+ },
746
+ {
747
+ "epoch": 0.1267118904390119,
748
+ "grad_norm": 1.2289057970046997,
749
+ "learning_rate": 4.220183486238532e-06,
750
+ "loss": 0.6605,
751
+ "step": 990
752
+ },
753
+ {
754
+ "epoch": 0.12799180852425446,
755
+ "grad_norm": 1.119125247001648,
756
+ "learning_rate": 4.262854704501814e-06,
757
+ "loss": 0.6564,
758
+ "step": 1000
759
+ },
760
+ {
761
+ "epoch": 0.12799180852425446,
762
+ "eval_accuracy": 0.69068,
763
+ "eval_loss": 0.6396247148513794,
764
+ "eval_macro_f1": 0.689588440856387,
765
+ "eval_precision": 0.6900076024604327,
766
+ "eval_recall": 0.6894114140312133,
767
+ "eval_runtime": 1634.4321,
768
+ "eval_samples_per_second": 61.183,
769
+ "eval_steps_per_second": 0.478,
770
+ "step": 1000
771
+ },
772
+ {
773
+ "epoch": 0.12927172660949698,
774
+ "grad_norm": 2.02966046333313,
775
+ "learning_rate": 4.3055259227650956e-06,
776
+ "loss": 0.646,
777
+ "step": 1010
778
+ },
779
+ {
780
+ "epoch": 0.13055164469473954,
781
+ "grad_norm": 0.7430800199508667,
782
+ "learning_rate": 4.348197141028377e-06,
783
+ "loss": 0.6453,
784
+ "step": 1020
785
+ },
786
+ {
787
+ "epoch": 0.1318315627799821,
788
+ "grad_norm": 1.031962275505066,
789
+ "learning_rate": 4.390868359291658e-06,
790
+ "loss": 0.6425,
791
+ "step": 1030
792
+ },
793
+ {
794
+ "epoch": 0.13311148086522462,
795
+ "grad_norm": 1.0249106884002686,
796
+ "learning_rate": 4.4335395775549394e-06,
797
+ "loss": 0.6426,
798
+ "step": 1040
799
+ },
800
+ {
801
+ "epoch": 0.13439139895046717,
802
+ "grad_norm": 0.7375448942184448,
803
+ "learning_rate": 4.476210795818221e-06,
804
+ "loss": 0.6401,
805
+ "step": 1050
806
+ },
807
+ {
808
+ "epoch": 0.13567131703570973,
809
+ "grad_norm": 1.295983076095581,
810
+ "learning_rate": 4.518882014081503e-06,
811
+ "loss": 0.6449,
812
+ "step": 1060
813
+ },
814
+ {
815
+ "epoch": 0.13695123512095225,
816
+ "grad_norm": 1.6978998184204102,
817
+ "learning_rate": 4.561553232344783e-06,
818
+ "loss": 0.6484,
819
+ "step": 1070
820
+ },
821
+ {
822
+ "epoch": 0.1382311532061948,
823
+ "grad_norm": 0.7747512459754944,
824
+ "learning_rate": 4.604224450608065e-06,
825
+ "loss": 0.6526,
826
+ "step": 1080
827
+ },
828
+ {
829
+ "epoch": 0.13951107129143736,
830
+ "grad_norm": 0.7895913124084473,
831
+ "learning_rate": 4.6468956688713465e-06,
832
+ "loss": 0.6311,
833
+ "step": 1090
834
+ },
835
+ {
836
+ "epoch": 0.1407909893766799,
837
+ "grad_norm": 0.6975072026252747,
838
+ "learning_rate": 4.689566887134628e-06,
839
+ "loss": 0.6454,
840
+ "step": 1100
841
+ },
842
+ {
843
+ "epoch": 0.14207090746192244,
844
+ "grad_norm": 0.8316811919212341,
845
+ "learning_rate": 4.73223810539791e-06,
846
+ "loss": 0.6447,
847
+ "step": 1110
848
+ },
849
+ {
850
+ "epoch": 0.143350825547165,
851
+ "grad_norm": 0.6887441277503967,
852
+ "learning_rate": 4.774909323661191e-06,
853
+ "loss": 0.6325,
854
+ "step": 1120
855
+ },
856
+ {
857
+ "epoch": 0.14463074363240752,
858
+ "grad_norm": 1.0218420028686523,
859
+ "learning_rate": 4.817580541924473e-06,
860
+ "loss": 0.6308,
861
+ "step": 1130
862
+ },
863
+ {
864
+ "epoch": 0.14591066171765008,
865
+ "grad_norm": 0.7663633823394775,
866
+ "learning_rate": 4.8602517601877535e-06,
867
+ "loss": 0.6227,
868
+ "step": 1140
869
+ },
870
+ {
871
+ "epoch": 0.1471905798028926,
872
+ "grad_norm": 1.5115224123001099,
873
+ "learning_rate": 4.902922978451035e-06,
874
+ "loss": 0.6315,
875
+ "step": 1150
876
+ },
877
+ {
878
+ "epoch": 0.14847049788813516,
879
+ "grad_norm": 0.8011078238487244,
880
+ "learning_rate": 4.945594196714317e-06,
881
+ "loss": 0.6428,
882
+ "step": 1160
883
+ },
884
+ {
885
+ "epoch": 0.1497504159733777,
886
+ "grad_norm": 1.3246116638183594,
887
+ "learning_rate": 4.9882654149775974e-06,
888
+ "loss": 0.6397,
889
+ "step": 1170
890
+ },
891
+ {
892
+ "epoch": 0.15103033405862024,
893
+ "grad_norm": 0.979498565196991,
894
+ "learning_rate": 5.030936633240879e-06,
895
+ "loss": 0.6308,
896
+ "step": 1180
897
+ },
898
+ {
899
+ "epoch": 0.1523102521438628,
900
+ "grad_norm": 0.8756135106086731,
901
+ "learning_rate": 5.073607851504161e-06,
902
+ "loss": 0.6373,
903
+ "step": 1190
904
+ },
905
+ {
906
+ "epoch": 0.15359017022910534,
907
+ "grad_norm": 0.845967710018158,
908
+ "learning_rate": 5.116279069767442e-06,
909
+ "loss": 0.6322,
910
+ "step": 1200
911
+ },
912
+ {
913
+ "epoch": 0.15359017022910534,
914
+ "eval_accuracy": 0.69931,
915
+ "eval_loss": 0.6224722862243652,
916
+ "eval_macro_f1": 0.6985627236986651,
917
+ "eval_precision": 0.6986454338713279,
918
+ "eval_recall": 0.6984991383538209,
919
+ "eval_runtime": 692.1526,
920
+ "eval_samples_per_second": 144.477,
921
+ "eval_steps_per_second": 2.258,
922
+ "step": 1200
923
+ },
924
+ {
925
+ "epoch": 0.15487008831434787,
926
+ "grad_norm": 2.0670483112335205,
927
+ "learning_rate": 1.032010243277849e-05,
928
+ "loss": 0.6386,
929
+ "step": 1210
930
+ },
931
+ {
932
+ "epoch": 0.15615000639959042,
933
+ "grad_norm": 2.482168436050415,
934
+ "learning_rate": 1.0405463081519421e-05,
935
+ "loss": 0.6281,
936
+ "step": 1220
937
+ },
938
+ {
939
+ "epoch": 0.15742992448483298,
940
+ "grad_norm": 1.4530706405639648,
941
+ "learning_rate": 1.049082373026035e-05,
942
+ "loss": 0.6277,
943
+ "step": 1230
944
+ },
945
+ {
946
+ "epoch": 0.1587098425700755,
947
+ "grad_norm": 2.9233577251434326,
948
