Hugging Face's logo

Forecast-ing
/
modernBERT-content-regression

Text Classification
Transformers
TensorBoard
Safetensors
modernbert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Forecast-ing commited on
Commit
1a8ed53
·
verified ·
1 Parent(s): 55ecd49

Training in progress, epoch 5

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. model.safetensors +1 -1
  2. run-0/checkpoint-620/config.json +53 -0
  3. run-0/checkpoint-620/model.safetensors +3 -0
  4. run-0/checkpoint-620/optimizer.pt +3 -0
  5. run-0/checkpoint-620/rng_state.pth +3 -0
  6. run-0/checkpoint-620/scheduler.pt +3 -0
  7. run-0/checkpoint-620/special_tokens_map.json +37 -0
  8. run-0/checkpoint-620/tokenizer.json +0 -0
  9. run-0/checkpoint-620/tokenizer_config.json +945 -0
  10. run-0/checkpoint-620/trainer_state.json +0 -0
  11. run-0/checkpoint-620/training_args.bin +3 -0
  12. run-1/checkpoint-248/config.json +53 -0
  13. run-1/checkpoint-248/model.safetensors +3 -0
  14. run-1/checkpoint-248/optimizer.pt +3 -0
  15. run-1/checkpoint-248/rng_state.pth +3 -0
  16. run-1/checkpoint-248/scheduler.pt +3 -0
  17. run-1/checkpoint-248/special_tokens_map.json +37 -0
  18. run-1/checkpoint-248/tokenizer.json +0 -0
  19. run-1/checkpoint-248/tokenizer_config.json +945 -0
  20. run-1/checkpoint-248/trainer_state.json +1797 -0
  21. run-1/checkpoint-248/training_args.bin +3 -0
  22. run-2/checkpoint-496/config.json +53 -0
  23. run-2/checkpoint-496/model.safetensors +3 -0
  24. run-2/checkpoint-496/optimizer.pt +3 -0
  25. run-2/checkpoint-496/rng_state.pth +3 -0
  26. run-2/checkpoint-496/scheduler.pt +3 -0
  27. run-2/checkpoint-496/special_tokens_map.json +37 -0
  28. run-2/checkpoint-496/tokenizer.json +0 -0
  29. run-2/checkpoint-496/tokenizer_config.json +945 -0
  30. run-2/checkpoint-496/trainer_state.json +3559 -0
  31. run-2/checkpoint-496/training_args.bin +3 -0
  32. run-3/checkpoint-248/config.json +53 -0
  33. run-3/checkpoint-248/model.safetensors +3 -0
  34. run-3/checkpoint-248/optimizer.pt +3 -0
  35. run-3/checkpoint-248/rng_state.pth +3 -0
  36. run-3/checkpoint-248/scheduler.pt +3 -0
  37. run-3/checkpoint-248/special_tokens_map.json +37 -0
  38. run-3/checkpoint-248/tokenizer.json +0 -0
  39. run-3/checkpoint-248/tokenizer_config.json +945 -0
  40. run-3/checkpoint-248/trainer_state.json +1797 -0
  41. run-3/checkpoint-248/training_args.bin +3 -0
  42. run-4/checkpoint-620/config.json +53 -0
  43. run-4/checkpoint-620/model.safetensors +3 -0
  44. run-4/checkpoint-620/optimizer.pt +3 -0
  45. run-4/checkpoint-620/rng_state.pth +3 -0
  46. run-4/checkpoint-620/scheduler.pt +3 -0
  47. run-4/checkpoint-620/special_tokens_map.json +37 -0
  48. run-4/checkpoint-620/tokenizer.json +0 -0
  49. run-4/checkpoint-620/tokenizer_config.json +945 -0
  50. run-4/checkpoint-620/trainer_state.json +0 -0
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:63ce04291ff3cde47e004ce4a78062278319b1f2df3a143ec551200c85d881c3
3
  size 598436708
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f61b6742fda1cb36ad4ec9d282f082abd58406f1812932fdda295db23bf3ea3
3
  size 598436708
run-0/checkpoint-620/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-base",
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": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0"
25
+ },
26
+ "initializer_cutoff_factor": 2.0,
27
+ "initializer_range": 0.02,
28
+ "intermediate_size": 1152,
29
+ "label2id": {
30
+ "LABEL_0": 0
31
+ },
32
+ "layer_norm_eps": 1e-05,
33
+ "local_attention": 128,
34
+ "local_rope_theta": 10000.0,
35
+ "max_position_embeddings": 8192,
36
+ "mlp_bias": false,
37
+ "mlp_dropout": 0.0,
38
+ "model_type": "modernbert",
39
+ "norm_bias": false,
40
+ "norm_eps": 1e-05,
41
+ "num_attention_heads": 12,
42
+ "num_hidden_layers": 22,
43
+ "pad_token_id": 50283,
44
+ "position_embedding_type": "absolute",
45
+ "problem_type": "regression",
46
+ "reference_compile": true,
47
+ "sep_token_id": 50282,
48
+ "sparse_pred_ignore_index": -100,
49
+ "sparse_prediction": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.48.0.dev0",
52
+ "vocab_size": 50368
53
+ }
run-0/checkpoint-620/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b70fc61aeff6b2d61e696669fb13d9cfcbe148ec1ba38e32a0bdb490c979aa3
3
+ size 598436708
run-0/checkpoint-620/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3af5d8906b3d198b38fcffe62e94e30e993d9a9a83d3b847e0606b5d7d2debbb
3
+ size 1196963194
run-0/checkpoint-620/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5c27b160b159e371c7d5122ee5052944383a2bf6a7498d51d058a4a6a3dd7d
3
+ size 14244
run-0/checkpoint-620/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d55ba6c7a22353d027b4a953fa237e77598e69657c325b52522dba1693ec4d02
3
+ size 1064
run-0/checkpoint-620/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": true,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
run-0/checkpoint-620/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-0/checkpoint-620/tokenizer_config.json ADDED
@@ -0,0 +1,945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "|||IP_ADDRESS|||",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1": {
12
+ "content": "<|padding|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "50254": {
20
+ "content": " ",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": false
26
+ },
27
+ "50255": {
28
+ "content": " ",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": false
34
+ },
35
+ "50256": {
36
+ "content": " ",
37
+ "lstrip": false,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": false
42
+ },
43
+ "50257": {
44
+ "content": " ",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ },
51
+ "50258": {
52
+ "content": " ",
53
+ "lstrip": false,
54
+ "normalized": true,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": false
58
+ },
59
+ "50259": {
60
+ "content": " ",
61
+ "lstrip": false,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": false
66
+ },
67
+ "50260": {
68
+ "content": " ",
69
+ "lstrip": false,
70
+ "normalized": true,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": false
74
+ },
75
+ "50261": {
76
+ "content": " ",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "50262": {
84
+ "content": " ",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "50263": {
