AhmedSSoliman commited on
Commit
bde5afc
·
verified ·
1 Parent(s): e0c231d

Model save

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +71 -0
  2. config.json +47 -0
  3. model.safetensors +3 -0
  4. training_args.bin +3 -0
  5. trial-0/checkpoint-3011/config.json +47 -0
  6. trial-0/checkpoint-3011/model.safetensors +3 -0
  7. trial-0/checkpoint-3011/optimizer.pt +3 -0
  8. trial-0/checkpoint-3011/rng_state.pth +3 -0
  9. trial-0/checkpoint-3011/scheduler.pt +3 -0
  10. trial-0/checkpoint-3011/trainer_state.json +465 -0
  11. trial-0/checkpoint-3011/training_args.bin +3 -0
  12. trial-1/checkpoint-15055/config.json +47 -0
  13. trial-1/checkpoint-15055/model.safetensors +3 -0
  14. trial-1/checkpoint-15055/optimizer.pt +3 -0
  15. trial-1/checkpoint-15055/rng_state.pth +3 -0
  16. trial-1/checkpoint-15055/scheduler.pt +3 -0
  17. trial-1/checkpoint-15055/trainer_state.json +2200 -0
  18. trial-1/checkpoint-15055/training_args.bin +3 -0
  19. trial-2/checkpoint-9033/config.json +47 -0
  20. trial-2/checkpoint-9033/model.safetensors +3 -0
  21. trial-2/checkpoint-9033/optimizer.pt +3 -0
  22. trial-2/checkpoint-9033/rng_state.pth +3 -0
  23. trial-2/checkpoint-9033/scheduler.pt +3 -0
  24. trial-2/checkpoint-9033/trainer_state.json +1329 -0
  25. trial-2/checkpoint-9033/training_args.bin +3 -0
  26. trial-3/checkpoint-1506/config.json +47 -0
  27. trial-3/checkpoint-1506/model.safetensors +3 -0
  28. trial-3/checkpoint-1506/optimizer.pt +3 -0
  29. trial-3/checkpoint-1506/rng_state.pth +3 -0
  30. trial-3/checkpoint-1506/scheduler.pt +3 -0
  31. trial-3/checkpoint-1506/trainer_state.json +255 -0
  32. trial-3/checkpoint-1506/training_args.bin +3 -0
  33. trial-4/checkpoint-3012/config.json +47 -0
  34. trial-4/checkpoint-3012/model.safetensors +3 -0
  35. trial-4/checkpoint-3012/optimizer.pt +3 -0
  36. trial-4/checkpoint-3012/rng_state.pth +3 -0
  37. trial-4/checkpoint-3012/scheduler.pt +3 -0
  38. trial-4/checkpoint-3012/trainer_state.json +477 -0
  39. trial-4/checkpoint-3012/training_args.bin +3 -0
  40. trial-5/checkpoint-1506/config.json +47 -0
  41. trial-5/checkpoint-1506/model.safetensors +3 -0
  42. trial-5/checkpoint-1506/optimizer.pt +3 -0
  43. trial-5/checkpoint-1506/rng_state.pth +3 -0
  44. trial-5/checkpoint-1506/scheduler.pt +3 -0
  45. trial-5/checkpoint-1506/trainer_state.json +255 -0
  46. trial-5/checkpoint-1506/training_args.bin +3 -0
  47. trial-6/checkpoint-1506/config.json +47 -0
  48. trial-6/checkpoint-1506/model.safetensors +3 -0
  49. trial-6/checkpoint-1506/optimizer.pt +3 -0
  50. trial-6/checkpoint-1506/rng_state.pth +3 -0
README.md ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: answerdotai/ModernBERT-large
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ - precision
10
+ - recall
11
+ - f1
12
+ model-index:
13
+ - name: answerdotai-ModernBERT-large-finetuned
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ # answerdotai-ModernBERT-large-finetuned
21
+
22
+ This model is a fine-tuned version of [answerdotai/ModernBERT-large](https://huggingface.co/answerdotai/ModernBERT-large) on the None dataset.
23
+ It achieves the following results on the evaluation set:
24
+ - Loss: 0.0153
25
+ - Accuracy: 0.9980
26
+ - Precision: 0.9980
27
+ - Recall: 0.9980
28
+ - F1: 0.9980
29
+
30
+ ## Model description
31
+
32
+ More information needed
33
+
34
+ ## Intended uses & limitations
35
+
36
+ More information needed
37
+
38
+ ## Training and evaluation data
39
+
40
+ More information needed
41
+
42
+ ## Training procedure
43
+
44
+ ### Training hyperparameters
45
+
46
+ The following hyperparameters were used during training:
47
+ - learning_rate: 4.1905207188250686e-05
48
+ - train_batch_size: 16
49
+ - eval_batch_size: 16
50
+ - seed: 42
51
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
52
+ - lr_scheduler_type: linear
53
+ - num_epochs: 5
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
58
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|
59
+ | 0.0046 | 1.0 | 3011 | 0.0257 | 0.9962 | 0.9962 | 0.9962 | 0.9962 |
60
+ | 0.021 | 2.0 | 6022 | 0.0234 | 0.9959 | 0.9960 | 0.9959 | 0.9960 |
61
+ | 0.0001 | 3.0 | 9033 | 0.0194 | 0.9979 | 0.9978 | 0.9979 | 0.9978 |
62
+ | 0.0002 | 4.0 | 12044 | 0.0181 | 0.9979 | 0.9978 | 0.9979 | 0.9978 |
63
+ | 0.0 | 5.0 | 15055 | 0.0177 | 0.9980 | 0.9980 | 0.9980 | 0.9980 |
64
+
65
+
66
+ ### Framework versions
67
+
68
+ - Transformers 4.48.0.dev0
69
+ - Pytorch 2.5.1+cu124
70
+ - Datasets 3.2.0
71
+ - Tokenizers 0.21.0
config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e73302dfe575e2fae3ec24034f72088b2adb6e6f5c4e4a35e949d0d8c4b78dbb
3
+ size 1583351632
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d77dcef85d4aa9fa4696da036cb727ea773883db77c9535ae75875fc4a5a5c11
3
+ size 5432
trial-0/checkpoint-3011/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-0/checkpoint-3011/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e79c24376368ad25cd3c338648c1e61fcbf3c81c96c917991d54a7bad08de0de
3
+ size 1583351632
trial-0/checkpoint-3011/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:022244260a3503756427f87fda63355cfa09334e6a2699f150afdd10b1cc0cb3
3
+ size 3166813178
trial-0/checkpoint-3011/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568428d80a25211a390c359ca51b0b20b38ca0607fbc196f106c9841c02d3e59
3
+ size 14244
trial-0/checkpoint-3011/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f8a6aef7194053a2f13849a184537c69f114fdea257f4aa4eef32c5d0218928
3
+ size 1064
trial-0/checkpoint-3011/trainer_state.json ADDED
@@ -0,0 +1,465 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.02747241035103798,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-0/checkpoint-3011",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 3011,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016605778811026237,
13
+ "grad_norm": 8.580768585205078,
14
+ "learning_rate": 5.396357633246935e-06,
15
+ "loss": 0.4089,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.033211557622052475,
20
+ "grad_norm": 4.984262466430664,
21
+ "learning_rate": 5.385132993856705e-06,
22
+ "loss": 0.1911,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.04981733643307871,
27
+ "grad_norm": 145.00692749023438,
28
+ "learning_rate": 5.373908354466475e-06,
29
+ "loss": 0.1545,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.06642311524410495,
34
+ "grad_norm": 2.051618814468384,
35
+ "learning_rate": 5.3626837150762455e-06,
36
+ "loss": 0.0941,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.08302889405513118,
41
+ "grad_norm": 0.19901786744594574,
42
+ "learning_rate": 5.351459075686016e-06,
43
+ "loss": 0.0327,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.09963467286615742,
48
+ "grad_norm": 0.003997017629444599,
49
+ "learning_rate": 5.340234436295786e-06,
50
+ "loss": 0.0585,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.11624045167718366,
55
+ "grad_norm": 0.016547370702028275,
56
+ "learning_rate": 5.329009796905555e-06,
57
+ "loss": 0.0461,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.1328462304882099,
62
+ "grad_norm": 5.060605049133301,
63
+ "learning_rate": 5.317785157515326e-06,
64
+ "loss": 0.0634,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.14945200929923613,
69
+ "grad_norm": 0.013804232701659203,
70
+ "learning_rate": 5.306560518125096e-06,
71
+ "loss": 0.0538,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.16605778811026237,
76
+ "grad_norm": 0.056022025644779205,
77
+ "learning_rate": 5.295335878734865e-06,
78
+ "loss": 0.0283,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.1826635669212886,
83
+ "grad_norm": 0.0030180325265973806,
84
+ "learning_rate": 5.284111239344636e-06,
85
+ "loss": 0.0278,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.19926934573231483,
90
+ "grad_norm": 0.14190103113651276,
91
+ "learning_rate": 5.272886599954406e-06,
92
+ "loss": 0.0486,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.2158751245433411,
97
+ "grad_norm": 23.922273635864258,
98
+ "learning_rate": 5.261661960564176e-06,
99
+ "loss": 0.0757,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.23248090335436733,
104
+ "grad_norm": 0.017755718901753426,
105
+ "learning_rate": 5.250437321173947e-06,
106
+ "loss": 0.0503,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.24908668216539356,
111
+ "grad_norm": 0.1306287795305252,
112
+ "learning_rate": 5.239212681783716e-06,
113
+ "loss": 0.0369,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.2656924609764198,
118
+ "grad_norm": 0.0014153249794617295,
119
+ "learning_rate": 5.227988042393486e-06,
120
+ "loss": 0.0284,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.282298239787446,
125
+ "grad_norm": 13.222740173339844,
126
+ "learning_rate": 5.216763403003257e-06,
127
+ "loss": 0.0685,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.29890401859847227,
132
+ "grad_norm": 0.01663159392774105,
133
+ "learning_rate": 5.205538763613026e-06,
134
+ "loss": 0.0377,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.3155097974094985,
139
+ "grad_norm": 9.587738037109375,
140
+ "learning_rate": 5.194314124222796e-06,
141
+ "loss": 0.0284,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.33211557622052473,
146
+ "grad_norm": 0.0031755813397467136,
147
+ "learning_rate": 5.183089484832566e-06,
148
+ "loss": 0.0287,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.348721355031551,
153
+ "grad_norm": 0.002594751538708806,
154
+ "learning_rate": 5.1718648454423365e-06,
155
+ "loss": 0.0239,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.3653271338425772,
160
+ "grad_norm": 0.0008383135427720845,
161
+ "learning_rate": 5.160640206052107e-06,
162
+ "loss": 0.0163,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.38193291265360346,
167
+ "grad_norm": 0.0018693436868488789,
168
+ "learning_rate": 5.149415566661877e-06,
169
+ "loss": 0.0173,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.39853869146462967,
174
+ "grad_norm": 3.5105800634482875e-05,
175
+ "learning_rate": 5.138190927271647e-06,
176
+ "loss": 0.0144,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.41514447027565593,
181
+ "grad_norm": 0.04262986406683922,
182
+ "learning_rate": 5.126966287881418e-06,
183
+ "loss": 0.0225,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.4317502490866822,
188
+ "grad_norm": 0.257622092962265,
189
+ "learning_rate": 5.115741648491187e-06,
190
+ "loss": 0.0458,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.4483560278977084,
195
+ "grad_norm": 0.057092875242233276,
196
+ "learning_rate": 5.104517009100957e-06,
197
+ "loss": 0.015,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.46496180670873466,
202
+ "grad_norm": 0.009775767102837563,
203
+ "learning_rate": 5.093292369710727e-06,
204
+ "loss": 0.0002,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.48156758551976087,
209
+ "grad_norm": 0.16755907237529755,
210
+ "learning_rate": 5.082067730320497e-06,
211
+ "loss": 0.0119,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.4981733643307871,
216
+ "grad_norm": 0.0004505734541453421,
217
+ "learning_rate": 5.0708430909302675e-06,
218
+ "loss": 0.0107,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.5147791431418134,
223
+ "grad_norm": 0.00015653851733077317,
224
+ "learning_rate": 5.059618451540038e-06,
225
+ "loss": 0.0002,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.5313849219528396,
230
+ "grad_norm": 0.1482504904270172,
231
+ "learning_rate": 5.048393812149808e-06,
232
+ "loss": 0.0221,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.5479907007638658,
237
+ "grad_norm": 0.00016298597620334476,
238
+ "learning_rate": 5.037169172759578e-06,
239
+ "loss": 0.0281,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.564596479574892,
244
+ "grad_norm": 0.0018974934937432408,
245
+ "learning_rate": 5.025944533369348e-06,
246
+ "loss": 0.0152,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.5812022583859183,
251
+ "grad_norm": 0.020836347714066505,
252
+ "learning_rate": 5.014719893979118e-06,
253
+ "loss": 0.0298,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.5978080371969445,
258
+ "grad_norm": 0.0017976267263293266,
259
+ "learning_rate": 5.003495254588888e-06,
260
+ "loss": 0.0163,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.6144138160079707,
265
+ "grad_norm": 0.007485950365662575,
266
+ "learning_rate": 4.992270615198658e-06,
267
+ "loss": 0.0196,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.631019594818997,
272
+ "grad_norm": 0.00043688452569767833,
273
+ "learning_rate": 4.981045975808428e-06,
274
+ "loss": 0.0118,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.6476253736300233,
279
+ "grad_norm": 0.0006113427225500345,
280
+ "learning_rate": 4.9698213364181985e-06,
281
+ "loss": 0.0084,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.6642311524410495,
286
+ "grad_norm": 0.001146040391176939,
287
+ "learning_rate": 4.958596697027969e-06,
288
+ "loss": 0.0001,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.6808369312520757,
293
+ "grad_norm": 0.022079484537243843,
294
+ "learning_rate": 4.947372057637739e-06,
295
+ "loss": 0.0135,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.697442710063102,
300
+ "grad_norm": 0.000534936785697937,
301
+ "learning_rate": 4.936147418247508e-06,
302
+ "loss": 0.0013,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.7140484888741282,
307
+ "grad_norm": 0.19137096405029297,
308
+ "learning_rate": 4.924922778857279e-06,
309
+ "loss": 0.0242,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.7306542676851544,
314
+ "grad_norm": 2.9649052521563135e-05,
315
+ "learning_rate": 4.913698139467049e-06,
316
+ "loss": 0.0048,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.7472600464961807,
321
+ "grad_norm": 0.0003890593070536852,
322
+ "learning_rate": 4.902473500076818e-06,
323
+ "loss": 0.0055,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.7638658253072069,
328
+ "grad_norm": 7.795329293003306e-05,
329
+ "learning_rate": 4.891248860686589e-06,
330
+ "loss": 0.0062,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.7804716041182331,
335
+ "grad_norm": 3.511108661768958e-05,
336
+ "learning_rate": 4.8800242212963586e-06,
337
+ "loss": 0.0032,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.7970773829292593,
342
+ "grad_norm": 0.0023988874163478613,
343
+ "learning_rate": 4.868799581906129e-06,
344
+ "loss": 0.0256,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.8136831617402857,
349
+ "grad_norm": 2.8709350772260223e-06,
350
+ "learning_rate": 4.8575749425159e-06,
351
+ "loss": 0.0,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.8302889405513119,
356
+ "grad_norm": 6.131248665042222e-05,
357
+ "learning_rate": 4.846350303125669e-06,
358
+ "loss": 0.0002,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.8468947193623381,
363
+ "grad_norm": 0.4060337245464325,
364
+ "learning_rate": 4.835125663735439e-06,
365
+ "loss": 0.0113,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.8635004981733644,
370
+ "grad_norm": 0.14530286192893982,
371
+ "learning_rate": 4.823901024345209e-06,
372
+ "loss": 0.0177,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.8801062769843906,
377
+ "grad_norm": 0.00015053209790494293,
378
+ "learning_rate": 4.812676384954979e-06,
379
+ "loss": 0.0083,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.8967120557954168,
384
+ "grad_norm": 2.9383112632785924e-05,
385
+ "learning_rate": 4.801451745564749e-06,
386
+ "loss": 0.0,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.913317834606443,
391
+ "grad_norm": 0.0002987831539940089,
392
+ "learning_rate": 4.7902271061745195e-06,
393
+ "loss": 0.0013,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.9299236134174693,
398
+ "grad_norm": 9.85327574198891e-07,
399
+ "learning_rate": 4.7790024667842896e-06,
400
+ "loss": 0.0009,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.9465293922284955,
405
+ "grad_norm": 0.0004204445576760918,
406
+ "learning_rate": 4.76777782739406e-06,
407
+ "loss": 0.0,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.9631351710395217,
412
+ "grad_norm": 0.02836577780544758,
413
+ "learning_rate": 4.75655318800383e-06,
414
+ "loss": 0.0017,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.9797409498505479,
419
+ "grad_norm": 12.173595428466797,
420
+ "learning_rate": 4.7453285486136e-06,
421
+ "loss": 0.0284,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 0.9963467286615743,
426
+ "grad_norm": 0.0006245356635190547,
427
+ "learning_rate": 4.73410390922337e-06,
428
+ "loss": 0.0003,
429
+ "step": 3000
430
+ },
431
+ {
432
+ "epoch": 1.0,
433
+ "eval_accuracy": 0.9964217557251909,
434
+ "eval_f1": 0.9964322542520268,
435
+ "eval_loss": 0.02747241035103798,
436
+ "eval_precision": 0.9964456382767726,
437
+ "eval_recall": 0.9964217557251909,
438
+ "eval_runtime": 70.9146,
439
+ "eval_samples_per_second": 118.227,
440
+ "eval_steps_per_second": 7.389,
441
+ "step": 3011
442
+ }
443
+ ],
444
+ "logging_steps": 50,
445
+ "max_steps": 24088,
446
+ "num_input_tokens_seen": 0,
447
+ "num_train_epochs": 8,
448
+ "save_steps": 500,
449
+ "stateful_callbacks": {
450
+ "TrainerControl": {
451
+ "args": {
452
+ "should_epoch_stop": false,
453
+ "should_evaluate": false,
454
+ "should_log": false,
455
+ "should_save": true,
456
+ "should_training_stop": false
457
+ },
458
+ "attributes": {}
459
+ }
460
+ },
461
+ "total_flos": 5.094247305025536e+16,
462
+ "train_batch_size": 16,
463
+ "trial_name": null,
464
+ "trial_params": null
465
+ }
trial-0/checkpoint-3011/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2d43680079f69ace9b061a9cddc7889b3b2058a373dcb9cabe67e7a0b586646
3
+ size 5368
trial-1/checkpoint-15055/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-1/checkpoint-15055/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce6a1060ef93e1b5dc13c4ed9a3b6c77d2c792f107b20ff17933b9d4c15ecf0a