+ "learning_rate": 1.0576184379001282e-05,
949
+ "loss": 0.6363,
950
+ "step": 1240
951
+ },
952
+ {
953
+ "epoch": 0.15998976065531806,
954
+ "grad_norm": 0.6975642442703247,
955
+ "learning_rate": 1.0661545027742212e-05,
956
+ "loss": 0.6344,
957
+ "step": 1250
958
+ },
959
+ {
960
+ "epoch": 0.1612696787405606,
961
+ "grad_norm": 2.3634731769561768,
962
+ "learning_rate": 1.0746905676483142e-05,
963
+ "loss": 0.6217,
964
+ "step": 1260
965
+ },
966
+ {
967
+ "epoch": 0.16254959682580314,
968
+ "grad_norm": 0.8812585473060608,
969
+ "learning_rate": 1.0832266325224072e-05,
970
+ "loss": 0.6311,
971
+ "step": 1270
972
+ },
973
+ {
974
+ "epoch": 0.1638295149110457,
975
+ "grad_norm": 1.8986798524856567,
976
+ "learning_rate": 1.0917626973965003e-05,
977
+ "loss": 0.6304,
978
+ "step": 1280
979
+ },
980
+ {
981
+ "epoch": 0.16510943299628825,
982
+ "grad_norm": 0.8174984455108643,
983
+ "learning_rate": 1.1002987622705935e-05,
984
+ "loss": 0.6176,
985
+ "step": 1290
986
+ },
987
+ {
988
+ "epoch": 0.16638935108153077,
989
+ "grad_norm": 1.0268642902374268,
990
+ "learning_rate": 1.1088348271446865e-05,
991
+ "loss": 0.6233,
992
+ "step": 1300
993
+ },
994
+ {
995
+ "epoch": 0.16766926916677333,
996
+ "grad_norm": 1.047821044921875,
997
+ "learning_rate": 1.1173708920187793e-05,
998
+ "loss": 0.6127,
999
+ "step": 1310
1000
+ },
1001
+ {
1002
+ "epoch": 0.16894918725201588,
1003
+ "grad_norm": 0.9875057935714722,
1004
+ "learning_rate": 1.1259069568928724e-05,
1005
+ "loss": 0.6211,
1006
+ "step": 1320
1007
+ },
1008
+ {
1009
+ "epoch": 0.1702291053372584,
1010
+ "grad_norm": 1.0082708597183228,
1011
+ "learning_rate": 1.1344430217669656e-05,
1012
+ "loss": 0.6202,
1013
+ "step": 1330
1014
+ },
1015
+ {
1016
+ "epoch": 0.17150902342250096,
1017
+ "grad_norm": 0.7229795455932617,
1018
+ "learning_rate": 1.1429790866410586e-05,
1019
+ "loss": 0.6071,
1020
+ "step": 1340
1021
+ },
1022
+ {
1023
+ "epoch": 0.17278894150774352,
1024
+ "grad_norm": 0.7687681317329407,
1025
+ "learning_rate": 1.1515151515151517e-05,
1026
+ "loss": 0.6218,
1027
+ "step": 1350
1028
+ },
1029
+ {
1030
+ "epoch": 0.17406885959298604,
1031
+ "grad_norm": 1.1643378734588623,
1032
+ "learning_rate": 1.1600512163892445e-05,
1033
+ "loss": 0.6208,
1034
+ "step": 1360
1035
+ },
1036
+ {
1037
+ "epoch": 0.1753487776782286,
1038
+ "grad_norm": 0.7338509559631348,
1039
+ "learning_rate": 1.1685872812633377e-05,
1040
+ "loss": 0.6105,
1041
+ "step": 1370
1042
+ },
1043
+ {
1044
+ "epoch": 0.17662869576347115,
1045
+ "grad_norm": 0.7444656491279602,
1046
+ "learning_rate": 1.1771233461374307e-05,
1047
+ "loss": 0.6183,
1048
+ "step": 1380
1049
+ },
1050
+ {
1051
+ "epoch": 0.17790861384871368,
1052
+ "grad_norm": 1.1902283430099487,
1053
+ "learning_rate": 1.1856594110115238e-05,
1054
+ "loss": 0.6209,
1055
+ "step": 1390
1056
+ },
1057
+ {
1058
+ "epoch": 0.17918853193395623,
1059
+ "grad_norm": 1.0872228145599365,
1060
+ "learning_rate": 1.194195475885617e-05,
1061
+ "loss": 0.6172,
1062
+ "step": 1400
1063
+ },
1064
+ {
1065
+ "epoch": 0.18046845001919878,
1066
+ "grad_norm": 1.8439041376113892,
1067
+ "learning_rate": 1.2027315407597098e-05,
1068
+ "loss": 0.6086,
1069
+ "step": 1410
1070
+ },
1071
+ {
1072
+ "epoch": 0.1817483681044413,
1073
+ "grad_norm": 0.8699899911880493,
1074
+ "learning_rate": 1.211267605633803e-05,
1075
+ "loss": 0.6247,
1076
+ "step": 1420
1077
+ },
1078
+ {
1079
+ "epoch": 0.18302828618968386,
1080
+ "grad_norm": 1.1468650102615356,
1081
+ "learning_rate": 1.2198036705078959e-05,
1082
+ "loss": 0.6114,
1083
+ "step": 1430
1084
+ },
1085
+ {
1086
+ "epoch": 0.18430820427492642,
1087
+ "grad_norm": 1.0204615592956543,
1088
+ "learning_rate": 1.228339735381989e-05,
1089
+ "loss": 0.6015,
1090
+ "step": 1440
1091
+ },
1092
+ {
1093
+ "epoch": 0.18558812236016894,
1094
+ "grad_norm": 2.660583257675171,
1095
+ "learning_rate": 1.236875800256082e-05,
1096
+ "loss": 0.6002,
1097
+ "step": 1450
1098
+ },
1099
+ {
1100
+ "epoch": 0.1868680404454115,
1101
+ "grad_norm": 1.1916371583938599,
1102
+ "learning_rate": 1.2454118651301752e-05,
1103
+ "loss": 0.6076,
1104
+ "step": 1460
1105
+ },
1106
+ {
1107
+ "epoch": 0.18814795853065402,
1108
+ "grad_norm": 0.8097987174987793,
1109
+ "learning_rate": 1.253947930004268e-05,
1110
+ "loss": 0.6126,
1111
+ "step": 1470
1112
+ },
1113
+ {
1114
+ "epoch": 0.18942787661589658,
1115
+ "grad_norm": 1.978747010231018,
1116
+ "learning_rate": 1.2624839948783612e-05,
1117
+ "loss": 0.5962,
1118
+ "step": 1480
1119
+ },
1120
+ {
1121
+ "epoch": 0.19070779470113913,
1122
+ "grad_norm": 1.0975425243377686,
1123
+ "learning_rate": 1.2710200597524543e-05,
1124
+ "loss": 0.587,
1125
+ "step": 1490
1126
+ },
1127
+ {
1128
+ "epoch": 0.19198771278638166,
1129
+ "grad_norm": 2.150351047515869,
1130
+ "learning_rate": 1.2795561246265473e-05,
1131
+ "loss": 0.5962,
1132
+ "step": 1500
1133
+ },
1134
+ {
1135
+ "epoch": 0.19198771278638166,
1136
+ "eval_accuracy": 0.72526,
1137
+ "eval_loss": 0.5823243856430054,
1138
+ "eval_macro_f1": 0.7252418839884757,
1139
+ "eval_precision": 0.7275813809678697,
1140
+ "eval_recall": 0.7271797416587706,
1141
+ "eval_runtime": 708.4119,
1142
+ "eval_samples_per_second": 141.161,
1143
+ "eval_steps_per_second": 1.104,
1144
+ "step": 1500
1145
+ },
1146
+ {
1147
+ "epoch": 0.1932676308716242,
1148