92
+ "content": " ",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "50264": {
100
+ "content": " ",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "50265": {
108
+ "content": " ",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "50266": {
116
+ "content": " ",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "50267": {
124
+ "content": " ",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "50268": {
132
+ "content": " ",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "50269": {
140
+ "content": " ",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "50270": {
148
+ "content": " ",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "50271": {
156
+ "content": " ",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "50272": {
164
+ "content": " ",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "50273": {
172
+ "content": " ",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "50274": {
180
+ "content": " ",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "50275": {
188
+ "content": " ",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "50276": {
196
+ "content": " ",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "50277": {
204
+ "content": "|||EMAIL_ADDRESS|||",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "50278": {
212
+ "content": "|||PHONE_NUMBER|||",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "50279": {
220
+ "content": "<|endoftext|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "50280": {
228
+ "content": "[UNK]",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "50281": {
236
+ "content": "[CLS]",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "50282": {
244
+ "content": "[SEP]",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "50283": {
252
+ "content": "[PAD]",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "50284": {
260
+ "content": "[MASK]",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "50285": {
268
+ "content": "[unused0]",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "50286": {
276
+ "content": "[unused1]",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "50287": {
284
+ "content": "[unused2]",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "50288": {
292
+ "content": "[unused3]",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "50289": {
300
+ "content": "[unused4]",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "50290": {
308
+ "content": "[unused5]",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "50291": {
316
+ "content": "[unused6]",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "50292": {
324
+ "content": "[unused7]",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "50293": {
332
+ "content": "[unused8]",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "50294": {
340
+ "content": "[unused9]",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "50295": {
348
+ "content": "[unused10]",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "50296": {
356
+ "content": "[unused11]",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "50297": {
364
+ "content": "[unused12]",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "50298": {
372
+ "content": "[unused13]",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "50299": {
380
+ "content": "[unused14]",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "50300": {
388
+ "content": "[unused15]",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "50301": {
396
+ "content": "[unused16]",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "50302": {
404
+ "content": "[unused17]",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "50303": {
412
+ "content": "[unused18]",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "50304": {
420
+ "content": "[unused19]",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "50305": {
428
+ "content": "[unused20]",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "50306": {
436
+ "content": "[unused21]",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "50307": {
444
+ "content": "[unused22]",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "50308": {
452
+ "content": "[unused23]",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "50309": {
460
+ "content": "[unused24]",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "50310": {
468
+ "content": "[unused25]",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "50311": {
476
+ "content": "[unused26]",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "50312": {
484
+ "content": "[unused27]",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "50313": {
492
+ "content": "[unused28]",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "50314": {
500
+ "content": "[unused29]",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "50315": {
508
+ "content": "[unused30]",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "50316": {
516
+ "content": "[unused31]",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "50317": {
524
+ "content": "[unused32]",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "50318": {
532
+ "content": "[unused33]",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "50319": {
540
+ "content": "[unused34]",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "50320": {
548
+ "content": "[unused35]",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "50321": {
556
+ "content": "[unused36]",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "50322": {
564
+ "content": "[unused37]",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "50323": {
572
+ "content": "[unused38]",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "50324": {
580
+ "content": "[unused39]",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "50325": {
588
+ "content": "[unused40]",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "50326": {
596
+ "content": "[unused41]",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "50327": {
604
+ "content": "[unused42]",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "50328": {
612
+ "content": "[unused43]",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "50329": {
620
+ "content": "[unused44]",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "50330": {
628