3
+ size 1583351632
trial-1/checkpoint-15055/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92795446fb40fea40977d388996f597e156f884cc50faed369041fd42add878a
3
+ size 3166813178
trial-1/checkpoint-15055/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb45f9d32a7f047001ed27e329de64c1bf9ebcb398e7f4734ee49c0dcd24d49
3
+ size 14244
trial-1/checkpoint-15055/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f4201dceb29ec1f3f21db3295b8c8fbe60d56c450ea048b5457e0cc4bfa0108
3
+ size 1064
trial-1/checkpoint-15055/trainer_state.json ADDED
@@ -0,0 +1,2200 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.026163995265960693,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-1/checkpoint-15055",
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 15055,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016605778811026237,
13
+ "grad_norm": 97.8153076171875,
14
+ "learning_rate": 4.1822378752775576e-05,
15
+ "loss": 0.332,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.033211557622052475,
20
+ "grad_norm": 0.13549020886421204,
21
+ "learning_rate": 4.1722929523801454e-05,
22
+ "loss": 0.1561,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.04981733643307871,
27
+ "grad_norm": 30.323139190673828,
28
+ "learning_rate": 4.162348029482734e-05,
29
+ "loss": 0.1222,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.06642311524410495,
34
+ "grad_norm": 0.0052522895857691765,
35
+ "learning_rate": 4.1524031065853216e-05,
36
+ "loss": 0.0938,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.08302889405513118,
41
+ "grad_norm": 8.804322242736816,
42
+ "learning_rate": 4.14245818368791e-05,
43
+ "loss": 0.0887,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.09963467286615742,
48
+ "grad_norm": 0.09965868294239044,
49
+ "learning_rate": 4.1325132607904985e-05,
50
+ "loss": 0.0998,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.11624045167718366,
55
+ "grad_norm": 0.03749964013695717,
56
+ "learning_rate": 4.122568337893087e-05,
57
+ "loss": 0.0552,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.1328462304882099,
62
+ "grad_norm": 0.2192620187997818,
63
+ "learning_rate": 4.112623414995675e-05,
64
+ "loss": 0.0717,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.14945200929923613,
69
+ "grad_norm": 136.17337036132812,
70
+ "learning_rate": 4.102678492098263e-05,
71
+ "loss": 0.0343,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.16605778811026237,
76
+ "grad_norm": 0.006200531497597694,
77
+ "learning_rate": 4.092733569200851e-05,
78
+ "loss": 0.0584,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.1826635669212886,
83
+ "grad_norm": 0.032046183943748474,
84
+ "learning_rate": 4.0827886463034394e-05,
85
+ "loss": 0.0458,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.19926934573231483,
90
+ "grad_norm": 3.6847121715545654,
91
+ "learning_rate": 4.072843723406027e-05,
92
+ "loss": 0.0463,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.2158751245433411,
97
+ "grad_norm": 9.91704273223877,
98
+ "learning_rate": 4.0628988005086156e-05,
99
+ "loss": 0.044,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.23248090335436733,
104
+ "grad_norm": 0.02639286033809185,
105
+ "learning_rate": 4.052953877611204e-05,
106
+ "loss": 0.0353,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.24908668216539356,
111
+ "grad_norm": 0.007340637035667896,
112
+ "learning_rate": 4.0430089547137925e-05,
113
+ "loss": 0.0519,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.2656924609764198,
118
+ "grad_norm": 0.003980421461164951,
119
+ "learning_rate": 4.033064031816381e-05,
120
+ "loss": 0.0588,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.282298239787446,
125
+ "grad_norm": 1.6867443323135376,
126
+ "learning_rate": 4.023119108918969e-05,
127
+ "loss": 0.0766,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.29890401859847227,
132
+ "grad_norm": 0.07216636836528778,
133
+ "learning_rate": 4.013174186021557e-05,
134
+ "loss": 0.0646,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.3155097974094985,
139
+ "grad_norm": 0.0015164370415732265,
140
+ "learning_rate": 4.003229263124145e-05,
141
+ "loss": 0.0342,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.33211557622052473,
146
+ "grad_norm": 0.004687592852860689,
147
+ "learning_rate": 3.993284340226733e-05,
148
+ "loss": 0.0327,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.348721355031551,
153
+ "grad_norm": 0.0013525976100936532,
154
+ "learning_rate": 3.983339417329321e-05,
155
+ "loss": 0.0362,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.3653271338425772,
160
+ "grad_norm": 0.021217679604887962,
161
+ "learning_rate": 3.9733944944319096e-05,
162
+ "loss": 0.0251,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.38193291265360346,
167
+ "grad_norm": 0.0011278189485892653,
168
+ "learning_rate": 3.963449571534498e-05,
169
+ "loss": 0.0136,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.39853869146462967,
174
+ "grad_norm": 0.0034522530622780323,
175
+ "learning_rate": 3.9535046486370864e-05,
176
+ "loss": 0.021,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.41514447027565593,
181
+ "grad_norm": 0.20626066625118256,
182
+ "learning_rate": 3.943559725739674e-05,
183
+ "loss": 0.0188,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.4317502490866822,
188
+ "grad_norm": 0.11095874011516571,
189
+ "learning_rate": 3.9336148028422627e-05,
190
+ "loss": 0.0275,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.4483560278977084,
195
+ "grad_norm": 0.09171418100595474,
196
+ "learning_rate": 3.9236698799448504e-05,
197
+ "loss": 0.0386,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.46496180670873466,
202
+ "grad_norm": 0.004308766219764948,
203
+ "learning_rate": 3.913724957047439e-05,
204
+ "loss": 0.0199,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.48156758551976087,
209
+ "grad_norm": 0.099054716527462,
210
+ "learning_rate": 3.9037800341500266e-05,
211
+ "loss": 0.0349,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.4981733643307871,
216
+ "grad_norm": 0.000510412035509944,
217
+ "learning_rate": 3.893835111252615e-05,
218
+ "loss": 0.0405,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.5147791431418134,
223
+ "grad_norm": 0.00045745610259473324,
224
+ "learning_rate": 3.8838901883552035e-05,
225
+ "loss": 0.0081,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.5313849219528396,
230
+ "grad_norm": 2.640720844268799,
231
+ "learning_rate": 3.873945265457791e-05,
232
+ "loss": 0.0312,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.5479907007638658,
237
+ "grad_norm": 0.0317668542265892,
238
+ "learning_rate": 3.86400034256038e-05,
239
+ "loss": 0.0231,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.564596479574892,
244
+ "grad_norm": 0.0374116413295269,
245
+ "learning_rate": 3.854055419662968e-05,
246
+ "loss": 0.0062,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.5812022583859183,
251
+ "grad_norm": 3.7858481407165527,
252
+ "learning_rate": 3.8441104967655566e-05,
253
+ "loss": 0.0172,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.5978080371969445,
258
+ "grad_norm": 0.006008578464388847,
259
+ "learning_rate": 3.8341655738681444e-05,
260
+ "loss": 0.0317,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.6144138160079707,
265
+ "grad_norm": 0.009135287255048752,
266
+ "learning_rate": 3.824220650970733e-05,
267
+ "loss": 0.0267,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.631019594818997,
272
+ "grad_norm": 0.004011407028883696,
273
+ "learning_rate": 3.8142757280733206e-05,
274
+ "loss": 0.0218,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.6476253736300233,
279
+ "grad_norm": 1.2146693468093872,
280
+ "learning_rate": 3.804330805175909e-05,
281
+ "loss": 0.0284,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.6642311524410495,
286
+ "grad_norm": 0.0033848509192466736,
287
+ "learning_rate": 3.794385882278497e-05,
288
+ "loss": 0.0241,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.6808369312520757,
293
+ "grad_norm": 0.0062151020392775536,
294
+ "learning_rate": 3.784440959381085e-05,
295
+ "loss": 0.0126,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.697442710063102,
300
+ "grad_norm": 0.004486024379730225,
301
+ "learning_rate": 3.774496036483674e-05,
302
+ "loss": 0.01,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.7140484888741282,
307
+ "grad_norm": 0.2388181835412979,
308
+ "learning_rate": 3.764551113586262e-05,
309
+ "loss": 0.029,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.7306542676851544,
314
+ "grad_norm": 0.0008192298118956387,
315
+ "learning_rate": 3.75460619068885e-05,
316
+ "loss": 0.0042,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.7472600464961807,
321
+ "grad_norm": 5.3397710871649906e-05,
322
+ "learning_rate": 3.7446612677914384e-05,
323
+ "loss": 0.003,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.7638658253072069,
328
+ "grad_norm": 0.020504744723439217,
329
+ "learning_rate": 3.734716344894026e-05,
330
+ "loss": 0.0262,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.7804716041182331,
335
+ "grad_norm": 0.0005639814771711826,
336
+ "learning_rate": 3.7247714219966146e-05,
337
+ "loss": 0.0043,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.7970773829292593,
342
+ "grad_norm": 0.0031704490538686514,
343
+ "learning_rate": 3.714826499099203e-05,
344
+ "loss": 0.0284,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.8136831617402857,
349
+ "grad_norm": 0.001501227729022503,
350
+ "learning_rate": 3.704881576201791e-05,
351
+ "loss": 0.0139,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.8302889405513119,
356
+ "grad_norm": 0.0010695902165025473,
357
+ "learning_rate": 3.694936653304379e-05,
358
+ "loss": 0.0002,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.8468947193623381,
363
+ "grad_norm": 0.0022703870199620724,
364
+ "learning_rate": 3.684991730406968e-05,
365
+ "loss": 0.0127,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.8635004981733644,
370
+ "grad_norm": 0.057379350066185,
371
+ "learning_rate": 3.675046807509556e-05,
372
+ "loss": 0.0185,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.8801062769843906,
377
+ "grad_norm": 0.00022332114167511463,
378
+ "learning_rate": 3.665101884612144e-05,
379
+ "loss": 0.0305,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.8967120557954168,
384
+ "grad_norm": 0.0001592856424394995,
385
+ "learning_rate": 3.6551569617147324e-05,
386
+ "loss": 0.0123,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.913317834606443,
391
+ "grad_norm": 17.401182174682617,
392
+ "learning_rate": 3.64521203881732e-05,
393
+ "loss": 0.0185,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.9299236134174693,
398
+ "grad_norm": 6.72700916766189e-05,
399
+ "learning_rate": 3.6352671159199086e-05,
400
+ "loss": 0.0069,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.9465293922284955,
405
+ "grad_norm": 0.004428895190358162,
406
+ "learning_rate": 3.6253221930224964e-05,
407
+ "loss": 0.0007,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.9631351710395217,
412
+ "grad_norm": 1.0974332094192505,
413
+ "learning_rate": 3.615377270125085e-05,
414
+ "loss": 0.0404,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.9797409498505479,
419
+ "grad_norm": 1.9213542938232422,
420
+ "learning_rate": 3.605432347227673e-05,
421
+ "loss": 0.0347,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 0.9963467286615743,
426
+ "grad_norm": 0.00922548957169056,
427
+ "learning_rate": 3.595487424330261e-05,
428
+ "loss": 0.0012,
429
+ "step": 3000
430
+ },
431
+ {
432
+ "epoch": 1.0,
433
+ "eval_accuracy": 0.9964217557251909,
434
+ "eval_f1": 0.9964374533901769,
435
+ "eval_loss": 0.0292544886469841,
436
+ "eval_precision": 0.9964596067772543,
437
+ "eval_recall": 0.9964217557251909,
438
+ "eval_runtime": 66.9274,
439
+ "eval_samples_per_second": 125.27,
440
+ "eval_steps_per_second": 7.829,
441
+ "step": 3011
442
+ },
443
+ {
444
+ "epoch": 1.0129525074726005,
445
+ "grad_norm": 0.0020659081637859344,
446
+ "learning_rate": 3.5855425014328495e-05,
447
+ "loss": 0.006,
448
+ "step": 3050
449
+ },
450
+ {
451
+ "epoch": 1.0295582862836268,
452
+ "grad_norm": 0.0007023397483862936,
453
+ "learning_rate": 3.575597578535438e-05,
454
+ "loss": 0.0252,
455
+ "step": 3100
456
+ },
457
+ {
458
+ "epoch": 1.0461640650946529,
459
+ "grad_norm": 0.0009625868406146765,
460
+ "learning_rate": 3.565652655638026e-05,
461
+ "loss": 0.0009,
462
+ "step": 3150
463
+ },
464
+ {
465
+ "epoch": 1.0627698439056792,
466
+ "grad_norm": 0.003358361078426242,
467
+ "learning_rate": 3.555707732740614e-05,
468
+ "loss": 0.0023,
469
+ "step": 3200
470
+ },
471
+ {
472
+ "epoch": 1.0793756227167055,
473
+ "grad_norm": 0.0069133201614022255,
474
+ "learning_rate": 3.5457628098432026e-05,
475
+ "loss": 0.0173,
476
+ "step": 3250
477
+ },
478
+ {
479
+ "epoch": 1.0959814015277316,
480
+ "grad_norm": 0.002833959646522999,
481
+ "learning_rate": 3.5358178869457903e-05,
482
+ "loss": 0.0103,
483
+ "step": 3300
484
+ },
485
+ {
486
+ "epoch": 1.112587180338758,
487
+ "grad_norm": 0.3491974174976349,
488
+ "learning_rate": 3.525872964048379e-05,
489
+ "loss": 0.0106,
490
+ "step": 3350
491
+ },
492
+ {
493
+ "epoch": 1.1291929591497842,
494
+ "grad_norm": 0.002338677179068327,
495
+ "learning_rate": 3.5159280411509666e-05,
496
+ "loss": 0.0113,
497
+ "step": 3400
498
+ },
499
+ {
500
+ "epoch": 1.1457987379608103,
501
+ "grad_norm": 0.01391940750181675,
502
+ "learning_rate": 3.505983118253555e-05,
503
+ "loss": 0.0629,
504
+ "step": 3450
505
+ },
506
+ {
507
+ "epoch": 1.1624045167718366,
508
+ "grad_norm": 0.01091580931097269,
509
+ "learning_rate": 3.4960381953561435e-05,
510
+ "loss": 0.0124,
511
+ "step": 3500
512
+ },
513
+ {
514
+ "epoch": 1.1790102955828627,
515
+ "grad_norm": 0.08861105889081955,
516
+ "learning_rate": 3.486093272458732e-05,
517
+ "loss": 0.005,
518
+ "step": 3550
519
+ },
520
+ {
521
+ "epoch": 1.195616074393889,
522
+ "grad_norm": 13.801424980163574,
523
+ "learning_rate": 3.47614834956132e-05,
524
+ "loss": 0.0198,
525
+ "step": 3600
526
+ },
527
+ {
528
+ "epoch": 1.2122218532049154,
529
+ "grad_norm": 0.0011831964366137981,
530
+ "learning_rate": 3.466203426663908e-05,
531
+ "loss": 0.0199,
532
+ "step": 3650
533
+ },
534
+ {
535
+ "epoch": 1.2288276320159415,
536
+ "grad_norm": 0.0007713422528468072,
537
+ "learning_rate": 3.456258503766496e-05,
538
+ "loss": 0.0002,
539
+ "step": 3700
540
+ },
541
+ {
542
+ "epoch": 1.2454334108269678,
543
+ "grad_norm": 0.012040040455758572,
544
+ "learning_rate": 3.446313580869084e-05,
545
+ "loss": 0.0145,
546
+ "step": 3750
547
+ },
548
+ {
549
+ "epoch": 1.2620391896379939,
550
+ "grad_norm": 0.002991499612107873,
551
+ "learning_rate": 3.436368657971672e-05,
552
+ "loss": 0.0019,
553
+ "step": 3800
554
+ },
555
+ {
556
+ "epoch": 1.2786449684490202,
557
+ "grad_norm": 0.0011237855069339275,
558
+ "learning_rate": 3.4264237350742605e-05,
559
+ "loss": 0.0121,
560
+ "step": 3850
561
+ },
562
+ {
563
+ "epoch": 1.2952507472600465,
564
+ "grad_norm": 0.0003940507594961673,
565
+ "learning_rate": 3.416478812176849e-05,
566
+ "loss": 0.0005,
567
+ "step": 3900
568
+ },
569
+ {
570
+ "epoch": 1.3118565260710726,
571
+ "grad_norm": 0.000680712575558573,
572
+ "learning_rate": 3.4065338892794374e-05,
573
+ "loss": 0.0064,
574
+ "step": 3950
575
+ },
576
+ {
577
+ "epoch": 1.328462304882099,
578
+ "grad_norm": 0.0005092715146020055,
579
+ "learning_rate": 3.396588966382026e-05,
580
+ "loss": 0.0015,
581
+ "step": 4000
582
+ },
583
+ {
584
+ "epoch": 1.3450680836931252,
585
+ "grad_norm": 0.05994075909256935,
586
+ "learning_rate": 3.3866440434846137e-05,
587
+ "loss": 0.012,
588
+ "step": 4050
589
+ },
590
+ {
591
+ "epoch": 1.3616738625041513,
592
+ "grad_norm": 0.00185777444858104,
593
+ "learning_rate": 3.376699120587202e-05,
594
+ "loss": 0.011,
595
+ "step": 4100
596
+ },
597
+ {
598
+ "epoch": 1.3782796413151777,
599
+ "grad_norm": 0.000904878368601203,
600
+ "learning_rate": 3.36675419768979e-05,
601
+ "loss": 0.005,
602
+ "step": 4150
603
+ },
604
+ {
605
+ "epoch": 1.394885420126204,
606
+ "grad_norm": 0.0007558612269349396,
607
+ "learning_rate": 3.356809274792378e-05,
608
+ "loss": 0.0047,
609
+ "step": 4200
610
+ },
611
+ {
612
+ "epoch": 1.41149119893723,
613
+ "grad_norm": 0.000539736298378557,
614
+ "learning_rate": 3.346864351894966e-05,
615
+ "loss": 0.0129,
616
+ "step": 4250
617
+ },
618
+ {
619
+ "epoch": 1.4280969777482564,
620
+ "grad_norm": 0.009128957986831665,
621
+ "learning_rate": 3.3369194289975545e-05,
622
+ "loss": 0.0243,
623
+ "step": 4300
624
+ },
625
+ {
626
+ "epoch": 1.4447027565592827,
627
+ "grad_norm": 0.0029375357553362846,
628
+ "learning_rate": 3.326974506100143e-05,
629
+ "loss": 0.0265,
630
+ "step": 4350
631
+ },
632
+ {
633
+ "epoch": 1.4613085353703088,
634
+ "grad_norm": 0.005572469439357519,
635
+ "learning_rate": 3.317029583202731e-05,
636
+ "loss": 0.0208,
637
+ "step": 4400
638
+ },
639
+ {
640
+ "epoch": 1.4779143141813351,
641
+ "grad_norm": 0.022462476044893265,
642
+ "learning_rate": 3.307084660305319e-05,
643
+ "loss": 0.028,
644
+ "step": 4450
645
+ },
646
+ {
647
+ "epoch": 1.4945200929923614,
648
+ "grad_norm": 0.03258312866091728,
649
+ "learning_rate": 3.2971397374079076e-05,
650
+ "loss": 0.0039,
651
+ "step": 4500
652
+ },
653
+ {
654
+ "epoch": 1.5111258718033875,
655
+ "grad_norm": 0.0003238330245949328,
656
+ "learning_rate": 3.2871948145104954e-05,
657
+ "loss": 0.0004,
658
+ "step": 4550
659
+ },
660
+ {
661
+ "epoch": 1.5277316506144138,
662
+ "grad_norm": 0.0003890040097758174,
663
+ "learning_rate": 3.277249891613084e-05,
664
+ "loss": 0.0019,
665
+ "step": 4600
666
+ },
667
+ {
668