+ "grad_norm": 1.8355706930160522,
1149
+ "learning_rate": 1.2880921895006405e-05,
1150
+ "loss": 0.6133,
1151
+ "step": 1510
1152
+ },
1153
+ {
1154
+ "epoch": 0.19454754895686677,
1155
+ "grad_norm": 1.4349753856658936,
1156
+ "learning_rate": 1.2966282543747333e-05,
1157
+ "loss": 0.6062,
1158
+ "step": 1520
1159
+ },
1160
+ {
1161
+ "epoch": 0.1958274670421093,
1162
+ "grad_norm": 1.8322596549987793,
1163
+ "learning_rate": 1.3051643192488264e-05,
1164
+ "loss": 0.5982,
1165
+ "step": 1530
1166
+ },
1167
+ {
1168
+ "epoch": 0.19710738512735185,
1169
+ "grad_norm": 1.509338617324829,
1170
+ "learning_rate": 1.3137003841229194e-05,
1171
+ "loss": 0.5961,
1172
+ "step": 1540
1173
+ },
1174
+ {
1175
+ "epoch": 0.1983873032125944,
1176
+ "grad_norm": 0.9844911098480225,
1177
+ "learning_rate": 1.3222364489970126e-05,
1178
+ "loss": 0.583,
1179
+ "step": 1550
1180
+ },
1181
+ {
1182
+ "epoch": 0.19966722129783693,
1183
+ "grad_norm": 1.035900354385376,
1184
+ "learning_rate": 1.3307725138711055e-05,
1185
+ "loss": 0.598,
1186
+ "step": 1560
1187
+ },
1188
+ {
1189
+ "epoch": 0.20094713938307948,
1190
+ "grad_norm": 0.7891924977302551,
1191
+ "learning_rate": 1.3393085787451985e-05,
1192
+ "loss": 0.6064,
1193
+ "step": 1570
1194
+ },
1195
+ {
1196
+ "epoch": 0.20222705746832204,
1197
+ "grad_norm": 1.4326112270355225,
1198
+ "learning_rate": 1.3478446436192915e-05,
1199
+ "loss": 0.5954,
1200
+ "step": 1580
1201
+ },
1202
+ {
1203
+ "epoch": 0.20350697555356456,
1204
+ "grad_norm": 0.9054266214370728,
1205
+ "learning_rate": 1.3563807084933847e-05,
1206
+ "loss": 0.59,
1207
+ "step": 1590
1208
+ },
1209
+ {
1210
+ "epoch": 0.20478689363880712,
1211
+ "grad_norm": 1.0216469764709473,
1212
+ "learning_rate": 1.3649167733674778e-05,
1213
+ "loss": 0.5845,
1214
+ "step": 1600
1215
+ },
1216
+ {
1217
+ "epoch": 0.20606681172404967,
1218
+ "grad_norm": 1.0591177940368652,
1219
+ "learning_rate": 1.3734528382415708e-05,
1220
+ "loss": 0.599,
1221
+ "step": 1610
1222
+ },
1223
+ {
1224
+ "epoch": 0.2073467298092922,
1225
+ "grad_norm": 0.7826931476593018,
1226
+ "learning_rate": 1.3819889031156638e-05,
1227
+ "loss": 0.5898,
1228
+ "step": 1620
1229
+ },
1230
+ {
1231
+ "epoch": 0.20862664789453475,
1232
+ "grad_norm": 0.755007803440094,
1233
+ "learning_rate": 1.3905249679897568e-05,
1234
+ "loss": 0.5769,
1235
+ "step": 1630
1236
+ },
1237
+ {
1238
+ "epoch": 0.2099065659797773,
1239
+ "grad_norm": 1.3526090383529663,
1240
+ "learning_rate": 1.3990610328638499e-05,
1241
+ "loss": 0.5819,
1242
+ "step": 1640
1243
+ },
1244
+ {
1245
+ "epoch": 0.21118648406501983,
1246
+ "grad_norm": 0.7078200578689575,
1247
+ "learning_rate": 1.4075970977379429e-05,
1248
+ "loss": 0.5884,
1249
+ "step": 1650
1250
+ },
1251
+ {
1252
+ "epoch": 0.21246640215026238,
1253
+ "grad_norm": 0.80736243724823,
1254
+ "learning_rate": 1.416133162612036e-05,
1255
+ "loss": 0.5991,
1256
+ "step": 1660
1257
+ },
1258
+ {
1259
+ "epoch": 0.21374632023550494,
1260
+ "grad_norm": 0.6882274150848389,
1261
+ "learning_rate": 1.4246692274861289e-05,
1262
+ "loss": 0.5825,
1263
+ "step": 1670
1264
+ },
1265
+ {
1266
+ "epoch": 0.21502623832074746,
1267
+ "grad_norm": 0.7223654985427856,
1268
+ "learning_rate": 1.433205292360222e-05,
1269
+ "loss": 0.5689,
1270
+ "step": 1680
1271
+ },
1272
+ {
1273
+ "epoch": 0.21630615640599002,
1274
+ "grad_norm": 0.6446322798728943,
1275
+ "learning_rate": 1.4417413572343152e-05,
1276
+ "loss": 0.5773,
1277
+ "step": 1690
1278
+ },
1279
+ {
1280
+ "epoch": 0.21758607449123257,
1281
+ "grad_norm": 1.505007266998291,
1282
+ "learning_rate": 1.4502774221084081e-05,
1283
+ "loss": 0.5677,
1284
+ "step": 1700
1285
+ },
1286
+ {
1287
+ "epoch": 0.2188659925764751,
1288
+ "grad_norm": 1.1939284801483154,
1289
+ "learning_rate": 1.4588134869825013e-05,
1290
+ "loss": 0.5805,
1291
+ "step": 1710
1292
+ },
1293
+ {
1294
+ "epoch": 0.22014591066171765,
1295
+ "grad_norm": 2.848264217376709,
1296
+ "learning_rate": 1.4673495518565943e-05,
1297
+ "loss": 0.5712,
1298
+ "step": 1720
1299
+ },
1300
+ {
1301
+ "epoch": 0.2214258287469602,
1302
+ "grad_norm": 1.2117749452590942,
1303
+ "learning_rate": 1.4758856167306873e-05,
1304
+ "loss": 0.582,
1305
+ "step": 1730
1306
+ },
1307
+ {
1308
+ "epoch": 0.22270574683220273,
1309
+ "grad_norm": 2.2297441959381104,
1310
+ "learning_rate": 1.4844216816047802e-05,
1311
+ "loss": 0.5751,
1312
+ "step": 1740
1313
+ },
1314
+ {
1315
+ "epoch": 0.2239856649174453,
1316
+ "grad_norm": 1.211978554725647,
1317
+ "learning_rate": 1.4929577464788734e-05,
1318
+ "loss": 0.5607,
1319
+ "step": 1750
1320
+ },
1321
+ {
1322
+ "epoch": 0.22526558300268784,
1323
+ "grad_norm": 0.8025451898574829,
1324
+ "learning_rate": 1.5014938113529664e-05,
1325
+ "loss": 0.5803,
1326
+ "step": 1760
1327
+ },
1328
+ {
1329
+ "epoch": 0.22654550108793037,
1330
+ "grad_norm": 1.2300900220870972,
1331
+ "learning_rate": 1.5100298762270595e-05,
1332
+ "loss": 0.5827,
1333
+ "step": 1770
1334
+ },
1335
+ {
1336
+ "epoch": 0.22782541917317292,
1337
+ "grad_norm": 1.0382864475250244,
1338
+ "learning_rate": 1.5185659411011523e-05,
1339
+ "loss": 0.5916,
1340
+ "step": 1780
1341
+ },
1342
+ {
1343
+ "epoch": 0.22910533725841545,
1344
+ "grad_norm": 1.4876656532287598,
1345
+ "learning_rate": 1.5271020059752453e-05,
1346
+ "loss": 0.5536,
1347
+ "step": 1790