+ "content": "[unused45]",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "50331": {
636
+ "content": "[unused46]",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "50332": {
644
+ "content": "[unused47]",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "50333": {
652
+ "content": "[unused48]",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "50334": {
660
+ "content": "[unused49]",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "50335": {
668
+ "content": "[unused50]",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "50336": {
676
+ "content": "[unused51]",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "50337": {
684
+ "content": "[unused52]",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "50338": {
692
+ "content": "[unused53]",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "50339": {
700
+ "content": "[unused54]",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "50340": {
708
+ "content": "[unused55]",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "50341": {
716
+ "content": "[unused56]",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "50342": {
724
+ "content": "[unused57]",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "50343": {
732
+ "content": "[unused58]",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "50344": {
740
+ "content": "[unused59]",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "50345": {
748
+ "content": "[unused60]",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "50346": {
756
+ "content": "[unused61]",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "50347": {
764
+ "content": "[unused62]",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "50348": {
772
+ "content": "[unused63]",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "50349": {
780
+ "content": "[unused64]",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "50350": {
788
+ "content": "[unused65]",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "50351": {
796
+ "content": "[unused66]",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "50352": {
804
+ "content": "[unused67]",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "50353": {
812
+ "content": "[unused68]",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "50354": {
820
+ "content": "[unused69]",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "50355": {
828
+ "content": "[unused70]",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "50356": {
836
+ "content": "[unused71]",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "50357": {
844
+ "content": "[unused72]",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "50358": {
852
+ "content": "[unused73]",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "50359": {
860
+ "content": "[unused74]",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "50360": {
868
+ "content": "[unused75]",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "50361": {
876
+ "content": "[unused76]",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "50362": {
884
+ "content": "[unused77]",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "50363": {
892
+ "content": "[unused78]",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "50364": {
900
+ "content": "[unused79]",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "50365": {
908
+ "content": "[unused80]",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "50366": {
916
+ "content": "[unused81]",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "50367": {
924
+ "content": "[unused82]",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ }
931
+ },
932
+ "clean_up_tokenization_spaces": true,
933
+ "cls_token": "[CLS]",
934
+ "extra_special_tokens": {},
935
+ "mask_token": "[MASK]",
936
+ "model_input_names": [
937
+ "input_ids",
938
+ "attention_mask"
939
+ ],
940
+ "model_max_length": 2048,
941
+ "pad_token": "[PAD]",
942
+ "sep_token": "[SEP]",
943
+ "tokenizer_class": "PreTrainedTokenizerFast",
944
+ "unk_token": "[UNK]"
945
+ }
run-0/checkpoint-620/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
run-0/checkpoint-620/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4d4476fb01884be49bbbc0370083d191538792d601f2fbc1aedc6941a21dbec
3
+ size 5432
run-1/checkpoint-248/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-base",
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": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0"
25
+ },
26
+ "initializer_cutoff_factor": 2.0,
27
+ "initializer_range": 0.02,
28
+ "intermediate_size": 1152,
29
+ "label2id": {
30
+ "LABEL_0": 0
31
+ },
32
+ "layer_norm_eps": 1e-05,
33
+ "local_attention": 128,
34
+ "local_rope_theta": 10000.0,
35
+ "max_position_embeddings": 8192,
36
+ "mlp_bias": false,
37
+ "mlp_dropout": 0.0,
38
+ "model_type": "modernbert",
39
+ "norm_bias": false,
40
+ "norm_eps": 1e-05,
41
+ "num_attention_heads": 12,
42
+ "num_hidden_layers": 22,
43
+ "pad_token_id": 50283,
44
+ "position_embedding_type": "absolute",
45
+ "problem_type": "regression",
46
+ "reference_compile": true,
47
+ "sep_token_id": 50282,
48
+ "sparse_pred_ignore_index": -100,
49
+ "sparse_prediction": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.48.0.dev0",
52
+ "vocab_size": 50368
53
+ }
run-1/checkpoint-248/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43e91c1956e55437751fe2a86332978fb00659e4aea08beee964c5e35a3d2534
3
+ size 598436708
run-1/checkpoint-248/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58c8263f607c78c42520b5be6053492cd19bd62728ddc2d65a05984615e5c829
3
+ size 1196963194
run-1/checkpoint-248/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c61318d640fc0fcb7f8a9868b31a885fae09243869e3f7810bd29eb4e61b970c
3
+ size 14244
run-1/checkpoint-248/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce0c77bd16241f4adceff44b5e8cadadfcdf7efd96614bf074821c7f2169dc45
3
+ size 1064
run-1/checkpoint-248/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": true,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
run-1/checkpoint-248/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-1/checkpoint-248/tokenizer_config.json ADDED
@@ -0,0 +1,945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "|||IP_ADDRESS|||",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1": {
12
+ "content": "<|padding|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "50254": {