+ "epoch": 1.5443374294254402,
669
+ "grad_norm": 0.0009322063415311277,
670
+ "learning_rate": 3.2673049687156716e-05,
671
+ "loss": 0.0179,
672
+ "step": 4650
673
+ },
674
+ {
675
+ "epoch": 1.5609432082364663,
676
+ "grad_norm": 0.01131466869264841,
677
+ "learning_rate": 3.25736004581826e-05,
678
+ "loss": 0.0104,
679
+ "step": 4700
680
+ },
681
+ {
682
+ "epoch": 1.5775489870474926,
683
+ "grad_norm": 13.522896766662598,
684
+ "learning_rate": 3.247415122920848e-05,
685
+ "loss": 0.0223,
686
+ "step": 4750
687
+ },
688
+ {
689
+ "epoch": 1.594154765858519,
690
+ "grad_norm": 0.02276255562901497,
691
+ "learning_rate": 3.237470200023436e-05,
692
+ "loss": 0.0158,
693
+ "step": 4800
694
+ },
695
+ {
696
+ "epoch": 1.610760544669545,
697
+ "grad_norm": 0.002166257705539465,
698
+ "learning_rate": 3.227525277126025e-05,
699
+ "loss": 0.0078,
700
+ "step": 4850
701
+ },
702
+ {
703
+ "epoch": 1.627366323480571,
704
+ "grad_norm": 1.3065482378005981,
705
+ "learning_rate": 3.217580354228613e-05,
706
+ "loss": 0.0141,
707
+ "step": 4900
708
+ },
709
+ {
710
+ "epoch": 1.6439721022915976,
711
+ "grad_norm": 0.003547689877450466,
712
+ "learning_rate": 3.2076354313312016e-05,
713
+ "loss": 0.015,
714
+ "step": 4950
715
+ },
716
+ {
717
+ "epoch": 1.6605778811026237,
718
+ "grad_norm": 0.014956770464777946,
719
+ "learning_rate": 3.1976905084337894e-05,
720
+ "loss": 0.0152,
721
+ "step": 5000
722
+ },
723
+ {
724
+ "epoch": 1.6771836599136498,
725
+ "grad_norm": 0.04130621254444122,
726
+ "learning_rate": 3.187745585536378e-05,
727
+ "loss": 0.0125,
728
+ "step": 5050
729
+ },
730
+ {
731
+ "epoch": 1.6937894387246761,
732
+ "grad_norm": 0.0008221206953749061,
733
+ "learning_rate": 3.1778006626389656e-05,
734
+ "loss": 0.0135,
735
+ "step": 5100
736
+ },
737
+ {
738
+ "epoch": 1.7103952175357025,
739
+ "grad_norm": 12.82378101348877,
740
+ "learning_rate": 3.167855739741554e-05,
741
+ "loss": 0.0068,
742
+ "step": 5150
743
+ },
744
+ {
745
+ "epoch": 1.7270009963467285,
746
+ "grad_norm": 0.00033835775684565306,
747
+ "learning_rate": 3.157910816844142e-05,
748
+ "loss": 0.0162,
749
+ "step": 5200
750
+ },
751
+ {
752
+ "epoch": 1.7436067751577549,
753
+ "grad_norm": 0.019336150959134102,
754
+ "learning_rate": 3.14796589394673e-05,
755
+ "loss": 0.0233,
756
+ "step": 5250
757
+ },
758
+ {
759
+ "epoch": 1.7602125539687812,
760
+ "grad_norm": 0.0025835856795310974,
761
+ "learning_rate": 3.138020971049319e-05,
762
+ "loss": 0.0049,
763
+ "step": 5300
764
+ },
765
+ {
766
+ "epoch": 1.7768183327798073,
767
+ "grad_norm": 0.001420403248630464,
768
+ "learning_rate": 3.128076048151907e-05,
769
+ "loss": 0.003,
770
+ "step": 5350
771
+ },
772
+ {
773
+ "epoch": 1.7934241115908336,
774
+ "grad_norm": 0.00038990622851997614,
775
+ "learning_rate": 3.118131125254495e-05,
776
+ "loss": 0.0006,
777
+ "step": 5400
778
+ },
779
+ {
780
+ "epoch": 1.81002989040186,
781
+ "grad_norm": 0.008520632050931454,
782
+ "learning_rate": 3.1081862023570834e-05,
783
+ "loss": 0.0017,
784
+ "step": 5450
785
+ },
786
+ {
787
+ "epoch": 1.826635669212886,
788
+ "grad_norm": 6.68273787596263e-05,
789
+ "learning_rate": 3.098241279459671e-05,
790
+ "loss": 0.0036,
791
+ "step": 5500
792
+ },
793
+ {
794
+ "epoch": 1.8432414480239123,
795
+ "grad_norm": 0.0003700813394971192,
796
+ "learning_rate": 3.0882963565622596e-05,
797
+ "loss": 0.0165,
798
+ "step": 5550
799
+ },
800
+ {
801
+ "epoch": 1.8598472268349386,
802
+ "grad_norm": 0.0006010610377416015,
803
+ "learning_rate": 3.078351433664848e-05,
804
+ "loss": 0.0087,
805
+ "step": 5600
806
+ },
807
+ {
808
+ "epoch": 1.8764530056459647,
809
+ "grad_norm": 0.003066062228754163,
810
+ "learning_rate": 3.068406510767436e-05,
811
+ "loss": 0.0146,
812
+ "step": 5650
813
+ },
814
+ {
815
+ "epoch": 1.893058784456991,
816
+ "grad_norm": 0.005507585126906633,
817
+ "learning_rate": 3.058461587870024e-05,
818
+ "loss": 0.0143,
819
+ "step": 5700
820
+ },
821
+ {
822
+ "epoch": 1.9096645632680174,
823
+ "grad_norm": 0.0004035944875795394,
824
+ "learning_rate": 3.0485166649726124e-05,
825
+ "loss": 0.0005,
826
+ "step": 5750
827
+ },
828
+ {
829
+ "epoch": 1.9262703420790435,
830
+ "grad_norm": 0.001009216532111168,
831
+ "learning_rate": 3.0385717420752008e-05,
832
+ "loss": 0.0001,
833
+ "step": 5800
834
+ },
835
+ {
836
+ "epoch": 1.9428761208900698,
837
+ "grad_norm": 0.000999059877358377,
838
+ "learning_rate": 3.0286268191777886e-05,
839
+ "loss": 0.0035,
840
+ "step": 5850
841
+ },
842
+ {
843
+ "epoch": 1.959481899701096,
844
+ "grad_norm": 0.009907165542244911,
845
+ "learning_rate": 3.018681896280377e-05,
846
+ "loss": 0.0169,
847
+ "step": 5900
848
+ },
849
+ {
850
+ "epoch": 1.9760876785121222,
851
+ "grad_norm": 0.0009444186580367386,
852
+ "learning_rate": 3.008736973382965e-05,
853
+ "loss": 0.0002,
854
+ "step": 5950
855
+ },
856
+ {
857
+ "epoch": 1.9926934573231485,
858
+ "grad_norm": 0.004821607377380133,
859
+ "learning_rate": 2.9987920504855536e-05,
860
+ "loss": 0.0279,
861
+ "step": 6000
862
+ },
863
+ {
864
+ "epoch": 2.0,
865
+ "eval_accuracy": 0.9960639312977099,
866
+ "eval_f1": 0.9960726064300177,
867
+ "eval_loss": 0.026200218126177788,
868
+ "eval_precision": 0.9960829017470937,
869
+ "eval_recall": 0.9960639312977099,
870
+ "eval_runtime": 66.8019,
871
+ "eval_samples_per_second": 125.505,
872
+ "eval_steps_per_second": 7.844,
873
+ "step": 6022
874
+ },
875
+ {
876
+ "epoch": 2.009299236134175,
877
+ "grad_norm": 0.005393081344664097,
878
+ "learning_rate": 2.9888471275881413e-05,
879
+ "loss": 0.0005,
880
+ "step": 6050
881
+ },
882
+ {
883
+ "epoch": 2.025905014945201,
884
+ "grad_norm": 0.0005983946030028164,
885
+ "learning_rate": 2.9789022046907298e-05,
886
+ "loss": 0.007,
887
+ "step": 6100
888
+ },
889
+ {
890
+ "epoch": 2.042510793756227,
891
+ "grad_norm": 0.009896568953990936,
892
+ "learning_rate": 2.968957281793318e-05,
893
+ "loss": 0.0126,
894
+ "step": 6150
895
+ },
896
+ {
897
+ "epoch": 2.0591165725672536,
898
+ "grad_norm": 0.0005078279646113515,
899
+ "learning_rate": 2.9590123588959063e-05,
900
+ "loss": 0.0036,
901
+ "step": 6200
902
+ },
903
+ {
904
+ "epoch": 2.0757223513782797,
905
+ "grad_norm": 0.003000972094014287,
906
+ "learning_rate": 2.949067435998494e-05,
907
+ "loss": 0.0084,
908
+ "step": 6250
909
+ },
910
+ {
911
+ "epoch": 2.0923281301893057,
912
+ "grad_norm": 0.000705482205376029,
913
+ "learning_rate": 2.9391225131010826e-05,
914
+ "loss": 0.0114,
915
+ "step": 6300
916
+ },
917
+ {
918
+ "epoch": 2.1089339090003323,
919
+ "grad_norm": 0.0010040552588179708,
920
+ "learning_rate": 2.929177590203671e-05,
921
+ "loss": 0.0001,
922
+ "step": 6350
923
+ },
924
+ {
925
+ "epoch": 2.1255396878113584,
926
+ "grad_norm": 0.00919084157794714,
927
+ "learning_rate": 2.919232667306259e-05,
928
+ "loss": 0.0132,
929
+ "step": 6400
930
+ },
931
+ {
932
+ "epoch": 2.1421454666223845,
933
+ "grad_norm": 0.0013779608998447657,
934
+ "learning_rate": 2.9092877444088476e-05,
935
+ "loss": 0.0005,
936
+ "step": 6450
937
+ },
938
+ {
939
+ "epoch": 2.158751245433411,
940
+ "grad_norm": 0.0005553108640015125,
941
+ "learning_rate": 2.8993428215114353e-05,
942
+ "loss": 0.0024,
943
+ "step": 6500
944
+ },
945
+ {
946
+ "epoch": 2.175357024244437,
947
+ "grad_norm": 0.0021922625601291656,
948
+ "learning_rate": 2.8893978986140238e-05,
949
+ "loss": 0.0001,
950
+ "step": 6550
951
+ },
952
+ {
953
+ "epoch": 2.191962803055463,
954
+ "grad_norm": 0.0002919725957326591,
955
+ "learning_rate": 2.879452975716612e-05,
956
+ "loss": 0.0016,
957
+ "step": 6600
958
+ },
959
+ {
960
+ "epoch": 2.2085685818664897,
961
+ "grad_norm": 0.0009692521998658776,
962
+ "learning_rate": 2.8695080528192003e-05,
963
+ "loss": 0.0,
964
+ "step": 6650
965
+ },
966
+ {
967
+ "epoch": 2.225174360677516,
968
+ "grad_norm": 0.0002649214584380388,
969
+ "learning_rate": 2.859563129921788e-05,
970
+ "loss": 0.0065,
971
+ "step": 6700
972
+ },
973
+ {
974
+ "epoch": 2.241780139488542,
975
+ "grad_norm": 0.00046473185648210347,
976
+ "learning_rate": 2.8496182070243765e-05,
977
+ "loss": 0.0045,
978
+ "step": 6750
979
+ },
980
+ {
981
+ "epoch": 2.2583859182995685,
982
+ "grad_norm": 0.00027421273989602923,
983
+ "learning_rate": 2.8396732841269647e-05,
984
+ "loss": 0.0001,
985
+ "step": 6800
986
+ },
987
+ {
988
+ "epoch": 2.2749916971105946,
989
+ "grad_norm": 0.0005349958664737642,
990
+ "learning_rate": 2.829728361229553e-05,
991
+ "loss": 0.0064,
992
+ "step": 6850
993
+ },
994
+ {
995
+ "epoch": 2.2915974759216207,
996
+ "grad_norm": 0.013357802294194698,
997
+ "learning_rate": 2.819783438332141e-05,
998
+ "loss": 0.0166,
999
+ "step": 6900
1000
+ },
1001
+ {
1002
+ "epoch": 2.308203254732647,
1003
+ "grad_norm": 0.00018077288405038416,
1004
+ "learning_rate": 2.8098385154347293e-05,
1005
+ "loss": 0.0003,
1006
+ "step": 6950
1007
+ },
1008
+ {
1009
+ "epoch": 2.3248090335436733,
1010
+ "grad_norm": 0.00022416921274270862,
1011
+ "learning_rate": 2.7998935925373174e-05,
1012
+ "loss": 0.0258,
1013
+ "step": 7000
1014
+ },
1015
+ {
1016
+ "epoch": 2.3414148123546994,
1017
+ "grad_norm": 0.0035074173938483,
1018
+ "learning_rate": 2.789948669639906e-05,
1019
+ "loss": 0.0235,
1020
+ "step": 7050
1021
+ },
1022
+ {
1023
+ "epoch": 2.3580205911657255,
1024
+ "grad_norm": 0.03189694508910179,
1025
+ "learning_rate": 2.7800037467424936e-05,
1026
+ "loss": 0.0092,
1027
+ "step": 7100
1028
+ },
1029
+ {
1030
+ "epoch": 2.374626369976752,
1031
+ "grad_norm": 0.003983413800597191,
1032
+ "learning_rate": 2.770058823845082e-05,
1033
+ "loss": 0.0178,
1034
+ "step": 7150
1035
+ },
1036
+ {
1037
+ "epoch": 2.391232148787778,
1038
+ "grad_norm": 0.002308150054886937,
1039
+ "learning_rate": 2.7601139009476705e-05,
1040
+ "loss": 0.0038,
1041
+ "step": 7200
1042
+ },
1043
+ {
1044
+ "epoch": 2.407837927598804,
1045
+ "grad_norm": 0.0013349172659218311,
1046
+ "learning_rate": 2.7501689780502583e-05,
1047
+ "loss": 0.001,
1048
+ "step": 7250
1049
+ },
1050
+ {
1051
+ "epoch": 2.4244437064098308,
1052
+ "grad_norm": 0.000329138885717839,
1053
+ "learning_rate": 2.7402240551528467e-05,
1054
+ "loss": 0.0,
1055
+ "step": 7300
1056
+ },
1057
+ {
1058
+ "epoch": 2.441049485220857,
1059
+ "grad_norm": 0.0008932758355513215,
1060
+ "learning_rate": 2.730279132255435e-05,
1061
+ "loss": 0.0,
1062
+ "step": 7350
1063
+ },
1064
+ {
1065
+ "epoch": 2.457655264031883,
1066
+ "grad_norm": 0.0005681042093783617,
1067
+ "learning_rate": 2.7203342093580233e-05,
1068
+ "loss": 0.0,
1069
+ "step": 7400
1070
+ },
1071
+ {
1072
+ "epoch": 2.4742610428429095,
1073
+ "grad_norm": 0.00035528288572095335,
1074
+ "learning_rate": 2.710389286460611e-05,
1075
+ "loss": 0.0006,
1076
+ "step": 7450
1077
+ },
1078
+ {
1079
+ "epoch": 2.4908668216539356,
1080
+ "grad_norm": 0.0064411167986691,
1081
+ "learning_rate": 2.7004443635631995e-05,
1082
+ "loss": 0.0057,
1083
+ "step": 7500
1084
+ },
1085
+ {
1086
+ "epoch": 2.5074726004649617,
1087
+ "grad_norm": 0.004782046191394329,
1088
+ "learning_rate": 2.6904994406657876e-05,
1089
+ "loss": 0.0104,
1090
+ "step": 7550
1091
+ },
1092
+ {
1093
+ "epoch": 2.5240783792759878,
1094
+ "grad_norm": 0.007697090040892363,
1095
+ "learning_rate": 2.680554517768376e-05,
1096
+ "loss": 0.0194,
1097
+ "step": 7600
1098
+ },
1099
+ {
1100
+ "epoch": 2.5406841580870143,
1101
+ "grad_norm": 0.0027768309228122234,
1102
+ "learning_rate": 2.670609594870964e-05,
1103
+ "loss": 0.0003,
1104
+ "step": 7650
1105
+ },
1106
+ {
1107
+ "epoch": 2.5572899368980404,
1108
+ "grad_norm": 0.004415275063365698,
1109
+ "learning_rate": 2.6606646719735523e-05,
1110
+ "loss": 0.0008,
1111
+ "step": 7700
1112
+ },
1113
+ {
1114
+ "epoch": 2.5738957157090665,
1115
+ "grad_norm": 0.0020732246339321136,
1116
+ "learning_rate": 2.6507197490761404e-05,
1117
+ "loss": 0.0099,
1118
+ "step": 7750
1119
+ },
1120
+ {
1121
+ "epoch": 2.590501494520093,
1122
+ "grad_norm": 0.01230535376816988,
1123
+ "learning_rate": 2.640774826178729e-05,
1124
+ "loss": 0.0013,
1125
+ "step": 7800
1126
+ },
1127
+ {
1128
+ "epoch": 2.607107273331119,
1129
+ "grad_norm": 0.00023796973982825875,
1130
+ "learning_rate": 2.6308299032813166e-05,
1131
+ "loss": 0.0001,
1132
+ "step": 7850
1133
+ },
1134
+ {
1135
+ "epoch": 2.6237130521421452,
1136
+ "grad_norm": 0.0007237173849716783,
1137
+ "learning_rate": 2.620884980383905e-05,
1138
+ "loss": 0.0,
1139
+ "step": 7900
1140
+ },
1141
+ {
1142
+ "epoch": 2.6403188309531718,
1143
+ "grad_norm": 0.012582485564053059,
1144
+ "learning_rate": 2.6109400574864935e-05,
1145
+ "loss": 0.0137,
1146
+ "step": 7950
1147
+ },
1148
+ {
1149
+ "epoch": 2.656924609764198,
1150
+ "grad_norm": 0.004291553515940905,
1151
+ "learning_rate": 2.6009951345890816e-05,
1152
+ "loss": 0.0004,
1153
+ "step": 8000
1154
+ },
1155
+ {
1156
+ "epoch": 2.673530388575224,
1157
+ "grad_norm": 0.001662308000959456,
1158
+ "learning_rate": 2.59105021169167e-05,
1159
+ "loss": 0.0001,
1160
+ "step": 8050
1161
+ },
1162
+ {
1163
+ "epoch": 2.6901361673862505,
1164
+ "grad_norm": 0.0007901078206487,
1165
+ "learning_rate": 2.5811052887942578e-05,
1166
+ "loss": 0.0001,
1167
+ "step": 8100
1168
+ },
1169
+ {
1170
+ "epoch": 2.7067419461972766,
1171
+ "grad_norm": 0.0007069796556606889,
1172
+ "learning_rate": 2.5711603658968463e-05,
1173
+ "loss": 0.0001,
1174
+ "step": 8150
1175
+ },
1176
+ {
1177
+ "epoch": 2.7233477250083027,
1178
+ "grad_norm": 0.0006272114696912467,
1179
+ "learning_rate": 2.5612154429994344e-05,
1180
+ "loss": 0.0,
1181
+ "step": 8200
1182
+ },
1183
+ {
1184
+ "epoch": 2.7399535038193292,
1185
+ "grad_norm": 0.0008849167497828603,
1186
+ "learning_rate": 2.5512705201020228e-05,
1187
+ "loss": 0.0108,
1188
+ "step": 8250
1189
+ },
1190
+ {
1191
+ "epoch": 2.7565592826303553,
1192
+ "grad_norm": 0.0023204211611300707,
1193
+ "learning_rate": 2.5413255972046106e-05,
1194
+ "loss": 0.0015,
1195
+ "step": 8300
1196
+ },
1197
+ {
1198
+ "epoch": 2.7731650614413814,
1199
+ "grad_norm": 0.0006785548175685108,
1200
+ "learning_rate": 2.531380674307199e-05,
1201
+ "loss": 0.0001,
1202
+ "step": 8350
1203
+ },
1204
+ {
1205
+ "epoch": 2.789770840252408,
1206
+ "grad_norm": 0.0012301064562052488,
1207
+ "learning_rate": 2.521435751409787e-05,
1208
+ "loss": 0.0001,
1209
+ "step": 8400
1210
+ },
1211
+ {
1212
+ "epoch": 2.806376619063434,
1213
+ "grad_norm": 0.0012826485326513648,
1214
+ "learning_rate": 2.5114908285123756e-05,
1215
+ "loss": 0.0005,
1216
+ "step": 8450
1217
+ },
1218
+ {
1219
+ "epoch": 2.82298239787446,
1220
+ "grad_norm": 0.00045186831266619265,
1221
+ "learning_rate": 2.5015459056149634e-05,
1222
+ "loss": 0.0,
1223
+ "step": 8500
1224
+ },
1225
+ {
1226
+ "epoch": 2.8395881766854867,
1227
+ "grad_norm": 0.004168716724961996,
1228
+ "learning_rate": 2.4916009827175518e-05,
1229
+ "loss": 0.0113,
1230
+ "step": 8550
1231
+ },
1232
+ {
1233
+ "epoch": 2.856193955496513,
1234
+ "grad_norm": 0.0053441463969647884,
1235
+ "learning_rate": 2.48165605982014e-05,
1236
+ "loss": 0.0148,
1237
+ "step": 8600
1238
+ },
1239
+ {
1240
+ "epoch": 2.872799734307539,
1241
+ "grad_norm": 0.002283324720337987,
1242
+ "learning_rate": 2.471711136922728e-05,
1243
+ "loss": 0.0001,
1244
+ "step": 8650
1245
+ },
1246
+ {
1247
+ "epoch": 2.8894055131185654,
1248
+ "grad_norm": 0.0046396502293646336,
1249
+ "learning_rate": 2.461766214025316e-05,
1250
+ "loss": 0.009,
1251
+ "step": 8700
1252
+ },
1253
+ {
1254
+ "epoch": 2.9060112919295915,
1255
+ "grad_norm": 0.01257924735546112,
1256
+ "learning_rate": 2.4518212911279046e-05,
1257
+ "loss": 0.015,
1258
+ "step": 8750
1259
+ },
1260
+ {
1261
+ "epoch": 2.9226170707406176,
1262
+ "grad_norm": 0.00245088548399508,
1263
+ "learning_rate": 2.441876368230493e-05,
1264
+ "loss": 0.0136,
1265
+ "step": 8800
1266
+ },
1267
+ {
1268
+ "epoch": 2.939222849551644,
1269
+ "grad_norm": 0.005676699336618185,
1270
+ "learning_rate": 2.4319314453330808e-05,
1271
+ "loss": 0.005,
1272
+ "step": 8850
1273
+ },
1274
+ {
1275
+ "epoch": 2.9558286283626702,
1276
+ "grad_norm": 0.0012465333566069603,
1277
+ "learning_rate": 2.4219865224356692e-05,
1278
+ "loss": 0.0002,
1279
+ "step": 8900
1280
+ },
1281
+ {
1282
+ "epoch": 2.9724344071736963,
1283
+ "grad_norm": 0.0007043919176794589,
1284
+ "learning_rate": 2.4120415995382573e-05,
1285
+ "loss": 0.0001,
1286
+ "step": 8950
1287
+ },
1288
+ {
1289
+ "epoch": 2.989040185984723,
1290
+ "grad_norm": 0.007725800387561321,
1291
+ "learning_rate": 2.4020966766408458e-05,
1292
+ "loss": 0.0,
1293
+ "step": 9000
1294
+ },
1295
+ {
1296
+ "epoch": 3.0,
1297
+ "eval_accuracy": 0.9971374045801527,
1298
+ "eval_f1": 0.9971159280148129,
1299
+ "eval_loss": 0.030433131381869316,
1300
+ "eval_precision": 0.9971134119828018,
1301
+ "eval_recall": 0.9971374045801527,
1302
+ "eval_runtime": 66.6186,
1303
+ "eval_samples_per_second": 125.851,
1304
+ "eval_steps_per_second": 7.866,
1305
+ "step": 9033
1306
+ },
1307
+ {
1308
+ "epoch": 3.005645964795749,
1309
+ "grad_norm": 0.015563803724944592,
1310
+ "learning_rate": 2.3921517537434336e-05,
1311
+ "loss": 0.0121,
1312
+ "step": 9050
1313
+ },
1314
+ {
1315
+ "epoch": 3.022251743606775,
1316
+ "grad_norm": 0.0009268106077797711,
1317
+ "learning_rate": 2.382206830846022e-05,
1318
+ "loss": 0.002,
1319
+ "step": 9100
1320
+ },
1321
+ {
1322
+ "epoch": 3.0388575224178016,
1323
+ "grad_norm": 0.003995839972048998,
1324
+ "learning_rate": 2.37226190794861e-05,
1325
+ "loss": 0.0001,
1326
+ "step": 9150
1327
+ },
1328
+ {
1329
+ "epoch": 3.0554633012288277,
1330
+ "grad_norm": 0.0004984468687325716,
1331
+ "learning_rate": 2.3623169850511986e-05,
1332
+ "loss": 0.0001,
1333
+ "step": 9200
1334
+ },
1335
+ {
1336
+ "epoch": 3.072069080039854,
1337
+ "grad_norm": 0.0007679828559048474,
1338
+ "learning_rate": 2.3523720621537863e-05,
1339