1348
+ },
1349
+ {
1350
+ "epoch": 0.230385255343658,
1351
+ "grad_norm": 1.1409252882003784,
1352
+ "learning_rate": 1.5356380708493386e-05,
1353
+ "loss": 0.5484,
1354
+ "step": 1800
1355
+ },
1356
+ {
1357
+ "epoch": 0.23166517342890056,
1358
+ "grad_norm": 0.7180280089378357,
1359
+ "learning_rate": 1.5441741357234316e-05,
1360
+ "loss": 0.5712,
1361
+ "step": 1810
1362
+ },
1363
+ {
1364
+ "epoch": 0.23294509151414308,
1365
+ "grad_norm": 1.5890289545059204,
1366
+ "learning_rate": 1.5527102005975246e-05,
1367
+ "loss": 0.5758,
1368
+ "step": 1820
1369
+ },
1370
+ {
1371
+ "epoch": 0.23422500959938564,
1372
+ "grad_norm": 0.6635769605636597,
1373
+ "learning_rate": 1.5612462654716176e-05,
1374
+ "loss": 0.5543,
1375
+ "step": 1830
1376
+ },
1377
+ {
1378
+ "epoch": 0.2355049276846282,
1379
+ "grad_norm": 0.8864602446556091,
1380
+ "learning_rate": 1.5697823303457106e-05,
1381
+ "loss": 0.5536,
1382
+ "step": 1840
1383
+ },
1384
+ {
1385
+ "epoch": 0.23678484576987072,
1386
+ "grad_norm": 0.8372675776481628,
1387
+ "learning_rate": 1.578318395219804e-05,
1388
+ "loss": 0.5743,
1389
+ "step": 1850
1390
+ },
1391
+ {
1392
+ "epoch": 0.23806476385511327,
1393
+ "grad_norm": 0.7532811164855957,
1394
+ "learning_rate": 1.586854460093897e-05,
1395
+ "loss": 0.5593,
1396
+ "step": 1860
1397
+ },
1398
+ {
1399
+ "epoch": 0.23934468194035582,
1400
+ "grad_norm": 0.7231799960136414,
1401
+ "learning_rate": 1.59539052496799e-05,
1402
+ "loss": 0.5831,
1403
+ "step": 1870
1404
+ },
1405
+ {
1406
+ "epoch": 0.24062460002559835,
1407
+ "grad_norm": 1.124228596687317,
1408
+ "learning_rate": 1.603926589842083e-05,
1409
+ "loss": 0.5936,
1410
+ "step": 1880
1411
+ },
1412
+ {
1413
+ "epoch": 0.2419045181108409,
1414
+ "grad_norm": 1.0107067823410034,
1415
+ "learning_rate": 1.612462654716176e-05,
1416
+ "loss": 0.5598,
1417
+ "step": 1890
1418
+ },
1419
+ {
1420
+ "epoch": 0.24318443619608346,
1421
+ "grad_norm": 1.6117154359817505,
1422
+ "learning_rate": 1.620998719590269e-05,
1423
+ "loss": 0.5605,
1424
+ "step": 1900
1425
+ },
1426
+ {
1427
+ "epoch": 0.24446435428132599,
1428
+ "grad_norm": 1.5564496517181396,
1429
+ "learning_rate": 1.629534784464362e-05,
1430
+ "loss": 0.5666,
1431
+ "step": 1910
1432
+ },
1433
+ {
1434
+ "epoch": 0.24574427236656854,
1435
+ "grad_norm": 0.8275917768478394,
1436
+ "learning_rate": 1.638070849338455e-05,
1437
+ "loss": 0.5597,
1438
+ "step": 1920
1439
+ },
1440
+ {
1441
+ "epoch": 0.2470241904518111,
1442
+ "grad_norm": 0.8956324458122253,
1443
+ "learning_rate": 1.646606914212548e-05,
1444
+ "loss": 0.5453,
1445
+ "step": 1930
1446
+ },
1447
+ {
1448
+ "epoch": 0.24830410853705362,
1449
+ "grad_norm": 1.9644063711166382,
1450
+ "learning_rate": 1.655142979086641e-05,
1451
+ "loss": 0.5732,
1452
+ "step": 1940
1453
+ },
1454
+ {
1455
+ "epoch": 0.24958402662229617,
1456
+ "grad_norm": 0.7495090365409851,
1457
+ "learning_rate": 1.663679043960734e-05,
1458
+ "loss": 0.5549,
1459
+ "step": 1950
1460
+ },
1461
+ {
1462
+ "epoch": 0.2508639447075387,
1463
+ "grad_norm": 0.9346863627433777,
1464
+ "learning_rate": 1.6722151088348274e-05,
1465
+ "loss": 0.5803,
1466
+ "step": 1960
1467
+ },
1468
+ {
1469
+ "epoch": 0.2521438627927813,
1470
+ "grad_norm": 0.5847868919372559,
1471
+ "learning_rate": 1.6807511737089204e-05,
1472
+ "loss": 0.5705,
1473
+ "step": 1970
1474
+ },
1475
+ {
1476
+ "epoch": 0.2534237808780238,
1477
+ "grad_norm": 1.5892306566238403,
1478
+ "learning_rate": 1.6892872385830134e-05,
1479
+ "loss": 0.5371,
1480
+ "step": 1980
1481
+ },
1482
+ {
1483
+ "epoch": 0.25470369896326633,
1484
+ "grad_norm": 0.8782947063446045,
1485
+ "learning_rate": 1.6978233034571063e-05,
1486
+ "loss": 0.565,
1487
+ "step": 1990
1488
+ },
1489
+ {
1490
+ "epoch": 0.2559836170485089,
1491
+ "grad_norm": 1.0143548250198364,
1492
+ "learning_rate": 1.7063593683311993e-05,
1493
+ "loss": 0.557,
1494
+ "step": 2000
1495
+ },
1496
+ {
1497
+ "epoch": 0.2559836170485089,
1498
+ "eval_accuracy": 0.74427,
1499
+ "eval_loss": 0.5308485627174377,
1500
+ "eval_macro_f1": 0.743516265066894,
1501
+ "eval_precision": 0.7538937627554241,
1502
+ "eval_recall": 0.7483595020974525,
1503
+ "eval_runtime": 708.2272,
1504
+ "eval_samples_per_second": 141.198,
1505
+ "eval_steps_per_second": 1.104,
1506
+ "step": 2000
1507
+ },
1508
+ {
1509
+ "epoch": 0.25726353513375144,
1510
+ "grad_norm": 0.9080437421798706,
1511
+ "learning_rate": 1.7148954332052926e-05,
1512
+ "loss": 0.5478,
1513
+ "step": 2010
1514
+ },
1515
+ {
1516
+ "epoch": 0.25854345321899397,
1517
+ "grad_norm": 1.1889078617095947,
1518
+ "learning_rate": 1.7234314980793856e-05,
1519
+ "loss": 0.5394,
1520
+ "step": 2020
1521
+ },
1522
+ {
1523
+ "epoch": 0.25982337130423655,
1524
+ "grad_norm": 1.7290714979171753,
1525
+ "learning_rate": 1.7319675629534786e-05,
1526
+ "loss": 0.5385,
1527
+ "step": 2030
1528
+ },
1529
+ {
1530
+ "epoch": 0.2611032893894791,
1531
+ "grad_norm": 0.9356957077980042,
1532
+ "learning_rate": 1.7405036278275716e-05,
1533
+ "loss": 0.5392,
1534
+ "step": 2040
1535
+ },
1536
+ {
1537
+ "epoch": 0.2623832074747216,
1538
+ "grad_norm": 0.765798032283783,
1539
+ "learning_rate": 1.7490396927016646e-05,
1540
+ "loss": 0.5442,
1541
+ "step": 2050
1542
+ },
1543
+ {
1544
+ "epoch": 0.2636631255599642,
1545
+ "grad_norm": 1.6606382131576538,
1546