20
+ "content": " ",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": false
26
+ },
27
+ "50255": {
28
+ "content": " ",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": false
34
+ },
35
+ "50256": {
36
+ "content": " ",
37
+ "lstrip": false,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": false
42
+ },
43
+ "50257": {
44
+ "content": " ",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ },
51
+ "50258": {
52
+ "content": " ",
53
+ "lstrip": false,
54
+ "normalized": true,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": false
58
+ },
59
+ "50259": {
60
+ "content": " ",
61
+ "lstrip": false,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": false
66
+ },
67
+ "50260": {
68
+ "content": " ",
69
+ "lstrip": false,
70
+ "normalized": true,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": false
74
+ },
75
+ "50261": {
76
+ "content": " ",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "50262": {
84
+ "content": " ",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "50263": {
92
+ "content": " ",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "50264": {
100
+ "content": " ",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "50265": {
108
+ "content": " ",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "50266": {
116
+ "content": " ",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "50267": {
124
+ "content": " ",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "50268": {
132
+ "content": " ",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "50269": {
140
+ "content": " ",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "50270": {
148
+ "content": " ",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "50271": {
156
+ "content": " ",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "50272": {
164
+ "content": " ",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "50273": {
172
+ "content": " ",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "50274": {
180
+ "content": " ",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "50275": {
188
+ "content": " ",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "50276": {
196
+ "content": " ",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "50277": {
204
+ "content": "|||EMAIL_ADDRESS|||",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "50278": {
212
+ "content": "|||PHONE_NUMBER|||",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "50279": {
220
+ "content": "<|endoftext|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "50280": {
228
+ "content": "[UNK]",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "50281": {
236
+ "content": "[CLS]",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "50282": {
244
+ "content": "[SEP]",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "50283": {
252
+ "content": "[PAD]",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "50284": {
260
+ "content": "[MASK]",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "50285": {
268
+ "content": "[unused0]",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "50286": {
276
+ "content": "[unused1]",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "50287": {
284
+ "content": "[unused2]",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "50288": {
292
+ "content": "[unused3]",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "50289": {
300
+ "content": "[unused4]",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "50290": {
308
+ "content": "[unused5]",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "50291": {
316
+ "content": "[unused6]",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "50292": {
324
+ "content": "[unused7]",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "50293": {
332
+ "content": "[unused8]",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "50294": {
340
+ "content": "[unused9]",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "50295": {
348
+ "content": "[unused10]",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "50296": {
356
+ "content": "[unused11]",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "50297": {
364
+ "content": "[unused12]",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "50298": {
372
+ "content": "[unused13]",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "50299": {
380
+ "content": "[unused14]",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "50300": {
388
+ "content": "[unused15]",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "50301": {
396
+ "content": "[unused16]",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "50302": {
404
+ "content": "[unused17]",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "50303": {
412
+ "content": "[unused18]",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "50304": {
420
+ "content": "[unused19]",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "50305": {
428
+ "content": "[unused20]",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "50306": {
436
+ "content": "[unused21]",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "50307": {
444
+ "content": "[unused22]",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "50308": {
452
+ "content": "[unused23]",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "50309": {
460
+ "content": "[unused24]",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "50310": {
468
+ "content": "[unused25]",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "50311": {
476
+ "content": "[unused26]",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "50312": {
484
+ "content": "[unused27]",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "50313": {
492
+ "content": "[unused28]",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "50314": {
500
+ "content": "[unused29]",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "50315": {
508
+ "content": "[unused30]",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "50316": {
516
+ "content": "[unused31]",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "50317": {
524
+ "content": "[unused32]",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "50318": {
532
+ "content": "[unused33]",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "50319": {
540