+ "loss": 0.0,
1340
+ "step": 9250
1341
+ },
1342
+ {
1343
+ "epoch": 3.0886748588508803,
1344
+ "grad_norm": 0.0004945154651068151,
1345
+ "learning_rate": 2.3424271392563748e-05,
1346
+ "loss": 0.0,
1347
+ "step": 9300
1348
+ },
1349
+ {
1350
+ "epoch": 3.1052806376619064,
1351
+ "grad_norm": 0.00045879624667577446,
1352
+ "learning_rate": 2.332482216358963e-05,
1353
+ "loss": 0.0,
1354
+ "step": 9350
1355
+ },
1356
+ {
1357
+ "epoch": 3.1218864164729325,
1358
+ "grad_norm": 0.0004470401909202337,
1359
+ "learning_rate": 2.3225372934615513e-05,
1360
+ "loss": 0.0,
1361
+ "step": 9400
1362
+ },
1363
+ {
1364
+ "epoch": 3.1384921952839586,
1365
+ "grad_norm": 0.00023802775831427425,
1366
+ "learning_rate": 2.312592370564139e-05,
1367
+ "loss": 0.0,
1368
+ "step": 9450
1369
+ },
1370
+ {
1371
+ "epoch": 3.155097974094985,
1372
+ "grad_norm": 0.00043214907054789364,
1373
+ "learning_rate": 2.3026474476667275e-05,
1374
+ "loss": 0.0,
1375
+ "step": 9500
1376
+ },
1377
+ {
1378
+ "epoch": 3.1717037529060113,
1379
+ "grad_norm": 0.00024678235058672726,
1380
+ "learning_rate": 2.292702524769316e-05,
1381
+ "loss": 0.0007,
1382
+ "step": 9550
1383
+ },
1384
+ {
1385
+ "epoch": 3.1883095317170373,
1386
+ "grad_norm": 0.0003429341595619917,
1387
+ "learning_rate": 2.282757601871904e-05,
1388
+ "loss": 0.0,
1389
+ "step": 9600
1390
+ },
1391
+ {
1392
+ "epoch": 3.204915310528064,
1393
+ "grad_norm": 0.00019195489585399628,
1394
+ "learning_rate": 2.2728126789744925e-05,
1395
+ "loss": 0.0001,
1396
+ "step": 9650
1397
+ },
1398
+ {
1399
+ "epoch": 3.22152108933909,
1400
+ "grad_norm": 0.00021821403061039746,
1401
+ "learning_rate": 2.2628677560770803e-05,
1402
+ "loss": 0.0,
1403
+ "step": 9700
1404
+ },
1405
+ {
1406
+ "epoch": 3.238126868150116,
1407
+ "grad_norm": 0.0004384716448839754,
1408
+ "learning_rate": 2.2529228331796688e-05,
1409
+ "loss": 0.0,
1410
+ "step": 9750
1411
+ },
1412
+ {
1413
+ "epoch": 3.2547326469611426,
1414
+ "grad_norm": 0.00017058267258107662,
1415
+ "learning_rate": 2.242977910282257e-05,
1416
+ "loss": 0.0,
1417
+ "step": 9800
1418
+ },
1419
+ {
1420
+ "epoch": 3.2713384257721687,
1421
+ "grad_norm": 0.0007424860959872603,
1422
+ "learning_rate": 2.2330329873848453e-05,
1423
+ "loss": 0.0162,
1424
+ "step": 9850
1425
+ },
1426
+ {
1427
+ "epoch": 3.287944204583195,
1428
+ "grad_norm": 0.0007187157752923667,
1429
+ "learning_rate": 2.223088064487433e-05,
1430
+ "loss": 0.0114,
1431
+ "step": 9900
1432
+ },
1433
+ {
1434
+ "epoch": 3.3045499833942213,
1435
+ "grad_norm": 0.0006962522747926414,
1436
+ "learning_rate": 2.2131431415900215e-05,
1437
+ "loss": 0.003,
1438
+ "step": 9950
1439
+ },
1440
+ {
1441
+ "epoch": 3.3211557622052474,
1442
+ "grad_norm": 0.0007819280726835132,
1443
+ "learning_rate": 2.2031982186926096e-05,
1444
+ "loss": 0.005,
1445
+ "step": 10000
1446
+ },
1447
+ {
1448
+ "epoch": 3.3377615410162735,
1449
+ "grad_norm": 0.00020613332162611187,
1450
+ "learning_rate": 2.1932532957951977e-05,
1451
+ "loss": 0.0002,
1452
+ "step": 10050
1453
+ },
1454
+ {
1455
+ "epoch": 3.3543673198273,
1456
+ "grad_norm": 0.000297825870802626,
1457
+ "learning_rate": 2.183308372897786e-05,
1458
+ "loss": 0.0,
1459
+ "step": 10100
1460
+ },
1461
+ {
1462
+ "epoch": 3.370973098638326,
1463
+ "grad_norm": 7.085188553901389e-05,
1464
+ "learning_rate": 2.1733634500003743e-05,
1465
+ "loss": 0.0002,
1466
+ "step": 10150
1467
+ },
1468
+ {
1469
+ "epoch": 3.3875788774493523,
1470
+ "grad_norm": 0.11117817461490631,
1471
+ "learning_rate": 2.1634185271029624e-05,
1472
+ "loss": 0.0196,
1473
+ "step": 10200
1474
+ },
1475
+ {
1476
+ "epoch": 3.404184656260379,
1477
+ "grad_norm": 0.0007934737950563431,
1478
+ "learning_rate": 2.1534736042055505e-05,
1479
+ "loss": 0.0063,
1480
+ "step": 10250
1481
+ },
1482
+ {
1483
+ "epoch": 3.420790435071405,
1484
+ "grad_norm": 0.0008382880478166044,
1485
+ "learning_rate": 2.1435286813081386e-05,
1486
+ "loss": 0.0003,
1487
+ "step": 10300
1488
+ },
1489
+ {
1490
+ "epoch": 3.437396213882431,
1491
+ "grad_norm": 0.00943814031779766,
1492
+ "learning_rate": 2.133583758410727e-05,
1493
+ "loss": 0.0229,
1494
+ "step": 10350
1495
+ },
1496
+ {
1497
+ "epoch": 3.454001992693457,
1498
+ "grad_norm": 0.0025139478966593742,
1499
+ "learning_rate": 2.1236388355133155e-05,
1500
+ "loss": 0.0135,
1501
+ "step": 10400
1502
+ },
1503
+ {
1504
+ "epoch": 3.4706077715044836,
1505
+ "grad_norm": 0.0012301671085879207,
1506
+ "learning_rate": 2.1136939126159033e-05,
1507
+ "loss": 0.0112,
1508
+ "step": 10450
1509
+ },
1510
+ {
1511
+ "epoch": 3.4872135503155097,
1512
+ "grad_norm": 0.0006923259934410453,
1513
+ "learning_rate": 2.1037489897184917e-05,
1514
+ "loss": 0.0023,
1515
+ "step": 10500
1516
+ },
1517
+ {
1518
+ "epoch": 3.503819329126536,
1519
+ "grad_norm": 0.00143510103225708,
1520
+ "learning_rate": 2.09380406682108e-05,
1521
+ "loss": 0.0,
1522
+ "step": 10550
1523
+ },
1524
+ {
1525
+ "epoch": 3.5204251079375624,
1526
+ "grad_norm": 0.0003266233834438026,
1527
+ "learning_rate": 2.083859143923668e-05,
1528
+ "loss": 0.0001,
1529
+ "step": 10600
1530
+ },
1531
+ {
1532
+ "epoch": 3.5370308867485885,
1533
+ "grad_norm": 0.0002165662735933438,
1534
+ "learning_rate": 2.073914221026256e-05,
1535
+ "loss": 0.0001,
1536
+ "step": 10650
1537
+ },
1538
+ {
1539
+ "epoch": 3.5536366655596145,
1540
+ "grad_norm": 0.000500050897244364,
1541
+ "learning_rate": 2.0639692981288445e-05,
1542
+ "loss": 0.0053,
1543
+ "step": 10700
1544
+ },
1545
+ {
1546
+ "epoch": 3.570242444370641,
1547
+ "grad_norm": 0.00033526355400681496,
1548
+ "learning_rate": 2.0540243752314326e-05,
1549
+ "loss": 0.0001,
1550
+ "step": 10750
1551
+ },
1552
+ {
1553
+ "epoch": 3.586848223181667,
1554
+ "grad_norm": 0.000240692708757706,
1555
+ "learning_rate": 2.044079452334021e-05,
1556
+ "loss": 0.0,
1557
+ "step": 10800
1558
+ },
1559
+ {
1560
+ "epoch": 3.6034540019926933,
1561
+ "grad_norm": 0.0011404575780034065,
1562
+ "learning_rate": 2.034134529436609e-05,
1563
+ "loss": 0.0065,
1564
+ "step": 10850
1565
+ },
1566
+ {
1567
+ "epoch": 3.62005978080372,
1568
+ "grad_norm": 0.0008779458003118634,
1569
+ "learning_rate": 2.0241896065391973e-05,
1570
+ "loss": 0.0,
1571
+ "step": 10900
1572
+ },
1573
+ {
1574
+ "epoch": 3.636665559614746,
1575
+ "grad_norm": 0.000214042782317847,
1576
+ "learning_rate": 2.0142446836417854e-05,
1577
+ "loss": 0.0001,
1578
+ "step": 10950
1579
+ },
1580
+ {
1581
+ "epoch": 3.653271338425772,
1582
+ "grad_norm": 0.001132681965827942,
1583
+ "learning_rate": 2.0042997607443738e-05,
1584
+ "loss": 0.0001,
1585
+ "step": 11000
1586
+ },
1587
+ {
1588
+ "epoch": 3.6698771172367985,
1589
+ "grad_norm": 9.505786874797195e-05,
1590
+ "learning_rate": 1.994354837846962e-05,
1591
+ "loss": 0.0,
1592
+ "step": 11050
1593
+ },
1594
+ {
1595
+ "epoch": 3.6864828960478246,
1596
+ "grad_norm": 0.0014238933799788356,
1597
+ "learning_rate": 1.98440991494955e-05,
1598
+ "loss": 0.0108,
1599
+ "step": 11100
1600
+ },
1601
+ {
1602
+ "epoch": 3.7030886748588507,
1603
+ "grad_norm": 0.0021179679315537214,
1604
+ "learning_rate": 1.974464992052138e-05,
1605
+ "loss": 0.0004,
1606
+ "step": 11150
1607
+ },
1608
+ {
1609
+ "epoch": 3.7196944536698773,
1610
+ "grad_norm": 0.0012716053752228618,
1611
+ "learning_rate": 1.9645200691547266e-05,
1612
+ "loss": 0.0001,
1613
+ "step": 11200
1614
+ },
1615
+ {
1616
+ "epoch": 3.7363002324809034,
1617
+ "grad_norm": 0.0003722730907611549,
1618
+ "learning_rate": 1.9545751462573147e-05,
1619
+ "loss": 0.0001,
1620
+ "step": 11250
1621
+ },
1622
+ {
1623
+ "epoch": 3.7529060112919295,
1624
+ "grad_norm": 0.0004279191780369729,
1625
+ "learning_rate": 1.9446302233599028e-05,
1626
+ "loss": 0.0,
1627
+ "step": 11300
1628
+ },
1629
+ {
1630
+ "epoch": 3.769511790102956,
1631
+ "grad_norm": 0.00044420413905754685,
1632
+ "learning_rate": 1.934685300462491e-05,
1633
+ "loss": 0.0,
1634
+ "step": 11350
1635
+ },
1636
+ {
1637
+ "epoch": 3.786117568913982,
1638
+ "grad_norm": 0.0006720417295582592,
1639
+ "learning_rate": 1.9247403775650794e-05,
1640
+ "loss": 0.0001,
1641
+ "step": 11400
1642
+ },
1643
+ {
1644
+ "epoch": 3.802723347725008,
1645
+ "grad_norm": 0.00018277636263519526,
1646
+ "learning_rate": 1.9147954546676675e-05,
1647
+ "loss": 0.0,
1648
+ "step": 11450
1649
+ },
1650
+ {
1651
+ "epoch": 3.8193291265360347,
1652
+ "grad_norm": 0.0033536076080054045,
1653
+ "learning_rate": 1.904850531770256e-05,
1654
+ "loss": 0.0,
1655
+ "step": 11500
1656
+ },
1657
+ {
1658
+ "epoch": 3.835934905347061,
1659
+ "grad_norm": 0.00032338136225007474,
1660
+ "learning_rate": 1.894905608872844e-05,
1661
+ "loss": 0.0,
1662
+ "step": 11550
1663
+ },
1664
+ {
1665
+ "epoch": 3.852540684158087,
1666
+ "grad_norm": 8.492634515278041e-05,
1667
+ "learning_rate": 1.884960685975432e-05,
1668
+ "loss": 0.0,
1669
+ "step": 11600
1670
+ },
1671
+ {
1672
+ "epoch": 3.8691464629691135,
1673
+ "grad_norm": 0.00010078576451633126,
1674
+ "learning_rate": 1.8750157630780202e-05,
1675
+ "loss": 0.0,
1676
+ "step": 11650
1677
+ },
1678
+ {
1679
+ "epoch": 3.8857522417801396,
1680
+ "grad_norm": 0.0001655027299420908,
1681
+ "learning_rate": 1.8650708401806087e-05,
1682
+ "loss": 0.0,
1683
+ "step": 11700
1684
+ },
1685
+ {
1686
+ "epoch": 3.9023580205911657,
1687
+ "grad_norm": 0.00012061335291946307,
1688
+ "learning_rate": 1.8551259172831968e-05,
1689
+ "loss": 0.0,
1690
+ "step": 11750
1691
+ },
1692
+ {
1693
+ "epoch": 3.918963799402192,
1694
+ "grad_norm": 0.00044828641694039106,
1695
+ "learning_rate": 1.845180994385785e-05,
1696
+ "loss": 0.0,
1697
+ "step": 11800
1698
+ },
1699
+ {
1700
+ "epoch": 3.9355695782132183,
1701
+ "grad_norm": 5.701217378373258e-05,
1702
+ "learning_rate": 1.835236071488373e-05,
1703
+ "loss": 0.0,
1704
+ "step": 11850
1705
+ },
1706
+ {
1707
+ "epoch": 3.9521753570242444,
1708
+ "grad_norm": 5.7118391850963235e-05,
1709
+ "learning_rate": 1.8252911485909615e-05,
1710
+ "loss": 0.0,
1711
+ "step": 11900
1712
+ },
1713
+ {
1714
+ "epoch": 3.968781135835271,
1715
+ "grad_norm": 0.0007707001641392708,
1716
+ "learning_rate": 1.8153462256935496e-05,
1717
+ "loss": 0.0001,
1718
+ "step": 11950
1719
+ },
1720
+ {
1721
+ "epoch": 3.985386914646297,
1722
+ "grad_norm": 6.149195542093366e-05,
1723
+ "learning_rate": 1.8054013027961377e-05,
1724
+ "loss": 0.0,
1725
+ "step": 12000
1726
+ },
1727
+ {
1728
+ "epoch": 4.0,
1729
+ "eval_accuracy": 0.9973759541984732,
1730
+ "eval_f1": 0.9973482118175944,
1731
+ "eval_loss": 0.03656009957194328,
1732
+ "eval_precision": 0.9973582002448674,
1733
+ "eval_recall": 0.9973759541984732,
1734
+ "eval_runtime": 67.5118,
1735
+ "eval_samples_per_second": 124.186,
1736
+ "eval_steps_per_second": 7.762,
1737
+ "step": 12044
1738
+ },
1739
+ {
1740
+ "epoch": 4.001992693457323,
1741
+ "grad_norm": 5.8200610510539263e-05,
1742
+ "learning_rate": 1.7954563798987258e-05,
1743
+ "loss": 0.0,
1744
+ "step": 12050
1745
+ },
1746
+ {
1747
+ "epoch": 4.01859847226835,
1748
+ "grad_norm": 2.329714880033862e-05,
1749
+ "learning_rate": 1.7855114570013142e-05,
1750
+ "loss": 0.0,
1751
+ "step": 12100
1752
+ },
1753
+ {
1754
+ "epoch": 4.035204251079375,
1755
+ "grad_norm": 0.00018428650218993425,
1756
+ "learning_rate": 1.7755665341039023e-05,
1757
+ "loss": 0.0,
1758
+ "step": 12150
1759
+ },
1760
+ {
1761
+ "epoch": 4.051810029890402,
1762
+ "grad_norm": 3.634762470028363e-05,
1763
+ "learning_rate": 1.7656216112064904e-05,
1764
+ "loss": 0.0047,
1765
+ "step": 12200
1766
+ },
1767
+ {
1768
+ "epoch": 4.068415808701428,
1769
+ "grad_norm": 0.016488956287503242,
1770
+ "learning_rate": 1.7556766883090785e-05,
1771
+ "loss": 0.0185,
1772
+ "step": 12250
1773
+ },
1774
+ {
1775
+ "epoch": 4.085021587512454,
1776
+ "grad_norm": 0.0005068861646577716,
1777
+ "learning_rate": 1.745731765411667e-05,
1778
+ "loss": 0.0002,
1779
+ "step": 12300
1780
+ },
1781
+ {
1782
+ "epoch": 4.101627366323481,
1783
+ "grad_norm": 0.0002266648516524583,
1784
+ "learning_rate": 1.735786842514255e-05,
1785
+ "loss": 0.0001,
1786
+ "step": 12350
1787
+ },
1788
+ {
1789
+ "epoch": 4.118233145134507,
1790
+ "grad_norm": 0.000158867915160954,
1791
+ "learning_rate": 1.7258419196168435e-05,
1792
+ "loss": 0.0001,
1793
+ "step": 12400
1794
+ },
1795
+ {
1796
+ "epoch": 4.134838923945533,
1797
+ "grad_norm": 0.00020392602891661227,
1798
+ "learning_rate": 1.7158969967194317e-05,
1799
+ "loss": 0.0001,
1800
+ "step": 12450
1801
+ },
1802
+ {
1803
+ "epoch": 4.151444702756559,
1804
+ "grad_norm": 7.127954449970275e-05,
1805
+ "learning_rate": 1.7059520738220198e-05,
1806
+ "loss": 0.0,
1807
+ "step": 12500
1808
+ },
1809
+ {
1810
+ "epoch": 4.168050481567586,
1811
+ "grad_norm": 0.0027118742000311613,
1812
+ "learning_rate": 1.696007150924608e-05,
1813
+ "loss": 0.0005,
1814
+ "step": 12550
1815
+ },
1816
+ {
1817
+ "epoch": 4.1846562603786115,
1818
+ "grad_norm": 0.00021424326405394822,
1819
+ "learning_rate": 1.6860622280271963e-05,
1820
+ "loss": 0.0001,
1821
+ "step": 12600
1822
+ },
1823
+ {
1824
+ "epoch": 4.201262039189638,
1825
+ "grad_norm": 0.003089814679697156,
1826
+ "learning_rate": 1.6761173051297844e-05,
1827
+ "loss": 0.0136,
1828
+ "step": 12650
1829
+ },
1830
+ {
1831
+ "epoch": 4.217867818000665,
1832
+ "grad_norm": 0.0032255896367132664,
1833
+ "learning_rate": 1.6661723822323725e-05,
1834
+ "loss": 0.0001,
1835
+ "step": 12700
1836
+ },
1837
+ {
1838
+ "epoch": 4.23447359681169,
1839
+ "grad_norm": 0.00042820069938898087,
1840
+ "learning_rate": 1.6562274593349606e-05,
1841
+ "loss": 0.0,
1842
+ "step": 12750
1843
+ },
1844
+ {
1845
+ "epoch": 4.251079375622717,
1846
+ "grad_norm": 0.0002473096828907728,
1847
+ "learning_rate": 1.646282536437549e-05,
1848
+ "loss": 0.0001,
1849
+ "step": 12800
1850
+ },
1851
+ {
1852
+ "epoch": 4.267685154433743,
1853
+ "grad_norm": 0.00025876634754240513,
1854
+ "learning_rate": 1.6363376135401372e-05,
1855
+ "loss": 0.0069,
1856
+ "step": 12850
1857
+ },
1858
+ {
1859
+ "epoch": 4.284290933244769,
1860
+ "grad_norm": 0.00011665420606732368,
1861
+ "learning_rate": 1.6263926906427253e-05,
1862
+ "loss": 0.0,
1863
+ "step": 12900
1864
+ },
1865
+ {
1866
+ "epoch": 4.3008967120557955,
1867
+ "grad_norm": 9.87951279967092e-05,
1868
+ "learning_rate": 1.6164477677453134e-05,
1869
+ "loss": 0.0002,
1870
+ "step": 12950
1871
+ },
1872
+ {
1873
+ "epoch": 4.317502490866822,
1874
+ "grad_norm": 0.0003025582409463823,
1875
+ "learning_rate": 1.606502844847902e-05,
1876
+ "loss": 0.0204,
1877
+ "step": 13000
1878
+ },
1879
+ {
1880
+ "epoch": 4.334108269677848,
1881
+ "grad_norm": 0.002073385054245591,
1882
+ "learning_rate": 1.59655792195049e-05,
1883
+ "loss": 0.0209,
1884
+ "step": 13050
1885
+ },
1886
+ {
1887
+ "epoch": 4.350714048488874,
1888
+ "grad_norm": 0.001544980681501329,
1889
+ "learning_rate": 1.5866129990530784e-05,
1890
+ "loss": 0.0014,
1891
+ "step": 13100
1892
+ },
1893
+ {
1894
+ "epoch": 4.367319827299901,
1895
+ "grad_norm": 0.0011374803725630045,
1896
+ "learning_rate": 1.5766680761556665e-05,
1897
+ "loss": 0.0108,
1898
+ "step": 13150
1899
+ },
1900
+ {
1901
+ "epoch": 4.383925606110926,
1902
+ "grad_norm": 0.0015389297623187304,
1903
+ "learning_rate": 1.5667231532582546e-05,
1904
+ "loss": 0.0001,
1905
+ "step": 13200
1906
+ },
1907
+ {
1908
+ "epoch": 4.400531384921953,
1909
+ "grad_norm": 0.0011413008905947208,
1910
+ "learning_rate": 1.5567782303608427e-05,
1911
+ "loss": 0.0001,
1912
+ "step": 13250
1913
+ },
1914
+ {
1915
+ "epoch": 4.4171371637329795,
1916
+ "grad_norm": 0.001274819835089147,
1917
+ "learning_rate": 1.5468333074634312e-05,
1918
+ "loss": 0.0001,
1919
+ "step": 13300
1920
+ },
1921
+ {
1922
+ "epoch": 4.433742942544005,
1923
+ "grad_norm": 0.0003733636694960296,
1924
+ "learning_rate": 1.5368883845660193e-05,
1925
+ "loss": 0.0001,
1926
+ "step": 13350
1927
+ },
1928
+ {
1929
+ "epoch": 4.450348721355032,
1930
+ "grad_norm": 0.00023891785531304777,
1931
+ "learning_rate": 1.5269434616686074e-05,
1932
+ "loss": 0.0,
1933
+ "step": 13400
1934
+ },
1935
+ {
1936
+ "epoch": 4.466954500166057,
1937
+ "grad_norm": 0.0003116075531579554,
1938
+ "learning_rate": 1.5169985387711957e-05,
1939
+ "loss": 0.0,
1940
+ "step": 13450
1941
+ },
1942
+ {
1943
+ "epoch": 4.483560278977084,
1944
+ "grad_norm": 0.0002987791958730668,
1945
+ "learning_rate": 1.5070536158737838e-05,
1946
+ "loss": 0.0007,
1947
+ "step": 13500
1948
+ },
1949
+ {
1950
+ "epoch": 4.50016605778811,
1951
+ "grad_norm": 0.0006362134590744972,
1952
+ "learning_rate": 1.497108692976372e-05,
1953
+ "loss": 0.0115,
1954
+ "step": 13550
1955
+ },
1956
+ {
1957
+ "epoch": 4.516771836599137,
1958
+ "grad_norm": 0.0006330082542262971,
1959
+ "learning_rate": 1.4871637700789602e-05,
1960
+ "loss": 0.0093,
1961
+ "step": 13600
1962
+ },
1963
+ {
1964
+ "epoch": 4.533377615410163,
1965
+ "grad_norm": 0.0014981752028688788,
1966
+ "learning_rate": 1.4772188471815484e-05,
1967
+ "loss": 0.0076,
1968
+ "step": 13650
1969
+ },
1970
+ {
1971
+ "epoch": 4.549983394221189,
1972
+ "grad_norm": 0.0010505125392228365,
1973
+ "learning_rate": 1.4672739242841365e-05,
1974
+ "loss": 0.0001,
1975
+ "step": 13700
1976
+ },
1977
+ {
1978
+ "epoch": 4.566589173032215,
1979
+ "grad_norm": 0.0011327213142067194,
1980
+ "learning_rate": 1.4573290013867248e-05,
1981
+ "loss": 0.0,
1982
+ "step": 13750
1983
+ },
1984
+ {
1985
+ "epoch": 4.583194951843241,
1986
+ "grad_norm": 0.0005059707909822464,
1987
+ "learning_rate": 1.447384078489313e-05,
1988
+ "loss": 0.0,
1989
+ "step": 13800
1990
+ },
1991
+ {
1992
+ "epoch": 4.599800730654268,
1993
+ "grad_norm": 0.0011543643195182085,
1994
+ "learning_rate": 1.437439155591901e-05,
1995
+ "loss": 0.0,
1996
+ "step": 13850
1997
+ },
1998
+ {
1999
+ "epoch": 4.616406509465294,
2000
+ "grad_norm": 0.00020546128507703543,