+ "learning_rate": 1.7575757575757576e-05,
1547
+ "loss": 0.5331,
1548
+ "step": 2060
1549
+ },
1550
+ {
1551
+ "epoch": 0.2649430436452067,
1552
+ "grad_norm": 0.7928575277328491,
1553
+ "learning_rate": 1.766111822449851e-05,
1554
+ "loss": 0.5483,
1555
+ "step": 2070
1556
+ },
1557
+ {
1558
+ "epoch": 0.26622296173044924,
1559
+ "grad_norm": 2.2261486053466797,
1560
+ "learning_rate": 1.774647887323944e-05,
1561
+ "loss": 0.5536,
1562
+ "step": 2080
1563
+ },
1564
+ {
1565
+ "epoch": 0.2675028798156918,
1566
+ "grad_norm": 0.9795323610305786,
1567
+ "learning_rate": 1.783183952198037e-05,
1568
+ "loss": 0.5584,
1569
+ "step": 2090
1570
+ },
1571
+ {
1572
+ "epoch": 0.26878279790093434,
1573
+ "grad_norm": 2.373581886291504,
1574
+ "learning_rate": 1.7917200170721298e-05,
1575
+ "loss": 0.5475,
1576
+ "step": 2100
1577
+ },
1578
+ {
1579
+ "epoch": 0.27006271598617687,
1580
+ "grad_norm": 0.9658742547035217,
1581
+ "learning_rate": 1.8002560819462228e-05,
1582
+ "loss": 0.5577,
1583
+ "step": 2110
1584
+ },
1585
+ {
1586
+ "epoch": 0.27134263407141945,
1587
+ "grad_norm": 0.8572520017623901,
1588
+ "learning_rate": 1.808792146820316e-05,
1589
+ "loss": 0.5301,
1590
+ "step": 2120
1591
+ },
1592
+ {
1593
+ "epoch": 0.272622552156662,
1594
+ "grad_norm": 0.7225477695465088,
1595
+ "learning_rate": 1.817328211694409e-05,
1596
+ "loss": 0.5626,
1597
+ "step": 2130
1598
+ },
1599
+ {
1600
+ "epoch": 0.2739024702419045,
1601
+ "grad_norm": 1.4276026487350464,
1602
+ "learning_rate": 1.825864276568502e-05,
1603
+ "loss": 0.5208,
1604
+ "step": 2140
1605
+ },
1606
+ {
1607
+ "epoch": 0.2751823883271471,
1608
+ "grad_norm": 1.7469464540481567,
1609
+ "learning_rate": 1.834400341442595e-05,
1610
+ "loss": 0.5669,
1611
+ "step": 2150
1612
+ },
1613
+ {
1614
+ "epoch": 0.2764623064123896,
1615
+ "grad_norm": 1.250931978225708,
1616
+ "learning_rate": 1.842936406316688e-05,
1617
+ "loss": 0.5577,
1618
+ "step": 2160
1619
+ },
1620
+ {
1621
+ "epoch": 0.27774222449763214,
1622
+ "grad_norm": 1.011313557624817,
1623
+ "learning_rate": 1.8514724711907814e-05,
1624
+ "loss": 0.5494,
1625
+ "step": 2170
1626
+ },
1627
+ {
1628
+ "epoch": 0.2790221425828747,
1629
+ "grad_norm": 0.7606518268585205,
1630
+ "learning_rate": 1.8600085360648744e-05,
1631
+ "loss": 0.5391,
1632
+ "step": 2180
1633
+ },
1634
+ {
1635
+ "epoch": 0.28030206066811725,
1636
+ "grad_norm": 0.8173381090164185,
1637
+ "learning_rate": 1.8685446009389673e-05,
1638
+ "loss": 0.5262,
1639
+ "step": 2190
1640
+ },
1641
+ {
1642
+ "epoch": 0.2815819787533598,
1643
+ "grad_norm": 0.7178583145141602,
1644
+ "learning_rate": 1.8770806658130603e-05,
1645
+ "loss": 0.5182,
1646
+ "step": 2200
1647
+ },
1648
+ {
1649
+ "epoch": 0.28286189683860236,
1650
+ "grad_norm": 0.6852149367332458,
1651
+ "learning_rate": 1.8856167306871533e-05,
1652
+ "loss": 0.528,
1653
+ "step": 2210
1654
+ },
1655
+ {
1656
+ "epoch": 0.2841418149238449,
1657
+ "grad_norm": 1.9807002544403076,
1658
+ "learning_rate": 1.8941527955612463e-05,
1659
+ "loss": 0.5315,
1660
+ "step": 2220
1661
+ },
1662
+ {
1663
+ "epoch": 0.2854217330090874,
1664
+ "grad_norm": 0.7938176989555359,
1665
+ "learning_rate": 1.9026888604353396e-05,
1666
+ "loss": 0.5324,
1667
+ "step": 2230
1668
+ },
1669
+ {
1670
+ "epoch": 0.28670165109433,
1671
+ "grad_norm": 0.8181535005569458,
1672
+ "learning_rate": 1.9112249253094326e-05,
1673
+ "loss": 0.51,
1674
+ "step": 2240
1675
+ },
1676
+ {
1677
+ "epoch": 0.2879815691795725,
1678
+ "grad_norm": 0.8722071051597595,
1679
+ "learning_rate": 1.9197609901835256e-05,
1680
+ "loss": 0.5373,
1681
+ "step": 2250
1682
+ },
1683
+ {
1684
+ "epoch": 0.28926148726481504,
1685
+ "grad_norm": 1.371565818786621,
1686
+ "learning_rate": 1.9282970550576186e-05,
1687
+ "loss": 0.5227,
1688
+ "step": 2260
1689
+ },
1690
+ {
1691
+ "epoch": 0.29054140535005757,
1692
+ "grad_norm": 0.6769284605979919,
1693
+ "learning_rate": 1.9368331199317115e-05,
1694
+ "loss": 0.5421,
1695
+ "step": 2270
1696
+ },
1697
+ {
1698
+ "epoch": 0.29182132343530015,
1699
+ "grad_norm": 2.1016697883605957,
1700
+ "learning_rate": 1.945369184805805e-05,
1701
+ "loss": 0.5299,
1702
+ "step": 2280
1703
+ },
1704
+ {
1705
+ "epoch": 0.2931012415205427,
1706
+ "grad_norm": 1.7011477947235107,
1707
+ "learning_rate": 1.953905249679898e-05,
1708
+ "loss": 0.5403,
1709
+ "step": 2290
1710
+ },
1711
+ {
1712
+ "epoch": 0.2943811596057852,
1713
+ "grad_norm": 1.0961430072784424,
1714
+ "learning_rate": 1.962441314553991e-05,
1715
+ "loss": 0.534,
1716
+ "step": 2300
1717
+ },
1718
+ {
1719
+ "epoch": 0.2956610776910278,
1720
+ "grad_norm": 0.8536306023597717,
1721
+ "learning_rate": 1.9709773794280838e-05,
1722
+ "loss": 0.5562,
1723
+ "step": 2310
1724
+ },
1725
+ {
1726
+ "epoch": 0.2969409957762703,
1727
+ "grad_norm": 1.046201467514038,
1728
+ "learning_rate": 1.9795134443021768e-05,
1729
+ "loss": 0.5418,
1730
+ "step": 2320
1731
+ },
1732
+ {
1733
+ "epoch": 0.29822091386151284,
1734
+ "grad_norm": 1.38905930519104,
1735
+ "learning_rate": 1.9880495091762698e-05,
1736
+ "loss": 0.5557,
1737
+ "step": 2330
1738
+ },
1739
+ {
1740
+ "epoch": 0.2995008319467554,
1741
+ "grad_norm": 2.44215989112854,
1742
+ "learning_rate": 1.996585574050363e-05,
1743
+ "loss": 0.5285,
1744
+ "step": 2340
1745
+ },
1746
+ {
1747
+ "epoch": 0.30078075003199795,