2001
+ "learning_rate": 1.4274942326944895e-05,
2002
+ "loss": 0.0,
2003
+ "step": 13900
2004
+ },
2005
+ {
2006
+ "epoch": 4.63301228827632,
2007
+ "grad_norm": 0.0001845878578023985,
2008
+ "learning_rate": 1.4175493097970778e-05,
2009
+ "loss": 0.0,
2010
+ "step": 13950
2011
+ },
2012
+ {
2013
+ "epoch": 4.649618067087347,
2014
+ "grad_norm": 0.00035237689735367894,
2015
+ "learning_rate": 1.4076043868996659e-05,
2016
+ "loss": 0.0,
2017
+ "step": 14000
2018
+ },
2019
+ {
2020
+ "epoch": 4.666223845898372,
2021
+ "grad_norm": 0.00028683870914392173,
2022
+ "learning_rate": 1.3976594640022541e-05,
2023
+ "loss": 0.0,
2024
+ "step": 14050
2025
+ },
2026
+ {
2027
+ "epoch": 4.682829624709399,
2028
+ "grad_norm": 0.0003122540074400604,
2029
+ "learning_rate": 1.3877145411048423e-05,
2030
+ "loss": 0.0,
2031
+ "step": 14100
2032
+ },
2033
+ {
2034
+ "epoch": 4.699435403520425,
2035
+ "grad_norm": 0.00045101705472916365,
2036
+ "learning_rate": 1.3777696182074305e-05,
2037
+ "loss": 0.0002,
2038
+ "step": 14150
2039
+ },
2040
+ {
2041
+ "epoch": 4.716041182331451,
2042
+ "grad_norm": 0.0011838871287181973,
2043
+ "learning_rate": 1.3678246953100186e-05,
2044
+ "loss": 0.0,
2045
+ "step": 14200
2046
+ },
2047
+ {
2048
+ "epoch": 4.7326469611424775,
2049
+ "grad_norm": 0.00016272020002361387,
2050
+ "learning_rate": 1.3578797724126069e-05,
2051
+ "loss": 0.003,
2052
+ "step": 14250
2053
+ },
2054
+ {
2055
+ "epoch": 4.749252739953504,
2056
+ "grad_norm": 0.00011718720634235069,
2057
+ "learning_rate": 1.347934849515195e-05,
2058
+ "loss": 0.0,
2059
+ "step": 14300
2060
+ },
2061
+ {
2062
+ "epoch": 4.76585851876453,
2063
+ "grad_norm": 8.695230644661933e-05,
2064
+ "learning_rate": 1.3379899266177833e-05,
2065
+ "loss": 0.0,
2066
+ "step": 14350
2067
+ },
2068
+ {
2069
+ "epoch": 4.782464297575556,
2070
+ "grad_norm": 6.504805787699297e-05,
2071
+ "learning_rate": 1.3280450037203714e-05,
2072
+ "loss": 0.0,
2073
+ "step": 14400
2074
+ },
2075
+ {
2076
+ "epoch": 4.799070076386583,
2077
+ "grad_norm": 0.0013819790910929441,
2078
+ "learning_rate": 1.3181000808229597e-05,
2079
+ "loss": 0.0,
2080
+ "step": 14450
2081
+ },
2082
+ {
2083
+ "epoch": 4.815675855197608,
2084
+ "grad_norm": 8.097992395050824e-05,
2085
+ "learning_rate": 1.3081551579255478e-05,
2086
+ "loss": 0.0,
2087
+ "step": 14500
2088
+ },
2089
+ {
2090
+ "epoch": 4.832281634008635,
2091
+ "grad_norm": 6.052408934920095e-05,
2092
+ "learning_rate": 1.2982102350281359e-05,
2093
+ "loss": 0.0,
2094
+ "step": 14550
2095
+ },
2096
+ {
2097
+ "epoch": 4.8488874128196615,
2098
+ "grad_norm": 6.93105612299405e-05,
2099
+ "learning_rate": 1.2882653121307242e-05,
2100
+ "loss": 0.0,
2101
+ "step": 14600
2102
+ },
2103
+ {
2104
+ "epoch": 4.865493191630687,
2105
+ "grad_norm": 6.091671093599871e-05,
2106
+ "learning_rate": 1.2783203892333123e-05,
2107
+ "loss": 0.0,
2108
+ "step": 14650
2109
+ },
2110
+ {
2111
+ "epoch": 4.882098970441714,
2112
+ "grad_norm": 0.00023341408814303577,
2113
+ "learning_rate": 1.2683754663359007e-05,
2114
+ "loss": 0.0,
2115
+ "step": 14700
2116
+ },
2117
+ {
2118
+ "epoch": 4.89870474925274,
2119
+ "grad_norm": 4.871335477218963e-05,
2120
+ "learning_rate": 1.258430543438489e-05,
2121
+ "loss": 0.0,
2122
+ "step": 14750
2123
+ },
2124
+ {
2125
+ "epoch": 4.915310528063766,
2126
+ "grad_norm": 0.0001195693839690648,
2127
+ "learning_rate": 1.2484856205410771e-05,
2128
+ "loss": 0.0007,
2129
+ "step": 14800
2130
+ },
2131
+ {
2132
+ "epoch": 4.931916306874792,
2133
+ "grad_norm": 0.0006585444789379835,
2134
+ "learning_rate": 1.2385406976436654e-05,
2135
+ "loss": 0.0001,
2136
+ "step": 14850
2137
+ },
2138
+ {
2139
+ "epoch": 4.948522085685819,
2140
+ "grad_norm": 0.0009381878189742565,
2141
+ "learning_rate": 1.2285957747462535e-05,
2142
+ "loss": 0.0136,
2143
+ "step": 14900
2144
+ },
2145
+ {
2146
+ "epoch": 4.965127864496845,
2147
+ "grad_norm": 0.0007571703754365444,
2148
+ "learning_rate": 1.2186508518488418e-05,
2149
+ "loss": 0.0001,
2150
+ "step": 14950
2151
+ },
2152
+ {
2153
+ "epoch": 4.981733643307871,
2154
+ "grad_norm": 0.0002012668555835262,
2155
+ "learning_rate": 1.2087059289514299e-05,
2156
+ "loss": 0.0,
2157
+ "step": 15000
2158
+ },
2159
+ {
2160
+ "epoch": 4.998339422118898,
2161
+ "grad_norm": 0.0004453076981008053,
2162
+ "learning_rate": 1.1987610060540182e-05,
2163
+ "loss": 0.0011,
2164
+ "step": 15050
2165
+ },
2166
+ {
2167
+ "epoch": 5.0,
2168
+ "eval_accuracy": 0.997256679389313,
2169
+ "eval_f1": 0.9972255532951301,
2170
+ "eval_loss": 0.026163995265960693,
2171
+ "eval_precision": 0.997237804162915,
2172
+ "eval_recall": 0.997256679389313,
2173
+ "eval_runtime": 67.7679,
2174
+ "eval_samples_per_second": 123.716,
2175
+ "eval_steps_per_second": 7.732,
2176
+ "step": 15055
2177
+ }
2178
+ ],
2179
+ "logging_steps": 50,
2180
+ "max_steps": 21077,
2181
+ "num_input_tokens_seen": 0,
2182
+ "num_train_epochs": 7,
2183
+ "save_steps": 500,
2184
+ "stateful_callbacks": {
2185
+ "TrainerControl": {
2186
+ "args": {
2187
+ "should_epoch_stop": false,
2188
+ "should_evaluate": false,
2189
+ "should_log": false,
2190
+ "should_save": true,
2191
+ "should_training_stop": false
2192
+ },
2193
+ "attributes": {}
2194
+ }
2195
+ },
2196
+ "total_flos": 2.547123652512768e+17,
2197
+ "train_batch_size": 16,
2198
+ "trial_name": null,
2199
+ "trial_params": null
2200
+ }
trial-1/checkpoint-15055/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:535f8963c9933a98cbda6a7a60fb42b0daa9affc10889961579027cb42dfd7b9
3
+ size 5368
trial-2/checkpoint-9033/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-2/checkpoint-9033/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ceb5aea83d31becf51f33bad0101af19aa72e30825571029e2058839c2598942
3
+ size 1583351632
trial-2/checkpoint-9033/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d58ab59719b06a08e5012236adc8ead9857b2a6b939a9df2e7ad1fd2ea7f3856
3
+ size 3166813178
trial-2/checkpoint-9033/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:000f96e69b5b01f915aefc67b0e71455a9632e90db9f36791eb61370a67ffd58
3
+ size 14244
trial-2/checkpoint-9033/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef3fd896aed775d6359ff0434e2911a0c84bd73dffb2a8c635b32b0f9e87d938
3
+ size 1064
trial-2/checkpoint-9033/trainer_state.json ADDED
@@ -0,0 +1,1329 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.017316868528723717,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-2/checkpoint-9033",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 9033,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.016605778811026237,
13
+ "grad_norm": 21.98246192932129,
14
+ "learning_rate": 4.176603346793102e-05,
15
+ "loss": 0.3038,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.033211557622052475,
20
+ "grad_norm": 0.019962424412369728,
21
+ "learning_rate": 4.162685974761136e-05,
22
+ "loss": 0.1339,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.04981733643307871,
27
+ "grad_norm": 4.160129070281982,
28
+ "learning_rate": 4.1487686027291694e-05,
29
+ "loss": 0.1214,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.06642311524410495,
34
+ "grad_norm": 0.058199040591716766,
35
+ "learning_rate": 4.1348512306972037e-05,
36
+ "loss": 0.0991,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.08302889405513118,
41
+ "grad_norm": 4.1636199951171875,
42
+ "learning_rate": 4.120933858665237e-05,
43
+ "loss": 0.0483,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.09963467286615742,
48
+ "grad_norm": 0.009575674310326576,
49
+ "learning_rate": 4.107016486633271e-05,
50
+ "loss": 0.0655,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.11624045167718366,
55
+ "grad_norm": 0.005853955168277025,
56
+ "learning_rate": 4.0930991146013044e-05,
57
+ "loss": 0.0417,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.1328462304882099,
62
+ "grad_norm": 0.22517672181129456,
63
+ "learning_rate": 4.079181742569338e-05,
64
+ "loss": 0.0559,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.14945200929923613,
69
+ "grad_norm": 2.013570785522461,
70
+ "learning_rate": 4.0652643705373716e-05,
71
+ "loss": 0.0295,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.16605778811026237,
76
+ "grad_norm": 0.06311897933483124,
77
+ "learning_rate": 4.051346998505405e-05,
78
+ "loss": 0.0683,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.1826635669212886,
83
+ "grad_norm": 0.5382233262062073,
84
+ "learning_rate": 4.037429626473439e-05,
85
+ "loss": 0.03,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.19926934573231483,
90
+ "grad_norm": 1.0302859544754028,
91
+ "learning_rate": 4.0235122544414723e-05,
92
+ "loss": 0.0502,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.2158751245433411,
97
+ "grad_norm": 5.7028913497924805,
98
+ "learning_rate": 4.009594882409506e-05,
99
+ "loss": 0.0429,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.23248090335436733,
104
+ "grad_norm": 0.05518786981701851,
105
+ "learning_rate": 3.9956775103775395e-05,
106
+ "loss": 0.0384,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.24908668216539356,
111
+ "grad_norm": 0.0032245127949863672,
112
+ "learning_rate": 3.981760138345573e-05,
113
+ "loss": 0.048,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.2656924609764198,
118
+ "grad_norm": 0.12157031893730164,
119
+ "learning_rate": 3.967842766313607e-05,
120
+ "loss": 0.0739,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.282298239787446,
125
+ "grad_norm": 1.955265998840332,
126
+ "learning_rate": 3.95392539428164e-05,
127
+ "loss": 0.0607,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.29890401859847227,
132
+ "grad_norm": 0.03681463748216629,
133
+ "learning_rate": 3.9400080222496745e-05,
134
+ "loss": 0.0469,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.3155097974094985,
139
+ "grad_norm": 0.00883490964770317,
140
+ "learning_rate": 3.926090650217708e-05,
141
+ "loss": 0.0251,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.33211557622052473,
146
+ "grad_norm": 0.017260171473026276,
147
+ "learning_rate": 3.912173278185742e-05,
148
+ "loss": 0.0274,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.348721355031551,
153
+ "grad_norm": 0.0017717696027830243,
154
+ "learning_rate": 3.898255906153775e-05,
155
+ "loss": 0.0603,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.3653271338425772,
160
+ "grad_norm": 0.03903853893280029,
161
+ "learning_rate": 3.884338534121809e-05,
162
+ "loss": 0.0378,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.38193291265360346,
167
+ "grad_norm": 0.002946289489045739,
168
+ "learning_rate": 3.8704211620898425e-05,
169
+ "loss": 0.0362,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.39853869146462967,
174
+ "grad_norm": 0.013737122528254986,
175
+ "learning_rate": 3.856503790057876e-05,
176
+ "loss": 0.048,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.41514447027565593,
181
+ "grad_norm": 0.22858883440494537,
182
+ "learning_rate": 3.84258641802591e-05,
183
+ "loss": 0.0195,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.4317502490866822,
188
+ "grad_norm": 0.605839729309082,
189
+ "learning_rate": 3.828669045993944e-05,
190
+ "loss": 0.0422,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.4483560278977084,
195
+ "grad_norm": 0.045106545090675354,
196
+ "learning_rate": 3.8147516739619775e-05,
197
+ "loss": 0.0303,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.46496180670873466,
202
+ "grad_norm": 0.0033422100823372602,
203
+ "learning_rate": 3.800834301930011e-05,
204
+ "loss": 0.0109,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.48156758551976087,
209
+ "grad_norm": 0.04552963748574257,
210
+ "learning_rate": 3.786916929898045e-05,
211
+ "loss": 0.02,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.4981733643307871,
216
+ "grad_norm": 0.0007607729057781398,
217
+ "learning_rate": 3.7729995578660776e-05,
218
+ "loss": 0.0283,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 0.5147791431418134,
223
+ "grad_norm": 0.011046077124774456,
224
+ "learning_rate": 3.759082185834111e-05,
225
+ "loss": 0.0249,
226
+ "step": 1550
227
+ },
228
+ {
229
+ "epoch": 0.5313849219528396,
230
+ "grad_norm": 0.1515175700187683,
231
+ "learning_rate": 3.7451648138021454e-05,
232
+ "loss": 0.0128,
233
+ "step": 1600
234
+ },
235
+ {
236
+ "epoch": 0.5479907007638658,
237
+ "grad_norm": 0.0051905689761042595,
238
+ "learning_rate": 3.731247441770179e-05,
239
+ "loss": 0.0235,
240
+ "step": 1650
241
+ },
242
+ {
243
+ "epoch": 0.564596479574892,
244
+ "grad_norm": 0.03131992742419243,
245
+ "learning_rate": 3.7173300697382126e-05,
246
+ "loss": 0.0017,
247
+ "step": 1700
248
+ },
249
+ {
250
+ "epoch": 0.5812022583859183,
251
+ "grad_norm": 50.02948760986328,
252
+ "learning_rate": 3.703412697706246e-05,
253
+ "loss": 0.0119,
254
+ "step": 1750
255
+ },
256
+ {
257
+ "epoch": 0.5978080371969445,
258
+ "grad_norm": 0.017883650958538055,
259
+ "learning_rate": 3.68949532567428e-05,
260
+ "loss": 0.0374,
261
+ "step": 1800
262
+ },
263
+ {
264
+ "epoch": 0.6144138160079707,
265
+ "grad_norm": 0.002184309996664524,
266
+ "learning_rate": 3.6755779536423134e-05,
267
+ "loss": 0.0258,
268
+ "step": 1850
269
+ },
270
+ {
271
+ "epoch": 0.631019594818997,
272
+ "grad_norm": 0.009681240655481815,
273
+ "learning_rate": 3.661660581610347e-05,
274
+ "loss": 0.0246,
275
+ "step": 1900
276
+ },
277
+ {
278
+ "epoch": 0.6476253736300233,
279
+ "grad_norm": 0.0064963954500854015,
280
+ "learning_rate": 3.647743209578381e-05,
281
+ "loss": 0.0076,
282
+ "step": 1950
283
+ },
284
+ {
285
+ "epoch": 0.6642311524410495,
286
+ "grad_norm": 0.0018849828047677875,
287
+ "learning_rate": 3.633825837546415e-05,
288
+ "loss": 0.0146,
289
+ "step": 2000
290
+ },
291
+ {
292
+ "epoch": 0.6808369312520757,
293
+ "grad_norm": 0.0013440287439152598,
294
+ "learning_rate": 3.6199084655144484e-05,
295
+ "loss": 0.0042,
296
+ "step": 2050
297
+ },
298
+ {
299
+ "epoch": 0.697442710063102,
300
+ "grad_norm": 0.0016228831373155117,
301
+ "learning_rate": 3.605991093482482e-05,
302
+ "loss": 0.0136,
303
+ "step": 2100
304
+ },
305
+ {
306
+ "epoch": 0.7140484888741282,
307
+ "grad_norm": 0.0009898327989503741,
308
+ "learning_rate": 3.5920737214505156e-05,
309
+ "loss": 0.0187,
310
+ "step": 2150
311
+ },
312
+ {
313
+ "epoch": 0.7306542676851544,
314
+ "grad_norm": 0.001361815258860588,
315
+ "learning_rate": 3.578156349418549e-05,
316
+ "loss": 0.0167,
317
+ "step": 2200
318
+ },
319
+ {
320
+ "epoch": 0.7472600464961807,
321
+ "grad_norm": 0.004233693704009056,
322
+ "learning_rate": 3.564238977386583e-05,
323
+ "loss": 0.0142,
324
+ "step": 2250
325
+ },
326
+ {
327
+ "epoch": 0.7638658253072069,
328
+ "grad_norm": 0.002790838712826371,
329
+ "learning_rate": 3.550321605354616e-05,
330
+ "loss": 0.0235,
331
+ "step": 2300
332
+ },
333
+ {
334
+ "epoch": 0.7804716041182331,
335
+ "grad_norm": 0.005341578274965286,
336
+ "learning_rate": 3.53640423332265e-05,
337
+ "loss": 0.002,
338
+ "step": 2350
339
+ },
340
+ {
341
+ "epoch": 0.7970773829292593,
342
+ "grad_norm": 0.011184672825038433,
343
+ "learning_rate": 3.5224868612906835e-05,
344
+ "loss": 0.0325,
345
+ "step": 2400
346
+ },
347
+ {
348
+ "epoch": 0.8136831617402857,
349
+ "grad_norm": 0.001378358923830092,
350
+ "learning_rate": 3.508569489258717e-05,
351
+ "loss": 0.0012,
352
+ "step": 2450
353
+ },
354
+ {
355
+ "epoch": 0.8302889405513119,
356
+ "grad_norm": 0.0006328076124191284,
357
+ "learning_rate": 3.4946521172267507e-05,
358
+ "loss": 0.0028,
359
+ "step": 2500
360
+ },
361
+ {
362
+ "epoch": 0.8468947193623381,
363
+ "grad_norm": 0.003556261071935296,
364
+ "learning_rate": 3.480734745194784e-05,
365
+ "loss": 0.0103,
366
+ "step": 2550
367
+ },
368
+ {
369
+ "epoch": 0.8635004981733644,
370
+ "grad_norm": 0.09675312787294388,
371
+ "learning_rate": 3.466817373162818e-05,
372
+ "loss": 0.0154,
373
+ "step": 2600
374
+ },
375
+ {
376
+ "epoch": 0.8801062769843906,
377
+ "grad_norm": 0.0008714852156117558,
378
+ "learning_rate": 3.452900001130852e-05,
379
+ "loss": 0.0098,
380
+ "step": 2650
381
+ },
382
+ {
383
+ "epoch": 0.8967120557954168,
384
+ "grad_norm": 0.00011318879842292517,
385
+ "learning_rate": 3.438982629098886e-05,
386
+ "loss": 0.0007,
387
+ "step": 2700
388
+ },
389
+ {
390
+ "epoch": 0.913317834606443,
391
+ "grad_norm": 0.0007046264945529401,
392
+ "learning_rate": 3.425065257066919e-05,
393
+ "loss": 0.0163,
394
+ "step": 2750
395
+ },
396
+ {
397
+ "epoch": 0.9299236134174693,
398
+ "grad_norm": 0.0003467966744210571,
399
+ "learning_rate": 3.411147885034953e-05,
400
+ "loss": 0.0087,
401
+ "step": 2800
402
+ },
403
+ {
404
+ "epoch": 0.9465293922284955,
405
+ "grad_norm": 0.002381992759183049,
406
+ "learning_rate": 3.3972305130029864e-05,
407
+ "loss": 0.025,
408
+ "step": 2850
409
+ },
410
+ {
411
+ "epoch": 0.9631351710395217,
412
+ "grad_norm": 3.5343263149261475,
413
+ "learning_rate": 3.38331314097102e-05,
414
+ "loss": 0.0066,
415
+ "step": 2900
416
+ },
417
+ {
418
+ "epoch": 0.9797409498505479,
419
+ "grad_norm": 1.6730190515518188,
420
+ "learning_rate": 3.369395768939054e-05,
421
+ "loss": 0.028,
422
+ "step": 2950
423
+ },
424
+ {
425
+ "epoch": 0.9963467286615743,
426
+ "grad_norm": 0.006649247836321592,
427
+ "learning_rate": 3.355478396907088e-05,
428
+ "loss": 0.001,
429
+ "step": 3000
430
+ },
431
+ {
432
+ "epoch": 1.0,
433
+ "eval_accuracy": 0.9877146946564885,
434
+ "eval_f1": 0.9883925037796503,
435
+ "eval_loss": 0.056414589285850525,
436
+ "eval_precision": 0.9900276560214059,
437
+ "eval_recall": 0.9877146946564885,
438
+ "eval_runtime": 66.3519,
439
+ "eval_samples_per_second": 126.357,
440
+ "eval_steps_per_second": 7.897,
441
+ "step": 3011
442
+ },
443
+ {
444
+ "epoch": 1.0129525074726005,
445
+ "grad_norm": 0.03574318438768387,
446
+ "learning_rate": 3.3415610248751215e-05,