1748
+ "grad_norm": 1.4736480712890625,
1749
+ "learning_rate": 1.9994311717861208e-05,
1750
+ "loss": 0.513,
1751
+ "step": 2350
1752
+ },
1753
+ {
1754
+ "epoch": 0.30206066811724047,
1755
+ "grad_norm": 0.7968178987503052,
1756
+ "learning_rate": 1.9984831247629883e-05,
1757
+ "loss": 0.5249,
1758
+ "step": 2360
1759
+ },
1760
+ {
1761
+ "epoch": 0.30334058620248305,
1762
+ "grad_norm": 1.6530416011810303,
1763
+ "learning_rate": 1.997535077739856e-05,
1764
+ "loss": 0.5329,
1765
+ "step": 2370
1766
+ },
1767
+ {
1768
+ "epoch": 0.3046205042877256,
1769
+ "grad_norm": 1.0949759483337402,
1770
+ "learning_rate": 1.9965870307167238e-05,
1771
+ "loss": 0.5164,
1772
+ "step": 2380
1773
+ },
1774
+ {
1775
+ "epoch": 0.3059004223729681,
1776
+ "grad_norm": 1.2724019289016724,
1777
+ "learning_rate": 1.9956389836935916e-05,
1778
+ "loss": 0.5293,
1779
+ "step": 2390
1780
+ },
1781
+ {
1782
+ "epoch": 0.3071803404582107,
1783
+ "grad_norm": 1.8785524368286133,
1784
+ "learning_rate": 1.994690936670459e-05,
1785
+ "loss": 0.5286,
1786
+ "step": 2400
1787
+ }
1788
+ ],
1789
+ "logging_steps": 10,
1790
+ "max_steps": 23439,
1791
+ "num_input_tokens_seen": 0,
1792
+ "num_train_epochs": 3,
1793
+ "save_steps": 200,
1794
+ "stateful_callbacks": {
1795
+ "EarlyStoppingCallback": {
1796
+ "args": {
1797
+ "early_stopping_patience": 3,
1798
+ "early_stopping_threshold": 0.0
1799
+ },
1800
+ "attributes": {
1801
+ "early_stopping_patience_counter": 0
1802
+ }
1803
+ },
1804
+ "TrainerControl": {
1805
+ "args": {
1806
+ "should_epoch_stop": false,
1807
+ "should_evaluate": false,
1808
+ "should_log": false,
1809
+ "should_save": true,
1810
+ "should_training_stop": false
1811
+ },
1812
+ "attributes": {}
1813
+ }
1814
+ },
1815
+ "total_flos": 4.0413858103296e+16,
1816
+ "train_batch_size": 64,
1817
+ "trial_name": null,
1818
+ "trial_params": null
1819
+ }
codebert-base-cross-entropy/checkpoint-2400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66cbef846d2329e000727fbcf0db86e3ad5252ae9de8caa12e196b1fac3c9577
3
+ size 5841
codebert-base-cross-entropy/checkpoint-2400/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
codebert-base-cross-entropy/training.log ADDED
@@ -0,0 +1,324 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
2
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
3
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
4
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
5
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
6
+ RobertaForSequenceClassification(
7
+ (roberta): RobertaModel(
8
+ (embeddings): RobertaEmbeddings(
9
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
10
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
11
+ (token_type_embeddings): Embedding(1, 768)
12
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
13
+ (dropout): Dropout(p=0.1, inplace=False)
14
+ )
15
+ (encoder): RobertaEncoder(
16
+ (layer): ModuleList(
17
+ (0-11): 12 x RobertaLayer(
18
+ (attention): RobertaAttention(
19
+ (self): RobertaSdpaSelfAttention(
20
+ (query): Linear(in_features=768, out_features=768, bias=True)
21
+ (key): Linear(in_features=768, out_features=768, bias=True)
22
+ (value): Linear(in_features=768, out_features=768, bias=True)
23
+ (dropout): Dropout(p=0.1, inplace=False)
24
+ )
25
+ (output): RobertaSelfOutput(
26
+ (dense): Linear(in_features=768, out_features=768, bias=True)
27
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
28
+ (dropout): Dropout(p=0.1, inplace=False)
29
+ )
30
+ )
31
+ (intermediate): RobertaIntermediate(
32
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
33
+ (intermediate_act_fn): GELUActivation()
34
+ )
35
+ (output): RobertaOutput(
36
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
37
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
38
+ (dropout): Dropout(p=0.1, inplace=False)
39
+ )
40
+ )
41
+ )
42
+ )
43
+ )
44
+ (classifier): RobertaClassificationHead(
45
+ (dense): Linear(in_features=768, out_features=768, bias=True)
46
+ (dropout): Dropout(p=0.1, inplace=False)
47
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
48
+ )
49
+ )
50
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
51
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
52
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
53
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
54
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===
55
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
56
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
57
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
58
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
59
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
60
+ RobertaForSequenceClassification(
61
+ (roberta): RobertaModel(
62
+ (embeddings): RobertaEmbeddings(
63
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
64
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
65
+ (token_type_embeddings): Embedding(1, 768)
66
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
67
+ (dropout): Dropout(p=0.1, inplace=False)
68
+ )
69
+ (encoder): RobertaEncoder(
70
+ (layer): ModuleList(
71
+ (0-11): 12 x RobertaLayer(
72
+ (attention): RobertaAttention(
73
+ (self): RobertaSdpaSelfAttention(
74
+ (query): Linear(in_features=768, out_features=768, bias=True)
75
+ (key): Linear(in_features=768, out_features=768, bias=True)