447
+ "loss": 0.0166,
448
+ "step": 3050
449
+ },
450
+ {
451
+ "epoch": 1.0295582862836268,
452
+ "grad_norm": 0.0022263473365455866,
453
+ "learning_rate": 3.327643652843155e-05,
454
+ "loss": 0.004,
455
+ "step": 3100
456
+ },
457
+ {
458
+ "epoch": 1.0461640650946529,
459
+ "grad_norm": 0.0038769582752138376,
460
+ "learning_rate": 3.3137262808111886e-05,
461
+ "loss": 0.0003,
462
+ "step": 3150
463
+ },
464
+ {
465
+ "epoch": 1.0627698439056792,
466
+ "grad_norm": 0.005572175141423941,
467
+ "learning_rate": 3.299808908779222e-05,
468
+ "loss": 0.001,
469
+ "step": 3200
470
+ },
471
+ {
472
+ "epoch": 1.0793756227167055,
473
+ "grad_norm": 0.1954014003276825,
474
+ "learning_rate": 3.285891536747256e-05,
475
+ "loss": 0.0323,
476
+ "step": 3250
477
+ },
478
+ {
479
+ "epoch": 1.0959814015277316,
480
+ "grad_norm": 0.0821860134601593,
481
+ "learning_rate": 3.2719741647152894e-05,
482
+ "loss": 0.0035,
483
+ "step": 3300
484
+ },
485
+ {
486
+ "epoch": 1.112587180338758,
487
+ "grad_norm": 0.04094453528523445,
488
+ "learning_rate": 3.258056792683323e-05,
489
+ "loss": 0.0025,
490
+ "step": 3350
491
+ },
492
+ {
493
+ "epoch": 1.1291929591497842,
494
+ "grad_norm": 0.0028018904849886894,
495
+ "learning_rate": 3.2441394206513566e-05,
496
+ "loss": 0.0092,
497
+ "step": 3400
498
+ },
499
+ {
500
+ "epoch": 1.1457987379608103,
501
+ "grad_norm": 0.002664918312802911,
502
+ "learning_rate": 3.23022204861939e-05,
503
+ "loss": 0.0194,
504
+ "step": 3450
505
+ },
506
+ {
507
+ "epoch": 1.1624045167718366,
508
+ "grad_norm": 6.756401538848877,
509
+ "learning_rate": 3.216304676587424e-05,
510
+ "loss": 0.0135,
511
+ "step": 3500
512
+ },
513
+ {
514
+ "epoch": 1.1790102955828627,
515
+ "grad_norm": 0.0011647256324067712,
516
+ "learning_rate": 3.202387304555457e-05,
517
+ "loss": 0.0009,
518
+ "step": 3550
519
+ },
520
+ {
521
+ "epoch": 1.195616074393889,
522
+ "grad_norm": 0.0007124203257262707,
523
+ "learning_rate": 3.188469932523491e-05,
524
+ "loss": 0.0004,
525
+ "step": 3600
526
+ },
527
+ {
528
+ "epoch": 1.2122218532049154,
529
+ "grad_norm": 0.00033606469514779747,
530
+ "learning_rate": 3.174552560491525e-05,
531
+ "loss": 0.0,
532
+ "step": 3650
533
+ },
534
+ {
535
+ "epoch": 1.2288276320159415,
536
+ "grad_norm": 0.0005367195117287338,
537
+ "learning_rate": 3.160635188459559e-05,
538
+ "loss": 0.0001,
539
+ "step": 3700
540
+ },
541
+ {
542
+ "epoch": 1.2454334108269678,
543
+ "grad_norm": 0.012845808640122414,
544
+ "learning_rate": 3.1467178164275924e-05,
545
+ "loss": 0.0132,
546
+ "step": 3750
547
+ },
548
+ {
549
+ "epoch": 1.2620391896379939,
550
+ "grad_norm": 0.0020410455763339996,
551
+ "learning_rate": 3.132800444395626e-05,
552
+ "loss": 0.0006,
553
+ "step": 3800
554
+ },
555
+ {
556
+ "epoch": 1.2786449684490202,
557
+ "grad_norm": 0.0007005564984865487,
558
+ "learning_rate": 3.1188830723636595e-05,
559
+ "loss": 0.0074,
560
+ "step": 3850
561
+ },
562
+ {
563
+ "epoch": 1.2952507472600465,
564
+ "grad_norm": 0.00040667568100616336,
565
+ "learning_rate": 3.104965700331693e-05,
566
+ "loss": 0.004,
567
+ "step": 3900
568
+ },
569
+ {
570
+ "epoch": 1.3118565260710726,
571
+ "grad_norm": 0.0005291103734634817,
572
+ "learning_rate": 3.091048328299727e-05,
573
+ "loss": 0.0035,
574
+ "step": 3950
575
+ },
576
+ {
577
+ "epoch": 1.328462304882099,
578
+ "grad_norm": 0.01364582683891058,
579
+ "learning_rate": 3.077130956267761e-05,
580
+ "loss": 0.0157,
581
+ "step": 4000
582
+ },
583
+ {
584
+ "epoch": 1.3450680836931252,
585
+ "grad_norm": 4.155655860900879,
586
+ "learning_rate": 3.0632135842357946e-05,
587
+ "loss": 0.015,
588
+ "step": 4050
589
+ },
590
+ {
591
+ "epoch": 1.3616738625041513,
592
+ "grad_norm": 0.22104530036449432,
593
+ "learning_rate": 3.0492962122038278e-05,
594
+ "loss": 0.0118,
595
+ "step": 4100
596
+ },
597
+ {
598
+ "epoch": 1.3782796413151777,
599
+ "grad_norm": 0.027948148548603058,
600
+ "learning_rate": 3.0353788401718614e-05,
601
+ "loss": 0.0238,
602
+ "step": 4150
603
+ },
604
+ {
605
+ "epoch": 1.394885420126204,
606
+ "grad_norm": 0.0006893076351843774,
607
+ "learning_rate": 3.021461468139895e-05,
608
+ "loss": 0.0001,
609
+ "step": 4200
610
+ },
611
+ {
612
+ "epoch": 1.41149119893723,
613
+ "grad_norm": 0.00040428817737847567,
614
+ "learning_rate": 3.0075440961079286e-05,
615
+ "loss": 0.0003,
616
+ "step": 4250
617
+ },
618
+ {
619
+ "epoch": 1.4280969777482564,
620
+ "grad_norm": 0.13213320076465607,
621
+ "learning_rate": 2.993626724075962e-05,
622
+ "loss": 0.0192,
623
+ "step": 4300
624
+ },
625
+ {
626
+ "epoch": 1.4447027565592827,
627
+ "grad_norm": 0.010131658986210823,
628
+ "learning_rate": 2.979709352043996e-05,
629
+ "loss": 0.0114,
630
+ "step": 4350
631
+ },
632
+ {
633
+ "epoch": 1.4613085353703088,
634
+ "grad_norm": 0.0037651287857443094,
635
+ "learning_rate": 2.9657919800120297e-05,
636
+ "loss": 0.0311,
637
+ "step": 4400
638
+ },
639
+ {
640
+ "epoch": 1.4779143141813351,
641
+ "grad_norm": 0.002042593201622367,
642
+ "learning_rate": 2.9518746079800632e-05,
643
+ "loss": 0.0004,
644
+ "step": 4450
645
+ },
646
+ {
647
+ "epoch": 1.4945200929923614,
648
+ "grad_norm": 0.0017547437455505133,
649
+ "learning_rate": 2.9379572359480968e-05,
650
+ "loss": 0.0003,
651
+ "step": 4500
652
+ },
653
+ {
654
+ "epoch": 1.5111258718033875,
655
+ "grad_norm": 0.0004074271419085562,
656
+ "learning_rate": 2.9240398639161304e-05,
657
+ "loss": 0.0,
658
+ "step": 4550
659
+ },
660
+ {
661
+ "epoch": 1.5277316506144138,
662
+ "grad_norm": 0.0007494412711821496,
663
+ "learning_rate": 2.910122491884164e-05,
664
+ "loss": 0.0,
665
+ "step": 4600
666
+ },
667
+ {
668
+ "epoch": 1.5443374294254402,
669
+ "grad_norm": 0.0027854584623128176,
670
+ "learning_rate": 2.8962051198521976e-05,
671
+ "loss": 0.0101,
672
+ "step": 4650
673
+ },
674
+ {
675
+ "epoch": 1.5609432082364663,
676
+ "grad_norm": 0.0018231570720672607,
677
+ "learning_rate": 2.882287747820232e-05,
678
+ "loss": 0.0038,
679
+ "step": 4700
680
+ },
681
+ {
682
+ "epoch": 1.5775489870474926,
683
+ "grad_norm": 0.0018910124199464917,
684
+ "learning_rate": 2.8683703757882654e-05,
685
+ "loss": 0.0101,
686
+ "step": 4750
687
+ },
688
+ {
689
+ "epoch": 1.594154765858519,
690
+ "grad_norm": 0.02916385605931282,
691
+ "learning_rate": 2.854453003756299e-05,
692
+ "loss": 0.0116,
693
+ "step": 4800
694
+ },
695
+ {
696
+ "epoch": 1.610760544669545,
697
+ "grad_norm": 0.000494068895932287,
698
+ "learning_rate": 2.8405356317243326e-05,
699
+ "loss": 0.0003,
700
+ "step": 4850
701
+ },
702
+ {
703
+ "epoch": 1.627366323480571,
704
+ "grad_norm": 0.003024607663974166,
705
+ "learning_rate": 2.8266182596923662e-05,
706
+ "loss": 0.0017,
707
+ "step": 4900
708
+ },
709
+ {
710
+ "epoch": 1.6439721022915976,
711
+ "grad_norm": 0.0006424608873203397,
712
+ "learning_rate": 2.8127008876603998e-05,
713
+ "loss": 0.0004,
714
+ "step": 4950
715
+ },
716
+ {
717
+ "epoch": 1.6605778811026237,
718
+ "grad_norm": 0.0013206731528043747,
719
+ "learning_rate": 2.7987835156284334e-05,
720
+ "loss": 0.0421,
721
+ "step": 5000
722
+ },
723
+ {
724
+ "epoch": 1.6771836599136498,
725
+ "grad_norm": 0.009919991716742516,
726
+ "learning_rate": 2.7848661435964673e-05,
727
+ "loss": 0.014,
728
+ "step": 5050
729
+ },
730
+ {
731
+ "epoch": 1.6937894387246761,
732
+ "grad_norm": 0.0010778785217553377,
733
+ "learning_rate": 2.770948771564501e-05,
734
+ "loss": 0.0111,
735
+ "step": 5100
736
+ },
737
+ {
738
+ "epoch": 1.7103952175357025,
739
+ "grad_norm": 0.048366744071245193,
740
+ "learning_rate": 2.7570313995325345e-05,
741
+ "loss": 0.0173,
742
+ "step": 5150
743
+ },
744
+ {
745
+ "epoch": 1.7270009963467285,
746
+ "grad_norm": 0.0018048906931653619,
747
+ "learning_rate": 2.743114027500568e-05,
748
+ "loss": 0.0004,
749
+ "step": 5200
750
+ },
751
+ {
752
+ "epoch": 1.7436067751577549,
753
+ "grad_norm": 0.0049881902523338795,
754
+ "learning_rate": 2.7291966554686016e-05,
755
+ "loss": 0.0055,
756
+ "step": 5250
757
+ },
758
+ {
759
+ "epoch": 1.7602125539687812,
760
+ "grad_norm": 0.035248152911663055,
761
+ "learning_rate": 2.7152792834366352e-05,
762
+ "loss": 0.0139,
763
+ "step": 5300
764
+ },
765
+ {
766
+ "epoch": 1.7768183327798073,
767
+ "grad_norm": 0.0055509209632873535,
768
+ "learning_rate": 2.7013619114046688e-05,
769
+ "loss": 0.0119,
770
+ "step": 5350
771
+ },
772
+ {
773
+ "epoch": 1.7934241115908336,
774
+ "grad_norm": 0.0015154307475313544,
775
+ "learning_rate": 2.6874445393727027e-05,
776
+ "loss": 0.0059,
777
+ "step": 5400
778
+ },
779
+ {
780
+ "epoch": 1.81002989040186,
781
+ "grad_norm": 0.002082501072436571,
782
+ "learning_rate": 2.6735271673407363e-05,
783
+ "loss": 0.0001,
784
+ "step": 5450
785
+ },
786
+ {
787
+ "epoch": 1.826635669212886,
788
+ "grad_norm": 0.0010320444125682116,
789
+ "learning_rate": 2.65960979530877e-05,
790
+ "loss": 0.0026,
791
+ "step": 5500
792
+ },
793
+ {
794
+ "epoch": 1.8432414480239123,
795
+ "grad_norm": 0.0015966288046911359,
796
+ "learning_rate": 2.6456924232768035e-05,
797
+ "loss": 0.0001,
798
+ "step": 5550
799
+ },
800
+ {
801
+ "epoch": 1.8598472268349386,
802
+ "grad_norm": 0.0002549632336013019,
803
+ "learning_rate": 2.631775051244837e-05,
804
+ "loss": 0.0,
805
+ "step": 5600
806
+ },
807
+ {
808
+ "epoch": 1.8764530056459647,
809
+ "grad_norm": 0.003573563415557146,
810
+ "learning_rate": 2.6178576792128707e-05,
811
+ "loss": 0.0119,
812
+ "step": 5650
813
+ },
814
+ {
815
+ "epoch": 1.893058784456991,
816
+ "grad_norm": 0.0006668689311482012,
817
+ "learning_rate": 2.6039403071809043e-05,
818
+ "loss": 0.0002,
819
+ "step": 5700
820
+ },
821
+ {
822
+ "epoch": 1.9096645632680174,
823
+ "grad_norm": 0.0003313050256110728,
824
+ "learning_rate": 2.5900229351489382e-05,
825
+ "loss": 0.0005,
826
+ "step": 5750
827
+ },
828
+ {
829
+ "epoch": 1.9262703420790435,
830
+ "grad_norm": 0.000735230278223753,
831
+ "learning_rate": 2.5761055631169718e-05,
832
+ "loss": 0.0033,
833
+ "step": 5800
834
+ },
835
+ {
836
+ "epoch": 1.9428761208900698,
837
+ "grad_norm": 0.0006316429935395718,
838
+ "learning_rate": 2.5621881910850054e-05,
839
+ "loss": 0.0,
840
+ "step": 5850
841
+ },
842
+ {
843
+ "epoch": 1.959481899701096,
844
+ "grad_norm": 0.07467895746231079,
845
+ "learning_rate": 2.548270819053039e-05,
846
+ "loss": 0.0239,
847
+ "step": 5900
848
+ },
849
+ {
850
+ "epoch": 1.9760876785121222,
851
+ "grad_norm": 0.0030303276143968105,
852
+ "learning_rate": 2.5343534470210725e-05,
853
+ "loss": 0.0051,
854
+ "step": 5950
855
+ },
856
+ {
857
+ "epoch": 1.9926934573231485,
858
+ "grad_norm": 0.028144309297204018,
859
+ "learning_rate": 2.520436074989106e-05,
860
+ "loss": 0.0411,
861
+ "step": 6000
862
+ },
863
+ {
864
+ "epoch": 2.0,
865
+ "eval_accuracy": 0.9971374045801527,
866
+ "eval_f1": 0.9970892145409969,
867
+ "eval_loss": 0.022991055622696877,
868
+ "eval_precision": 0.9971362128759713,
869
+ "eval_recall": 0.9971374045801527,
870
+ "eval_runtime": 67.1475,
871
+ "eval_samples_per_second": 124.859,
872
+ "eval_steps_per_second": 7.804,
873
+ "step": 6022
874
+ },
875
+ {
876
+ "epoch": 2.009299236134175,
877
+ "grad_norm": 0.01069187093526125,
878
+ "learning_rate": 2.5065187029571397e-05,
879
+ "loss": 0.0176,
880
+ "step": 6050
881
+ },
882
+ {
883
+ "epoch": 2.025905014945201,
884
+ "grad_norm": 0.002449960447847843,
885
+ "learning_rate": 2.492601330925174e-05,
886
+ "loss": 0.0034,
887
+ "step": 6100
888
+ },
889
+ {
890
+ "epoch": 2.042510793756227,
891
+ "grad_norm": 0.006188757251948118,
892
+ "learning_rate": 2.4786839588932076e-05,
893
+ "loss": 0.0001,
894
+ "step": 6150
895
+ },
896
+ {
897
+ "epoch": 2.0591165725672536,
898
+ "grad_norm": 0.0007646733429282904,
899
+ "learning_rate": 2.464766586861241e-05,
900
+ "loss": 0.0049,
901
+ "step": 6200
902
+ },
903
+ {
904
+ "epoch": 2.0757223513782797,
905
+ "grad_norm": 0.008807332254946232,
906
+ "learning_rate": 2.4508492148292747e-05,
907
+ "loss": 0.0108,
908
+ "step": 6250
909
+ },
910
+ {
911
+ "epoch": 2.0923281301893057,
912
+ "grad_norm": 0.0009894605027511716,
913
+ "learning_rate": 2.4369318427973083e-05,
914
+ "loss": 0.0002,
915
+ "step": 6300
916
+ },
917
+ {
918
+ "epoch": 2.1089339090003323,
919
+ "grad_norm": 0.0016475095180794597,
920
+ "learning_rate": 2.423014470765342e-05,
921
+ "loss": 0.0009,
922
+ "step": 6350
923
+ },
924
+ {
925
+ "epoch": 2.1255396878113584,
926
+ "grad_norm": 0.000374704715795815,
927
+ "learning_rate": 2.4090970987333755e-05,
928
+ "loss": 0.002,
929
+ "step": 6400
930
+ },
931
+ {
932
+ "epoch": 2.1421454666223845,
933
+ "grad_norm": 0.0004738509014714509,
934
+ "learning_rate": 2.3951797267014094e-05,
935
+ "loss": 0.0,
936
+ "step": 6450
937
+ },
938
+ {
939
+ "epoch": 2.158751245433411,
940
+ "grad_norm": 0.004183988086879253,
941
+ "learning_rate": 2.381262354669443e-05,
942
+ "loss": 0.0059,
943
+ "step": 6500
944
+ },
945
+ {
946
+ "epoch": 2.175357024244437,
947
+ "grad_norm": 0.0020687321666628122,
948
+ "learning_rate": 2.3673449826374766e-05,
949
+ "loss": 0.0079,
950
+ "step": 6550
951
+ },
952
+ {
953
+ "epoch": 2.191962803055463,
954
+ "grad_norm": 0.00045966755715198815,
955
+ "learning_rate": 2.35342761060551e-05,
956
+ "loss": 0.0,
957
+ "step": 6600
958
+ },
959
+ {
960
+ "epoch": 2.2085685818664897,
961
+ "grad_norm": 0.001768000889569521,
962
+ "learning_rate": 2.3395102385735438e-05,
963
+ "loss": 0.0,
964
+ "step": 6650
965
+ },
966
+ {
967
+ "epoch": 2.225174360677516,
968
+ "grad_norm": 0.00014257608563639224,
969
+ "learning_rate": 2.3255928665415773e-05,
970
+ "loss": 0.0,
971
+ "step": 6700
972
+ },
973
+ {
974
+ "epoch": 2.241780139488542,
975
+ "grad_norm": 0.0002034334756899625,
976
+ "learning_rate": 2.311675494509611e-05,
977
+ "loss": 0.0121,
978
+ "step": 6750
979
+ },
980
+ {
981
+ "epoch": 2.2583859182995685,
982
+ "grad_norm": 0.7104772329330444,
983
+ "learning_rate": 2.297758122477645e-05,
984
+ "loss": 0.0,
985
+ "step": 6800
986
+ },
987
+ {
988
+ "epoch": 2.2749916971105946,
989
+ "grad_norm": 0.0013439609901979566,
990
+ "learning_rate": 2.2838407504456784e-05,
991
+ "loss": 0.0116,
992
+ "step": 6850
993
+ },
994
+ {
995
+ "epoch": 2.2915974759216207,
996
+ "grad_norm": 0.003318699076771736,
997
+ "learning_rate": 2.269923378413712e-05,
998
+ "loss": 0.0001,
999
+ "step": 6900
1000
+ },
1001
+ {
1002
+ "epoch": 2.308203254732647,
1003
+ "grad_norm": 0.00031852992833592,
1004
+ "learning_rate": 2.2560060063817456e-05,
1005
+ "loss": 0.0065,
1006
+ "step": 6950
1007
+ },
1008
+ {
1009
+ "epoch": 2.3248090335436733,
1010
+ "grad_norm": 0.00043174950405955315,
1011
+ "learning_rate": 2.2420886343497792e-05,
1012
+ "loss": 0.0,
1013
+ "step": 7000
1014
+ },
1015
+ {
1016
+ "epoch": 2.3414148123546994,
1017
+ "grad_norm": 0.008213113993406296,
1018
+ "learning_rate": 2.2281712623178128e-05,
1019
+ "loss": 0.0164,
1020
+ "step": 7050
1021
+ },
1022
+ {
1023
+ "epoch": 2.3580205911657255,
1024
+ "grad_norm": 0.0064167445525527,
1025
+ "learning_rate": 2.2142538902858464e-05,
1026
+ "loss": 0.0035,
1027
+ "step": 7100
1028
+ },
1029
+ {
1030
+ "epoch": 2.374626369976752,
1031
+ "grad_norm": 0.00048106853500939906,
1032
+ "learning_rate": 2.2003365182538803e-05,
1033
+ "loss": 0.0002,
1034
+ "step": 7150
1035
+ },
1036
+ {
1037
+ "epoch": 2.391232148787778,
1038
+ "grad_norm": 0.00031906799995340407,
1039
+ "learning_rate": 2.186419146221914e-05,
1040
+ "loss": 0.0032,
1041
+ "step": 7200
1042
+ },
1043
+ {
1044
+ "epoch": 2.407837927598804,
1045
+ "grad_norm": 0.0002367593697272241,
1046
+ "learning_rate": 2.1725017741899475e-05,
1047
+ "loss": 0.0001,
1048
+ "step": 7250
1049
+ },
1050
+ {
1051
+ "epoch": 2.4244437064098308,
1052
+ "grad_norm": 0.0002154409303329885,
1053
+ "learning_rate": 2.158584402157981e-05,
1054
+ "loss": 0.0,
1055
+ "step": 7300
1056
+ },
1057
+ {
1058
+ "epoch": 2.441049485220857,
1059
+ "grad_norm": 0.004206398501992226,
1060
+ "learning_rate": 2.1446670301260146e-05,
1061
+ "loss": 0.0,
1062
+ "step": 7350
1063
+ },
1064
+ {
1065
+ "epoch": 2.457655264031883,
1066
+ "grad_norm": 0.00027124237385578454,
1067
+ "learning_rate": 2.1307496580940482e-05,
1068
+ "loss": 0.0,
1069
+ "step": 7400
1070
+ },
1071
+ {
1072
+ "epoch": 2.4742610428429095,
1073
+ "grad_norm": 0.00019591822638176382,
1074
+ "learning_rate": 2.1168322860620818e-05,
1075
+ "loss": 0.0,
1076
+ "step": 7450
1077
+ },
1078
+ {
1079
+ "epoch": 2.4908668216539356,
1080
+ "grad_norm": 0.004273345228284597,
1081
+ "learning_rate": 2.102914914030116e-05,
1082
+ "loss": 0.0069,
1083
+ "step": 7500
1084
+ },
1085
+ {
1086
+ "epoch": 2.5074726004649617,
1087
+ "grad_norm": 0.0011552508221939206,
1088
+ "learning_rate": 2.0889975419981497e-05,
1089
+ "loss": 0.0108,
1090
+ "step": 7550
1091
+ },
1092
+ {
1093
+ "epoch": 2.5240783792759878,
1094
+ "grad_norm": 0.0009868694469332695,
1095
+ "learning_rate": 2.0750801699661833e-05,
1096
+ "loss": 0.0023,
1097
+ "step": 7600
1098
+ },
1099
+ {
1100
+ "epoch": 2.5406841580870143,
1101
+ "grad_norm": 0.00020178337581455708,
1102
+ "learning_rate": 2.061162797934217e-05,
1103
+ "loss": 0.0131,
1104
+ "step": 7650
1105
+ },
1106
+ {
1107
+ "epoch": 2.5572899368980404,
1108
+ "grad_norm": 0.005387285258620977,
1109
+ "learning_rate": 2.0472454259022504e-05,
1110
+ "loss": 0.0,
1111
+ "step": 7700
1112
+ },
1113
+ {
1114
+ "epoch": 2.5738957157090665,
1115
+ "grad_norm": 5.118713670526631e-05,
1116
+ "learning_rate": 2.033328053870284e-05,
1117
+ "loss": 0.0001,
1118
+ "step": 7750
1119
+ },
1120
+ {
1121
+ "epoch": 2.590501494520093,
1122
+ "grad_norm": 0.00040173486922867596,
1123
+ "learning_rate": 2.0194106818383176e-05,
1124
+ "loss": 0.0001,
1125
+ "step": 7800
1126
+ },
1127
+ {
1128
+ "epoch": 2.607107273331119,
1129
+ "grad_norm": 6.548867531819269e-05,