76
+ (value): Linear(in_features=768, out_features=768, bias=True)
77
+ (dropout): Dropout(p=0.1, inplace=False)
78
+ )
79
+ (output): RobertaSelfOutput(
80
+ (dense): Linear(in_features=768, out_features=768, bias=True)
81
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
82
+ (dropout): Dropout(p=0.1, inplace=False)
83
+ )
84
+ )
85
+ (intermediate): RobertaIntermediate(
86
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
87
+ (intermediate_act_fn): GELUActivation()
88
+ )
89
+ (output): RobertaOutput(
90
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
91
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
92
+ (dropout): Dropout(p=0.1, inplace=False)
93
+ )
94
+ )
95
+ )
96
+ )
97
+ )
98
+ (classifier): RobertaClassificationHead(
99
+ (dense): Linear(in_features=768, out_features=768, bias=True)
100
+ (dropout): Dropout(p=0.1, inplace=False)
101
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
102
+ )
103
+ )
104
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
105
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
106
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
107
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
108
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===
109
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
110
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
111
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
112
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
113
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
114
+ RobertaForSequenceClassification(
115
+ (roberta): RobertaModel(
116
+ (embeddings): RobertaEmbeddings(
117
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
118
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
119
+ (token_type_embeddings): Embedding(1, 768)
120
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
121
+ (dropout): Dropout(p=0.1, inplace=False)
122
+ )
123
+ (encoder): RobertaEncoder(
124
+ (layer): ModuleList(
125
+ (0-11): 12 x RobertaLayer(
126
+ (attention): RobertaAttention(
127
+ (self): RobertaSdpaSelfAttention(
128
+ (query): Linear(in_features=768, out_features=768, bias=True)
129
+ (key): Linear(in_features=768, out_features=768, bias=True)
130
+ (value): Linear(in_features=768, out_features=768, bias=True)
131
+ (dropout): Dropout(p=0.1, inplace=False)
132
+ )
133
+ (output): RobertaSelfOutput(
134
+ (dense): Linear(in_features=768, out_features=768, bias=True)
135
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
136
+ (dropout): Dropout(p=0.1, inplace=False)
137
+ )
138
+ )
139
+ (intermediate): RobertaIntermediate(
140
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
141
+ (intermediate_act_fn): GELUActivation()
142
+ )
143
+ (output): RobertaOutput(
144
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
145
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
146
+ (dropout): Dropout(p=0.1, inplace=False)
147
+ )
148
+ )
149
+ )
150
+ )
151
+ )
152
+ (classifier): RobertaClassificationHead(
153
+ (dense): Linear(in_features=768, out_features=768, bias=True)
154
+ (dropout): Dropout(p=0.1, inplace=False)
155
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
156
+ )
157
+ )
158
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
159
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
160
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
161
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
162
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===
163
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
164
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
165
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
166
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
167
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
168
+ RobertaForSequenceClassification(
169
+ (roberta): RobertaModel(
170
+ (embeddings): RobertaEmbeddings(
171
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
172
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
173
+ (token_type_embeddings): Embedding(1, 768)
174
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
175
+ (dropout): Dropout(p=0.1, inplace=False)
176
+ )
177
+ (encoder): RobertaEncoder(
178
+ (layer): ModuleList(
179
+ (0-11): 12 x RobertaLayer(
180
+ (attention): RobertaAttention(
181
+ (self): RobertaSdpaSelfAttention(
182
+ (query): Linear(in_features=768, out_features=768, bias=True)
183
+ (key): Linear(in_features=768, out_features=768, bias=True)
184
+ (value): Linear(in_features=768, out_features=768, bias=True)
185
+ (dropout): Dropout(p=0.1, inplace=False)
186
+ )
187
+ (output): RobertaSelfOutput(
188
+ (dense): Linear(in_features=768, out_features=768, bias=True)
189
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
190
+ (dropout): Dropout(p=0.1, inplace=False)
191
+ )
192
+ )
193
+ (intermediate): RobertaIntermediate(
194
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
195
+ (intermediate_act_fn): GELUActivation()
196
+ )
197
+ (output): RobertaOutput(
198
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
199
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
200
+ (dropout): Dropout(p=0.1, inplace=False)
201
+ )
202
+ )
203
+ )
204
+ )
205
+ )
206
+ (classifier): RobertaClassificationHead(
207
+ (dense): Linear(in_features=768, out_features=768, bias=True)
208
+ (dropout): Dropout(p=0.1, inplace=False)
209
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
210
+ )
211
+ )
212
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
213
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
214
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
215
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
216
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===
217
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
218
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
219
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
220
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
221
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
222
+ RobertaForSequenceClassification(
223
+ (roberta): RobertaModel(
224
+ (embeddings): RobertaEmbeddings(
225
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
226
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
227
+ (token_type_embeddings): Embedding(1, 768)
228
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
229
+ (dropout): Dropout(p=0.1, inplace=False)
230
+ )
231
+ (encoder): RobertaEncoder(
232
+ (layer): ModuleList(
233
+ (0-11): 12 x RobertaLayer(
234
+ (attention): RobertaAttention(
235
+ (self): RobertaSdpaSelfAttention(
236
+ (query): Linear(in_features=768, out_features=768, bias=True)
237
+ (key): Linear(in_features=768, out_features=768, bias=True)
238
+ (value): Linear(in_features=768, out_features=768, bias=True)
239
+ (dropout): Dropout(p=0.1, inplace=False)
240
+ )
241
+ (output): RobertaSelfOutput(
242
+ (dense): Linear(in_features=768, out_features=768, bias=True)
243
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
244
+ (dropout): Dropout(p=0.1, inplace=False)
245
+ )
246
+ )
247
+ (intermediate): RobertaIntermediate(
248
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
249
+ (intermediate_act_fn): GELUActivation()
250
+ )
251
+ (output): RobertaOutput(
252
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
253
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
254
+ (dropout): Dropout(p=0.1, inplace=False)
255
+ )
256
+ )
257
+ )
258
+ )
259
+ )
260
+ (classifier): RobertaClassificationHead(
261
+ (dense): Linear(in_features=768, out_features=768, bias=True)
262
+ (dropout): Dropout(p=0.1, inplace=False)
263
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
264
+ )
265
+ )
266
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
267
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
268
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
269
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
270
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===
271
+ 2026-04-15 11:40:23,361 - INFO - train_pipeline - Logging to ./output_checkpoints/codebert-base-cross-entropy/training.log
272
+ 2026-04-15 11:40:23,369 - INFO - train_pipeline - Loading model & tokenizer for 'microsoft/codebert-base'
273
+ 2026-04-15 11:40:25,679 - INFO - train_pipeline - Model placed on cuda
274
+ 2026-04-15 11:40:25,694 - INFO - train_pipeline - ===== Model Architecture =====
275
+ 2026-04-15 11:40:25,708 - INFO - train_pipeline -
276
+ RobertaForSequenceClassification(
277
+ (roberta): RobertaModel(
278
+ (embeddings): RobertaEmbeddings(
279
+ (word_embeddings): Embedding(50265, 768, padding_idx=1)
280
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
281
+ (token_type_embeddings): Embedding(1, 768)
282
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
283
+ (dropout): Dropout(p=0.1, inplace=False)
284
+ )
285
+ (encoder): RobertaEncoder(
286
+ (layer): ModuleList(
287
+ (0-11): 12 x RobertaLayer(
288
+ (attention): RobertaAttention(
289
+ (self): RobertaSdpaSelfAttention(
290
+ (query): Linear(in_features=768, out_features=768, bias=True)
291
+ (key): Linear(in_features=768, out_features=768, bias=True)
292
+ (value): Linear(in_features=768, out_features=768, bias=True)
293
+ (dropout): Dropout(p=0.1, inplace=False)
294
+ )
295
+ (output): RobertaSelfOutput(
296
+ (dense): Linear(in_features=768, out_features=768, bias=True)
297
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
298
+ (dropout): Dropout(p=0.1, inplace=False)
299
+ )
300
+ )
301
+ (intermediate): RobertaIntermediate(
302
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
303
+ (intermediate_act_fn): GELUActivation()
304
+ )
305
+ (output): RobertaOutput(
306
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
307
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
308
+ (dropout): Dropout(p=0.1, inplace=False)
309
+ )
310
+ )
311
+ )
312
+ )
313
+ )
314
+ (classifier): RobertaClassificationHead(
315
+ (dense): Linear(in_features=768, out_features=768, bias=True)
316
+ (dropout): Dropout(p=0.1, inplace=False)
317
+ (out_proj): Linear(in_features=768, out_features=2, bias=True)
318
+ )
319
+ )
320
+ 2026-04-15 11:40:25,734 - INFO - train_pipeline - ===== Tokenizer Summary =====
321
+ 2026-04-15 11:40:25,811 - INFO - train_pipeline - Vocab size: 50265 | Special tokens: ['<s>', '</s>', '<unk>', '<pad>', '<mask>']
322
+ 2026-04-15 11:40:25,830 - INFO - train_pipeline - ===== End of Architecture Log =====
323
+ 2026-04-15 11:40:25,852 - INFO - train_pipeline - Base model weights frozen – only classifier head will be trained.
324
+ 2026-04-15 11:40:27,878 - INFO - train_pipeline - === Starting training ===