1130
+ "learning_rate": 2.0054933098063512e-05,
1131
+ "loss": 0.0,
1132
+ "step": 7850
1133
+ },
1134
+ {
1135
+ "epoch": 2.6237130521421452,
1136
+ "grad_norm": 0.00011618030839599669,
1137
+ "learning_rate": 1.991575937774385e-05,
1138
+ "loss": 0.0,
1139
+ "step": 7900
1140
+ },
1141
+ {
1142
+ "epoch": 2.6403188309531718,
1143
+ "grad_norm": 3.7753208744106814e-05,
1144
+ "learning_rate": 1.9776585657424187e-05,
1145
+ "loss": 0.0,
1146
+ "step": 7950
1147
+ },
1148
+ {
1149
+ "epoch": 2.656924609764198,
1150
+ "grad_norm": 0.00011960588017245755,
1151
+ "learning_rate": 1.9637411937104523e-05,
1152
+ "loss": 0.0,
1153
+ "step": 8000
1154
+ },
1155
+ {
1156
+ "epoch": 2.673530388575224,
1157
+ "grad_norm": 0.0005026832805015147,
1158
+ "learning_rate": 1.949823821678486e-05,
1159
+ "loss": 0.0,
1160
+ "step": 8050
1161
+ },
1162
+ {
1163
+ "epoch": 2.6901361673862505,
1164
+ "grad_norm": 5.5350832553813234e-05,
1165
+ "learning_rate": 1.9359064496465198e-05,
1166
+ "loss": 0.0,
1167
+ "step": 8100
1168
+ },
1169
+ {
1170
+ "epoch": 2.7067419461972766,
1171
+ "grad_norm": 0.00011889787128893659,
1172
+ "learning_rate": 1.9219890776145534e-05,
1173
+ "loss": 0.006,
1174
+ "step": 8150
1175
+ },
1176
+ {
1177
+ "epoch": 2.7233477250083027,
1178
+ "grad_norm": 0.001216597855091095,
1179
+ "learning_rate": 1.908071705582587e-05,
1180
+ "loss": 0.0,
1181
+ "step": 8200
1182
+ },
1183
+ {
1184
+ "epoch": 2.7399535038193292,
1185
+ "grad_norm": 0.000991505105048418,
1186
+ "learning_rate": 1.8941543335506206e-05,
1187
+ "loss": 0.028,
1188
+ "step": 8250
1189
+ },
1190
+ {
1191
+ "epoch": 2.7565592826303553,
1192
+ "grad_norm": 0.0008777762413956225,
1193
+ "learning_rate": 1.880236961518654e-05,
1194
+ "loss": 0.0017,
1195
+ "step": 8300
1196
+ },
1197
+ {
1198
+ "epoch": 2.7731650614413814,
1199
+ "grad_norm": 0.0007498575723730028,
1200
+ "learning_rate": 1.8663195894866877e-05,
1201
+ "loss": 0.0001,
1202
+ "step": 8350
1203
+ },
1204
+ {
1205
+ "epoch": 2.789770840252408,
1206
+ "grad_norm": 0.0012167900567874312,
1207
+ "learning_rate": 1.8524022174547213e-05,
1208
+ "loss": 0.0,
1209
+ "step": 8400
1210
+ },
1211
+ {
1212
+ "epoch": 2.806376619063434,
1213
+ "grad_norm": 0.007960589602589607,
1214
+ "learning_rate": 1.8384848454227552e-05,
1215
+ "loss": 0.0,
1216
+ "step": 8450
1217
+ },
1218
+ {
1219
+ "epoch": 2.82298239787446,
1220
+ "grad_norm": 0.00019711998174898326,
1221
+ "learning_rate": 1.8245674733907888e-05,
1222
+ "loss": 0.0,
1223
+ "step": 8500
1224
+ },
1225
+ {
1226
+ "epoch": 2.8395881766854867,
1227
+ "grad_norm": 0.009689416736364365,
1228
+ "learning_rate": 1.8106501013588224e-05,
1229
+ "loss": 0.012,
1230
+ "step": 8550
1231
+ },
1232
+ {
1233
+ "epoch": 2.856193955496513,
1234
+ "grad_norm": 0.0010081271175295115,
1235
+ "learning_rate": 1.796732729326856e-05,
1236
+ "loss": 0.0001,
1237
+ "step": 8600
1238
+ },
1239
+ {
1240
+ "epoch": 2.872799734307539,
1241
+ "grad_norm": 0.0037073129788041115,
1242
+ "learning_rate": 1.7828153572948896e-05,
1243
+ "loss": 0.0037,
1244
+ "step": 8650
1245
+ },
1246
+ {
1247
+ "epoch": 2.8894055131185654,
1248
+ "grad_norm": 0.000657514261547476,
1249
+ "learning_rate": 1.7688979852629232e-05,
1250
+ "loss": 0.0005,
1251
+ "step": 8700
1252
+ },
1253
+ {
1254
+ "epoch": 2.9060112919295915,
1255
+ "grad_norm": 0.0041756597347557545,
1256
+ "learning_rate": 1.7549806132309568e-05,
1257
+ "loss": 0.0001,
1258
+ "step": 8750
1259
+ },
1260
+ {
1261
+ "epoch": 2.9226170707406176,
1262
+ "grad_norm": 0.0014196323463693261,
1263
+ "learning_rate": 1.7410632411989907e-05,
1264
+ "loss": 0.0051,
1265
+ "step": 8800
1266
+ },
1267
+ {
1268
+ "epoch": 2.939222849551644,
1269
+ "grad_norm": 0.001023727236315608,
1270
+ "learning_rate": 1.7271458691670243e-05,
1271
+ "loss": 0.0112,
1272
+ "step": 8850
1273
+ },
1274
+ {
1275
+ "epoch": 2.9558286283626702,
1276
+ "grad_norm": 0.0005508707836270332,
1277
+ "learning_rate": 1.713228497135058e-05,
1278
+ "loss": 0.0001,
1279
+ "step": 8900
1280
+ },
1281
+ {
1282
+ "epoch": 2.9724344071736963,
1283
+ "grad_norm": 0.0005536659737117589,
1284
+ "learning_rate": 1.6993111251030918e-05,
1285
+ "loss": 0.0,
1286
+ "step": 8950
1287
+ },
1288
+ {
1289
+ "epoch": 2.989040185984723,
1290
+ "grad_norm": 0.002692180685698986,
1291
+ "learning_rate": 1.6853937530711254e-05,
1292
+ "loss": 0.0018,
1293
+ "step": 9000
1294
+ },
1295
+ {
1296
+ "epoch": 3.0,
1297
+ "eval_accuracy": 0.9977337786259542,
1298
+ "eval_f1": 0.997711567093862,
1299
+ "eval_loss": 0.017316868528723717,
1300
+ "eval_precision": 0.9977219427584941,
1301
+ "eval_recall": 0.9977337786259542,
1302
+ "eval_runtime": 67.8318,
1303
+ "eval_samples_per_second": 123.6,
1304
+ "eval_steps_per_second": 7.725,
1305
+ "step": 9033
1306
+ }
1307
+ ],
1308
+ "logging_steps": 50,
1309
+ "max_steps": 15055,
1310
+ "num_input_tokens_seen": 0,
1311
+ "num_train_epochs": 5,
1312
+ "save_steps": 500,
1313
+ "stateful_callbacks": {
1314
+ "TrainerControl": {
1315
+ "args": {
1316
+ "should_epoch_stop": false,
1317
+ "should_evaluate": false,
1318
+ "should_log": false,
1319
+ "should_save": true,
1320
+ "should_training_stop": false
1321
+ },
1322
+ "attributes": {}
1323
+ }
1324
+ },
1325
+ "total_flos": 1.5282741915076608e+17,
1326
+ "train_batch_size": 16,
1327
+ "trial_name": null,
1328
+ "trial_params": null
1329
+ }
trial-2/checkpoint-9033/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31b16aa5b08a9764ff6ccb4b3d56d2656dee1972624cbbcf01b67cb965ad3e06
3
+ size 5368
trial-3/checkpoint-1506/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-3/checkpoint-1506/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2633ff2f035355d1f491e943541f26f2b553f8bb1f21a963431b246a1dac02a3
3
+ size 1583351632
trial-3/checkpoint-1506/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8dbd147ad035350b88e1d53f3cf901e1b9d28549304b83bb06a4cc02825d41cf
3
+ size 3166813178
trial-3/checkpoint-1506/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568428d80a25211a390c359ca51b0b20b38ca0607fbc196f106c9841c02d3e59
3
+ size 14244
trial-3/checkpoint-1506/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baefd0e97bcced8a792772527e556de3be3ee540a7e8c67d6914b57db907793c
3
+ size 1064
trial-3/checkpoint-1506/trainer_state.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.029849544167518616,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-3/checkpoint-1506",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1506,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.033200531208499334,
13
+ "grad_norm": 23.8559627532959,
14
+ "learning_rate": 4.534517164997864e-06,
15
+ "loss": 0.4406,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.06640106241699867,
20
+ "grad_norm": 6.17399787902832,
21
+ "learning_rate": 4.51562019398187e-06,
22
+ "loss": 0.1819,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.099601593625498,
27
+ "grad_norm": 0.49522921442985535,
28
+ "learning_rate": 4.496723222965876e-06,
29
+ "loss": 0.107,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.13280212483399734,
34
+ "grad_norm": 7.754148960113525,
35
+ "learning_rate": 4.477826251949882e-06,
36
+ "loss": 0.0688,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.16600265604249667,
41
+ "grad_norm": 0.5113905668258667,
42
+ "learning_rate": 4.458929280933889e-06,
43
+ "loss": 0.0495,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.199203187250996,
48
+ "grad_norm": 4.4989800453186035,
49
+ "learning_rate": 4.440032309917895e-06,
50
+ "loss": 0.0415,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.23240371845949534,
55
+ "grad_norm": 9.69454288482666,
56
+ "learning_rate": 4.421135338901901e-06,
57
+ "loss": 0.0663,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.2656042496679947,
62
+ "grad_norm": 0.023875955492258072,
63
+ "learning_rate": 4.4022383678859074e-06,
64
+ "loss": 0.0373,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.29880478087649404,
69
+ "grad_norm": 0.09503920376300812,
70
+ "learning_rate": 4.383341396869914e-06,
71
+ "loss": 0.0444,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.33200531208499334,
76
+ "grad_norm": 0.008267635479569435,
77
+ "learning_rate": 4.36444442585392e-06,
78
+ "loss": 0.0286,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.3652058432934927,
83
+ "grad_norm": 0.12851744890213013,
84
+ "learning_rate": 4.345547454837926e-06,
85
+ "loss": 0.0204,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.398406374501992,
90
+ "grad_norm": 2.524458646774292,
91
+ "learning_rate": 4.3266504838219325e-06,
92
+ "loss": 0.0095,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.4316069057104914,
97
+ "grad_norm": 1.337737798690796,
98
+ "learning_rate": 4.307753512805939e-06,
99
+ "loss": 0.0249,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.4648074369189907,
104
+ "grad_norm": 0.08153943717479706,
105
+ "learning_rate": 4.288856541789945e-06,
106
+ "loss": 0.0092,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.49800796812749004,
111
+ "grad_norm": 0.018035605549812317,
112
+ "learning_rate": 4.269959570773951e-06,
113
+ "loss": 0.0132,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.5312084993359893,
118
+ "grad_norm": 0.22391293942928314,
119
+ "learning_rate": 4.251062599757957e-06,
120
+ "loss": 0.0103,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.5644090305444888,
125
+ "grad_norm": 7.64361047744751,
126
+ "learning_rate": 4.232165628741963e-06,
127
+ "loss": 0.0099,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.5976095617529881,
132
+ "grad_norm": 0.0029439961072057486,
133
+ "learning_rate": 4.213268657725969e-06,
134
+ "loss": 0.0099,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.6308100929614874,
139
+ "grad_norm": 0.02843591570854187,
140
+ "learning_rate": 4.194371686709975e-06,
141
+ "loss": 0.0051,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.6640106241699867,
146
+ "grad_norm": 0.21529018878936768,
147
+ "learning_rate": 4.175474715693982e-06,
148
+ "loss": 0.0165,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.6972111553784861,
153
+ "grad_norm": 0.0392189547419548,
154
+ "learning_rate": 4.156577744677988e-06,
155
+ "loss": 0.011,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.7304116865869854,
160
+ "grad_norm": 0.006516862660646439,
161
+ "learning_rate": 4.137680773661994e-06,
162
+ "loss": 0.002,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.7636122177954847,
167
+ "grad_norm": 0.0022248616442084312,
168
+ "learning_rate": 4.1187838026460004e-06,
169
+ "loss": 0.0117,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.796812749003984,
174
+ "grad_norm": 0.014311583712697029,
175
+ "learning_rate": 4.099886831630007e-06,
176
+ "loss": 0.0115,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.8300132802124834,
181
+ "grad_norm": 0.006024663802236319,
182
+ "learning_rate": 4.080989860614013e-06,
183
+ "loss": 0.0004,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.8632138114209827,
188
+ "grad_norm": 0.005676358472555876,
189
+ "learning_rate": 4.062092889598019e-06,
190
+ "loss": 0.0067,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.896414342629482,
195
+ "grad_norm": 0.0005013855989091098,
196
+ "learning_rate": 4.0431959185820255e-06,
197
+ "loss": 0.0008,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.9296148738379814,
202
+ "grad_norm": 0.0016384737100452185,
203
+ "learning_rate": 4.024298947566032e-06,
204
+ "loss": 0.001,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.9628154050464808,
209
+ "grad_norm": 3.2831873893737793,
210
+ "learning_rate": 4.005401976550038e-06,
211
+ "loss": 0.0015,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.9960159362549801,
216
+ "grad_norm": 0.036648038774728775,
217
+ "learning_rate": 3.986505005534044e-06,
218
+ "loss": 0.0061,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 1.0,
223
+ "eval_accuracy": 0.9970181297709924,
224
+ "eval_f1": 0.9969934810410721,
225
+ "eval_loss": 0.029849544167518616,
226
+ "eval_precision": 0.9969918072979542,
227
+ "eval_recall": 0.9970181297709924,
228
+ "eval_runtime": 59.3241,
229
+ "eval_samples_per_second": 141.325,
230
+ "eval_steps_per_second": 4.416,
231
+ "step": 1506
232
+ }
233
+ ],
234
+ "logging_steps": 50,
235
+ "max_steps": 12048,
236
+ "num_input_tokens_seen": 0,
237
+ "num_train_epochs": 8,
238
+ "save_steps": 500,
239
+ "stateful_callbacks": {
240
+ "TrainerControl": {
241
+ "args": {
242
+ "should_epoch_stop": false,
243
+ "should_evaluate": false,
244
+ "should_log": false,
245
+ "should_save": true,
246
+ "should_training_stop": false
247
+ },
248
+ "attributes": {}
249
+ }
250
+ },
251
+ "total_flos": 5.094247305025536e+16,
252
+ "train_batch_size": 32,
253
+ "trial_name": null,
254
+ "trial_params": null
255
+ }
trial-3/checkpoint-1506/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:862f96821c71b38bd72c3c10e43adbfe554490b2332bd838409265267f16997e
3
+ size 5368
trial-4/checkpoint-3012/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-4/checkpoint-3012/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84a78c605436363f9db346ac678e170cff4e009ca2331def7f65ff704e3349e4
3
+ size 1583351632
trial-4/checkpoint-3012/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4385bc937b3ea0adfa5c713df04eb100e89eb731c8f09a620e7ed3c4d32d4df3
3
+ size 3166813178
trial-4/checkpoint-3012/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:914f37830aa379563c31bd15a8b8f53b8ccc8e2de0f0aa6da9695369e4ad84ef
3
+ size 14244
trial-4/checkpoint-3012/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:159228974380d6d8a61d5fb2da4e5cc76ffc86287d9ca0582ad4c46cb816cc56
3
+ size 1064
trial-4/checkpoint-3012/trainer_state.json ADDED
@@ -0,0 +1,477 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.019555753096938133,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-4/checkpoint-3012",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 3012,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.033200531208499334,
13
+ "grad_norm": 111.04224395751953,
14
+ "learning_rate": 1.1277048866432555e-05,
15
+ "loss": 0.3116,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.06640106241699867,
20
+ "grad_norm": 0.14363102614879608,
21
+ "learning_rate": 1.123948374695743e-05,
22
+ "loss": 0.098,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.099601593625498,
27
+ "grad_norm": 0.10250398516654968,
28
+ "learning_rate": 1.1201918627482305e-05,
29
+ "loss": 0.0729,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.13280212483399734,
34
+ "grad_norm": 1.9557462930679321,
35
+ "learning_rate": 1.116435350800718e-05,
36
+ "loss": 0.0537,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.16600265604249667,
41
+ "grad_norm": 0.5240038633346558,
42
+ "learning_rate": 1.1126788388532055e-05,
43
+ "loss": 0.0442,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.199203187250996,
48
+ "grad_norm": 2.139970541000366,
49
+ "learning_rate": 1.1089223269056931e-05,
50
+ "loss": 0.036,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.23240371845949534,
55
+ "grad_norm": 0.04191768541932106,
56
+ "learning_rate": 1.1051658149581805e-05,
57
+ "loss": 0.0355,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.2656042496679947,
62
+ "grad_norm": 0.0023582959547638893,
63
+ "learning_rate": 1.1014093030106681e-05,
64
+ "loss": 0.0368,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.29880478087649404,
69
+ "grad_norm": 12.193012237548828,
70
+ "learning_rate": 1.0976527910631555e-05,
71
+ "loss": 0.0574,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.33200531208499334,
76
+ "grad_norm": 0.00729788513854146,
77
+ "learning_rate": 1.0938962791156431e-05,
78
+ "loss": 0.0238,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.3652058432934927,
83
+ "grad_norm": 1.7422609329223633,
84
+ "learning_rate": 1.0901397671681305e-05,
85
+ "loss": 0.0213,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.398406374501992,
90
+ "grad_norm": 0.06725198775529861,
91
+ "learning_rate": 1.086383255220618e-05,
92
+ "loss": 0.013,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.4316069057104914,
97
+ "grad_norm": 0.21940571069717407,
98
+ "learning_rate": 1.0826267432731055e-05,
99
+ "loss": 0.025,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.4648074369189907,
104
+ "grad_norm": 0.032884348183870316,
105
+ "learning_rate": 1.078870231325593e-05,
106
+ "loss": 0.0138,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.49800796812749004,
111
+ "grad_norm": 0.0014803586527705193,
112
+ "learning_rate": 1.0751137193780805e-05,
113
+ "loss": 0.0058,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.5312084993359893,
118
+ "grad_norm": 10.339282989501953,
119
+ "learning_rate": 1.071357207430568e-05,
120
+ "loss": 0.009,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.5644090305444888,
125
+ "grad_norm": 1.531823754310608,
126
+ "learning_rate": 1.0676006954830555e-05,
127
+ "loss": 0.0115,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.5976095617529881,
132
+ "grad_norm": 0.00043045339407399297,
133
+ "learning_rate": 1.063844183535543e-05,
134
+ "loss": 0.0068,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.6308100929614874,
139
+ "grad_norm": 0.025898275896906853,
140
+ "learning_rate": 1.0600876715880303e-05,
141
+ "loss": 0.0098,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.6640106241699867,
146
+ "grad_norm": 0.0007346518104895949,
147
+ "learning_rate": 1.0563311596405178e-05,
148
+ "loss": 0.0094,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.6972111553784861,
153
+ "grad_norm": 0.005919306073337793,
154
+ "learning_rate": 1.0525746476930054e-05,
155
+ "loss": 0.0115,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.7304116865869854,
160
+ "grad_norm": 0.04206903651356697,
161
+ "learning_rate": 1.0488181357454928e-05,
162
+ "loss": 0.0061,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.7636122177954847,
167
+ "grad_norm": 0.000557853898499161,
168
+ "learning_rate": 1.0450616237979804e-05,
169
+ "loss": 0.0024,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.796812749003984,
174
+ "grad_norm": 0.0012018937850371003,
175
+ "learning_rate": 1.0413051118504678e-05,
176
+ "loss": 0.0058,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.8300132802124834,
181
+ "grad_norm": 0.0013845885405316949,
182
+ "learning_rate": 1.0375485999029554e-05,
183
+ "loss": 0.0022,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.8632138114209827,
188
+ "grad_norm": 0.07051751017570496,
189
+ "learning_rate": 1.0337920879554428e-05,
190
+ "loss": 0.0049,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.896414342629482,
195
+ "grad_norm": 0.00019932868599426,
196
+ "learning_rate": 1.0300355760079302e-05,
197
+ "loss": 0.0026,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.9296148738379814,
202
+ "grad_norm": 2.3489619707106613e-05,
203
+ "learning_rate": 1.0262790640604178e-05,
204
+ "loss": 0.0027,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.9628154050464808,
209
+ "grad_norm": 0.0037654500920325518,
210
+ "learning_rate": 1.0225225521129052e-05,
211
+ "loss": 0.0079,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.9960159362549801,
216
+ "grad_norm": 0.05777144059538841,
217
+ "learning_rate": 1.0187660401653928e-05,
218
+ "loss": 0.0154,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 1.0,
223
+ "eval_accuracy": 0.9967795801526718,
224
+ "eval_f1": 0.9967913721697382,
225
+ "eval_loss": 0.022535286843776703,
226
+ "eval_precision": 0.9968076807415951,
227
+ "eval_recall": 0.9967795801526718,
228
+ "eval_runtime": 59.7088,
229
+ "eval_samples_per_second": 140.415,
230
+ "eval_steps_per_second": 4.388,
231
+ "step": 1506
232
+ },
233
+ {
234
+ "epoch": 1.0292164674634794,
235
+ "grad_norm": 0.020927241072058678,
236
+ "learning_rate": 1.0150095282178802e-05,
237
+ "loss": 0.0068,
238
+ "step": 1550
239
+ },
240
+ {
241
+ "epoch": 1.0624169986719787,
242
+ "grad_norm": 0.00013940146891400218,
243
+ "learning_rate": 1.0112530162703678e-05,
244
+ "loss": 0.0023,
245
+ "step": 1600
246
+ },
247
+ {
248
+ "epoch": 1.095617529880478,
249
+ "grad_norm": 0.01088524330407381,
250
+ "learning_rate": 1.0074965043228552e-05,
251
+ "loss": 0.0062,
252
+ "step": 1650
253
+ },
254
+ {
255
+ "epoch": 1.1288180610889773,
256
+ "grad_norm": 0.058450598269701004,
257
+ "learning_rate": 1.0037399923753428e-05,
258
+ "loss": 0.0153,
259
+ "step": 1700
260
+ },
261
+ {
262
+ "epoch": 1.1620185922974768,
263
+ "grad_norm": 0.011701447889208794,
264
+ "learning_rate": 9.999834804278302e-06,
265
+ "loss": 0.0105,
266
+ "step": 1750
267
+ },
268
+ {
269
+ "epoch": 1.1952191235059761,
270
+ "grad_norm": 3.078742742538452,
271
+ "learning_rate": 9.962269684803178e-06,
272
+ "loss": 0.0076,
273
+ "step": 1800
274
+ },
275
+ {
276
+ "epoch": 1.2284196547144755,
277
+ "grad_norm": 0.02360646426677704,
278
+ "learning_rate": 9.924704565328052e-06,
279
+ "loss": 0.0025,
280
+ "step": 1850
281
+ },
282
+ {
283
+ "epoch": 1.2616201859229748,
284
+ "grad_norm": 0.0027641034685075283,
285
+ "learning_rate": 9.887139445852926e-06,
286
+ "loss": 0.0114,
287
+ "step": 1900
288
+ },
289
+ {
290
+ "epoch": 1.294820717131474,
291
+ "grad_norm": 0.00172056641895324,
292
+ "learning_rate": 9.849574326377802e-06,
293
+ "loss": 0.0003,
294
+ "step": 1950
295
+ },
296
+ {
297
+ "epoch": 1.3280212483399734,
298
+ "grad_norm": 0.07806120812892914,
299
+ "learning_rate": 9.812009206902676e-06,
300
+ "loss": 0.0008,
301
+ "step": 2000
302
+ },
303
+ {
304
+ "epoch": 1.361221779548473,
305
+ "grad_norm": 0.000884020933881402,
306
+ "learning_rate": 9.774444087427552e-06,
307
+ "loss": 0.0038,
308
+ "step": 2050
309
+ },
310
+ {
311
+ "epoch": 1.3944223107569722,
312
+ "grad_norm": 0.1488543003797531,
313
+ "learning_rate": 9.736878967952426e-06,
314
+ "loss": 0.0106,
315
+ "step": 2100
316
+ },
317
+ {
318
+ "epoch": 1.4276228419654715,
319
+ "grad_norm": 0.037523552775382996,
320
+ "learning_rate": 9.699313848477302e-06,
321
+ "loss": 0.0007,
322
+ "step": 2150
323
+ },
324
+ {
325
+ "epoch": 1.4608233731739708,
326
+ "grad_norm": 0.00033480292768217623,
327
+ "learning_rate": 9.661748729002176e-06,
328
+ "loss": 0.0116,
329
+ "step": 2200
330
+ },
331
+ {
332
+ "epoch": 1.4940239043824701,
333
+ "grad_norm": 0.007270739413797855,
334
+ "learning_rate": 9.624183609527052e-06,
335
+ "loss": 0.0079,
336
+ "step": 2250
337
+ },
338
+ {
339
+ "epoch": 1.5272244355909694,
340
+ "grad_norm": 0.00751983979716897,
341
+ "learning_rate": 9.586618490051926e-06,
342
+ "loss": 0.0044,
343
+ "step": 2300
344
+ },
345
+ {
346
+ "epoch": 1.5604249667994687,
347
+ "grad_norm": 4.298997402191162,
348
+ "learning_rate": 9.549053370576802e-06,
349
+ "loss": 0.0014,
350
+ "step": 2350
351
+ },
352
+ {
353
+ "epoch": 1.593625498007968,
354
+ "grad_norm": 0.0014925749273970723,
355
+ "learning_rate": 9.511488251101676e-06,
356
+ "loss": 0.0001,
357
+ "step": 2400
358
+ },
359
+ {
360
+ "epoch": 1.6268260292164674,
361
+ "grad_norm": 0.003561707679182291,
362
+ "learning_rate": 9.47392313162655e-06,
363
+ "loss": 0.0077,
364
+ "step": 2450
365
+ },
366
+ {
367
+ "epoch": 1.6600265604249667,
368
+ "grad_norm": 0.0006838434492237866,
369
+ "learning_rate": 9.436358012151426e-06,
370
+ "loss": 0.0001,
371
+ "step": 2500
372
+ },
373
+ {
374
+ "epoch": 1.6932270916334662,
375
+ "grad_norm": 0.0002309294941369444,
376
+ "learning_rate": 9.3987928926763e-06,
377
+ "loss": 0.0029,
378
+ "step": 2550
379
+ },
380
+ {
381
+ "epoch": 1.7264276228419655,
382
+ "grad_norm": 0.0011594091774895787,
383
+ "learning_rate": 9.361227773201176e-06,
384
+ "loss": 0.0107,
385
+ "step": 2600
386
+ },
387
+ {
388
+ "epoch": 1.7596281540504648,
389
+ "grad_norm": 0.00012066392082488164,
390
+ "learning_rate": 9.32366265372605e-06,
391
+ "loss": 0.0006,
392
+ "step": 2650
393
+ },
394
+ {
395
+ "epoch": 1.792828685258964,
396
+ "grad_norm": 0.0021935878321528435,
397
+ "learning_rate": 9.286097534250926e-06,
398
+ "loss": 0.0065,
399
+ "step": 2700
400
+ },
401
+ {
402
+ "epoch": 1.8260292164674636,
403
+ "grad_norm": 0.0002105861931340769,
404
+ "learning_rate": 9.2485324147758e-06,
405
+ "loss": 0.0032,
406
+ "step": 2750
407
+ },
408
+ {
409
+ "epoch": 1.859229747675963,
410
+ "grad_norm": 0.0009871097281575203,
411
+ "learning_rate": 9.210967295300676e-06,
412
+ "loss": 0.0,
413
+ "step": 2800
414
+ },
415
+ {
416
+ "epoch": 1.8924302788844622,
417
+ "grad_norm": 5.9001271438319236e-05,
418
+ "learning_rate": 9.173402175825549e-06,
419
+ "loss": 0.0077,
420
+ "step": 2850
421
+ },
422
+ {
423
+ "epoch": 1.9256308100929616,
424
+ "grad_norm": 0.001614038716070354,
425
+ "learning_rate": 9.135837056350425e-06,
426
+ "loss": 0.0059,
427
+ "step": 2900
428
+ },
429
+ {
430
+ "epoch": 1.9588313413014609,
431
+ "grad_norm": 0.001071999897249043,
432
+ "learning_rate": 9.098271936875299e-06,
433
+ "loss": 0.008,
434
+ "step": 2950
435
+ },
436
+ {
437
+ "epoch": 1.9920318725099602,
438
+ "grad_norm": 0.018942702561616898,
439
+ "learning_rate": 9.060706817400175e-06,
440
+ "loss": 0.0107,
441
+ "step": 3000
442
+ },
443
+ {
444
+ "epoch": 2.0,
445
+ "eval_accuracy": 0.997256679389313,
446
+ "eval_f1": 0.997238185991172,
447
+ "eval_loss": 0.019555753096938133,
448
+ "eval_precision": 0.997235028769195,
449
+ "eval_recall": 0.997256679389313,
450
+ "eval_runtime": 59.293,
451
+ "eval_samples_per_second": 141.4,
452
+ "eval_steps_per_second": 4.419,
453
+ "step": 3012
454
+ }
455
+ ],
456
+ "logging_steps": 50,
457
+ "max_steps": 15060,
458
+ "num_input_tokens_seen": 0,
459
+ "num_train_epochs": 10,
460
+ "save_steps": 500,
461
+ "stateful_callbacks": {
462
+ "TrainerControl": {
463
+ "args": {
464
+ "should_epoch_stop": false,
465
+ "should_evaluate": false,
466
+ "should_log": false,
467
+ "should_save": true,
468
+ "should_training_stop": false
469
+ },
470
+ "attributes": {}
471
+ }
472
+ },
473
+ "total_flos": 1.0188494610051072e+17,
474
+ "train_batch_size": 32,
475
+ "trial_name": null,
476
+ "trial_params": null
477
+ }
trial-4/checkpoint-3012/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6980b2d26960e0944dcf6a709651feb944395421a5df274ddff277d8343607e
3
+ size 5368
trial-5/checkpoint-1506/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-5/checkpoint-1506/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f5ecd51b91e86db092fb98e5e0ed6bb61c710dc95a00fe365f829c3218537a1
3
+ size 1583351632
trial-5/checkpoint-1506/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0309590d2cd916de4a2f04ddce5837c3c90bd59129970b6cb4f3ed8ac4e03b7
3
+ size 3166813178
trial-5/checkpoint-1506/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568428d80a25211a390c359ca51b0b20b38ca0607fbc196f106c9841c02d3e59
3
+ size 14244
trial-5/checkpoint-1506/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c12ccce76e7b5dcd2d6e5c79dcffd36e8cc7c4912a60cfbc89196c37d5a03edc
3
+ size 1064
trial-5/checkpoint-1506/trainer_state.json ADDED
@@ -0,0 +1,255 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.022090721875429153,
3
+ "best_model_checkpoint": "./results/answerdotai/ModernBERT-large/trial-5/checkpoint-1506",
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1506,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.033200531208499334,
13
+ "grad_norm": 21.02681541442871,
14
+ "learning_rate": 1.1549761256591334e-05,
15
+ "loss": 0.3094,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.06640106241699867,
20
+ "grad_norm": 14.655999183654785,
21
+ "learning_rate": 1.1494720457617871e-05,
22
+ "loss": 0.0991,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.099601593625498,
27
+ "grad_norm": 0.3093714416027069,
28
+ "learning_rate": 1.143967965864441e-05,
29
+ "loss": 0.045,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.13280212483399734,
34
+ "grad_norm": 0.19615088403224945,
35
+ "learning_rate": 1.1384638859670947e-05,
36
+ "loss": 0.0505,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.16600265604249667,
41
+ "grad_norm": 1.4286335706710815,
42
+ "learning_rate": 1.1329598060697483e-05,
43
+ "loss": 0.0281,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.199203187250996,
48
+ "grad_norm": 2.1918282508850098,
49
+ "learning_rate": 1.1274557261724022e-05,
50
+ "loss": 0.0536,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.23240371845949534,
55
+ "grad_norm": 2.13693904876709,
56
+ "learning_rate": 1.1219516462750559e-05,
57
+ "loss": 0.0385,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.2656042496679947,
62
+ "grad_norm": 0.029252415522933006,
63
+ "learning_rate": 1.1164475663777096e-05,
64
+ "loss": 0.0326,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.29880478087649404,
69
+ "grad_norm": 0.5969660878181458,
70
+ "learning_rate": 1.1109434864803635e-05,
71
+ "loss": 0.0547,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.33200531208499334,
76
+ "grad_norm": 0.00738520547747612,
77
+ "learning_rate": 1.1054394065830171e-05,
78
+ "loss": 0.0115,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 0.3652058432934927,
83
+ "grad_norm": 1.9720779657363892,
84
+ "learning_rate": 1.099935326685671e-05,
85
+ "loss": 0.0239,
86
+ "step": 550
87
+ },
88
+ {
89
+ "epoch": 0.398406374501992,
90
+ "grad_norm": 0.6309007406234741,
91
+ "learning_rate": 1.0944312467883247e-05,
92
+ "loss": 0.0043,
93
+ "step": 600
94
+ },
95
+ {
96
+ "epoch": 0.4316069057104914,
97
+ "grad_norm": 0.011916632764041424,
98
+ "learning_rate": 1.0889271668909786e-05,
99
+ "loss": 0.0164,
100
+ "step": 650
101
+ },
102
+ {
103
+ "epoch": 0.4648074369189907,
104
+ "grad_norm": 0.009058245457708836,
105
+ "learning_rate": 1.0834230869936323e-05,
106
+ "loss": 0.0103,
107
+ "step": 700
108
+ },
109
+ {
110
+ "epoch": 0.49800796812749004,
111
+ "grad_norm": 0.0032912548631429672,
112
+ "learning_rate": 1.0779190070962861e-05,
113
+ "loss": 0.028,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 0.5312084993359893,
118
+ "grad_norm": 0.011074424721300602,
119
+ "learning_rate": 1.0724149271989398e-05,
120
+ "loss": 0.0104,
121
+ "step": 800
122
+ },
123
+ {
124
+ "epoch": 0.5644090305444888,
125
+ "grad_norm": 0.951624870300293,
126
+ "learning_rate": 1.0669108473015937e-05,
127
+ "loss": 0.0114,
128
+ "step": 850
129
+ },
130
+ {
131
+ "epoch": 0.5976095617529881,
132
+ "grad_norm": 0.0027369000017642975,
133
+ "learning_rate": 1.0614067674042474e-05,
134
+ "loss": 0.016,
135
+ "step": 900
136
+ },
137
+ {
138
+ "epoch": 0.6308100929614874,
139
+ "grad_norm": 0.012001908384263515,
140
+ "learning_rate": 1.055902687506901e-05,
141
+ "loss": 0.0094,
142
+ "step": 950
143
+ },
144
+ {
145
+ "epoch": 0.6640106241699867,
146
+ "grad_norm": 0.69849693775177,
147
+ "learning_rate": 1.050398607609555e-05,
148
+ "loss": 0.0199,
149
+ "step": 1000
150
+ },
151
+ {
152
+ "epoch": 0.6972111553784861,
153
+ "grad_norm": 0.036301348358392715,
154
+ "learning_rate": 1.0448945277122086e-05,
155
+ "loss": 0.0077,
156
+ "step": 1050
157
+ },
158
+ {
159
+ "epoch": 0.7304116865869854,
160
+ "grad_norm": 0.008320258930325508,
161
+ "learning_rate": 1.0393904478148625e-05,
162
+ "loss": 0.0043,
163
+ "step": 1100
164
+ },
165
+ {
166
+ "epoch": 0.7636122177954847,
167
+ "grad_norm": 0.0027414376381784678,
168
+ "learning_rate": 1.0338863679175162e-05,
169
+ "loss": 0.004,
170
+ "step": 1150
171
+ },
172
+ {
173
+ "epoch": 0.796812749003984,
174
+ "grad_norm": 0.0007768127834424376,
175
+ "learning_rate": 1.02838228802017e-05,
176
+ "loss": 0.0072,
177
+ "step": 1200
178
+ },
179
+ {
180
+ "epoch": 0.8300132802124834,
181
+ "grad_norm": 0.0004548605065792799,
182
+ "learning_rate": 1.0228782081228237e-05,
183
+ "loss": 0.0051,
184
+ "step": 1250
185
+ },
186
+ {
187
+ "epoch": 0.8632138114209827,
188
+ "grad_norm": 0.007403654046356678,
189
+ "learning_rate": 1.0173741282254776e-05,
190
+ "loss": 0.0112,
191
+ "step": 1300
192
+ },
193
+ {
194
+ "epoch": 0.896414342629482,
195
+ "grad_norm": 0.0017905730055645108,
196
+ "learning_rate": 1.0118700483281313e-05,
197
+ "loss": 0.0069,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 0.9296148738379814,
202
+ "grad_norm": 0.0004039919876959175,
203
+ "learning_rate": 1.0063659684307851e-05,
204
+ "loss": 0.0008,
205
+ "step": 1400
206
+ },
207
+ {
208
+ "epoch": 0.9628154050464808,
209
+ "grad_norm": 0.000610634742770344,
210
+ "learning_rate": 1.0008618885334388e-05,
211
+ "loss": 0.0011,
212
+ "step": 1450
213
+ },
214
+ {
215
+ "epoch": 0.9960159362549801,
216
+ "grad_norm": 1.0261385440826416,
217
+ "learning_rate": 9.953578086360925e-06,
218
+ "loss": 0.0207,
219
+ "step": 1500
220
+ },
221
+ {
222
+ "epoch": 1.0,
223
+ "eval_accuracy": 0.9971374045801527,
224
+ "eval_f1": 0.9971246021172431,
225
+ "eval_loss": 0.022090721875429153,
226
+ "eval_precision": 0.9971185448871154,
227
+ "eval_recall": 0.9971374045801527,
228
+ "eval_runtime": 59.3469,
229
+ "eval_samples_per_second": 141.271,
230
+ "eval_steps_per_second": 4.415,
231
+ "step": 1506
232
+ }
233
+ ],
234
+ "logging_steps": 50,
235
+ "max_steps": 10542,
236
+ "num_input_tokens_seen": 0,
237
+ "num_train_epochs": 7,
238
+ "save_steps": 500,
239
+ "stateful_callbacks": {
240
+ "TrainerControl": {
241
+ "args": {
242
+ "should_epoch_stop": false,
243
+ "should_evaluate": false,
244
+ "should_log": false,
245
+ "should_save": true,
246
+ "should_training_stop": false
247
+ },
248
+ "attributes": {}
249
+ }
250
+ },
251
+ "total_flos": 5.094247305025536e+16,
252
+ "train_batch_size": 32,
253
+ "trial_name": null,
254
+ "trial_params": null
255
+ }
trial-5/checkpoint-1506/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57df86703a932a26c855abebf91502e9da3c5daba164538c1fa2ecde95e9c014
3
+ size 5368
trial-6/checkpoint-1506/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-large",
3
+ "architectures": [
4
+ "ModernBertForSequenceClassification"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 50281,
9
+ "classifier_activation": "gelu",
10
+ "classifier_bias": false,
11
+ "classifier_dropout": 0.0,
12
+ "classifier_pooling": "mean",
13
+ "cls_token_id": 50281,
14
+ "decoder_bias": true,
15
+ "deterministic_flash_attn": false,
16
+ "embedding_dropout": 0.0,
17
+ "eos_token_id": 50282,
18
+ "global_attn_every_n_layers": 3,
19
+ "global_rope_theta": 160000.0,
20
+ "gradient_checkpointing": false,
21
+ "hidden_activation": "gelu",
22
+ "hidden_size": 1024,
23
+ "initializer_cutoff_factor": 2.0,
24
+ "initializer_range": 0.02,
25
+ "intermediate_size": 2624,
26
+ "layer_norm_eps": 1e-05,
27
+ "local_attention": 128,
28
+ "local_rope_theta": 10000.0,
29
+ "max_position_embeddings": 8192,
30
+ "mlp_bias": false,
31
+ "mlp_dropout": 0.0,
32
+ "model_type": "modernbert",
33
+ "norm_bias": false,
34
+ "norm_eps": 1e-05,
35
+ "num_attention_heads": 16,
36
+ "num_hidden_layers": 28,
37
+ "pad_token_id": 50283,
38
+ "position_embedding_type": "absolute",
39
+ "problem_type": "single_label_classification",
40
+ "reference_compile": true,
41
+ "sep_token_id": 50282,
42
+ "sparse_pred_ignore_index": -100,
43
+ "sparse_prediction": false,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.48.0.dev0",
46
+ "vocab_size": 50368
47
+ }
trial-6/checkpoint-1506/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d8dd9c50dd83347fd2f3d076bda888d52bc1f28720f6537878e55817f57b843
3
+ size 1583351632
trial-6/checkpoint-1506/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba4b588a87754fc94b6d70748348851ee39e0cd46fcba5bc4098f18c984a4eee
3
+ size 3166813178
trial-6/checkpoint-1506/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:568428d80a25211a390c359ca51b0b20b38ca0607fbc196f106c9841c02d3e59
3
+ size 14244