+ "content": "[unused34]",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "50320": {
548
+ "content": "[unused35]",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "50321": {
556
+ "content": "[unused36]",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "50322": {
564
+ "content": "[unused37]",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "50323": {
572
+ "content": "[unused38]",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "50324": {
580
+ "content": "[unused39]",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "50325": {
588
+ "content": "[unused40]",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "50326": {
596
+ "content": "[unused41]",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "50327": {
604
+ "content": "[unused42]",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "50328": {
612
+ "content": "[unused43]",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "50329": {
620
+ "content": "[unused44]",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "50330": {
628
+ "content": "[unused45]",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "50331": {
636
+ "content": "[unused46]",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "50332": {
644
+ "content": "[unused47]",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "50333": {
652
+ "content": "[unused48]",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "50334": {
660
+ "content": "[unused49]",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "50335": {
668
+ "content": "[unused50]",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "50336": {
676
+ "content": "[unused51]",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "50337": {
684
+ "content": "[unused52]",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "50338": {
692
+ "content": "[unused53]",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "50339": {
700
+ "content": "[unused54]",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "50340": {
708
+ "content": "[unused55]",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "50341": {
716
+ "content": "[unused56]",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "50342": {
724
+ "content": "[unused57]",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "50343": {
732
+ "content": "[unused58]",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "50344": {
740
+ "content": "[unused59]",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "50345": {
748
+ "content": "[unused60]",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "50346": {
756
+ "content": "[unused61]",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "50347": {
764
+ "content": "[unused62]",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "50348": {
772
+ "content": "[unused63]",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "50349": {
780
+ "content": "[unused64]",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "50350": {
788
+ "content": "[unused65]",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "50351": {
796
+ "content": "[unused66]",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "50352": {
804
+ "content": "[unused67]",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "50353": {
812
+ "content": "[unused68]",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "50354": {
820
+ "content": "[unused69]",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "50355": {
828
+ "content": "[unused70]",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "50356": {
836
+ "content": "[unused71]",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "50357": {
844
+ "content": "[unused72]",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "50358": {
852
+ "content": "[unused73]",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "50359": {
860
+ "content": "[unused74]",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "50360": {
868
+ "content": "[unused75]",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "50361": {
876
+ "content": "[unused76]",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "50362": {
884
+ "content": "[unused77]",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "50363": {
892
+ "content": "[unused78]",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "50364": {
900
+ "content": "[unused79]",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "50365": {
908
+ "content": "[unused80]",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "50366": {
916
+ "content": "[unused81]",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "50367": {
924
+ "content": "[unused82]",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ }
931
+ },
932
+ "clean_up_tokenization_spaces": true,
933
+ "cls_token": "[CLS]",
934
+ "extra_special_tokens": {},
935
+ "mask_token": "[MASK]",
936
+ "model_input_names": [
937
+ "input_ids",
938
+ "attention_mask"
939
+ ],
940
+ "model_max_length": 2048,
941
+ "pad_token": "[PAD]",
942
+ "sep_token": "[SEP]",
943
+ "tokenizer_class": "PreTrainedTokenizerFast",
944
+ "unk_token": "[UNK]"
945
+ }
run-1/checkpoint-248/trainer_state.json ADDED
@@ -0,0 +1,1797 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.8794732457657044,
3
+ "best_model_checkpoint": "./modernBERT-content-regression/run-1/checkpoint-248",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 248,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.008064516129032258,
13
+ "grad_norm": 312.8101501464844,
14
+ "learning_rate": 1.578455807125323e-06,
15
+ "loss": 21.3087,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.016129032258064516,
20
+ "grad_norm": 107.2333755493164,
21
+ "learning_rate": 1.5759057977438605e-06,
22
+ "loss": 1.8325,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.024193548387096774,
27
+ "grad_norm": 524.5511474609375,
28
+ "learning_rate": 1.573355788362398e-06,
29
+ "loss": 153.4526,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.03225806451612903,
34
+ "grad_norm": 200.25125122070312,
35
+ "learning_rate": 1.5708057789809354e-06,
36
+ "loss": 17.9352,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.04032258064516129,
41
+ "grad_norm": 106.64122009277344,
42
+ "learning_rate": 1.5682557695994728e-06,
43
+ "loss": 3.896,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.04838709677419355,
48
+ "grad_norm": 30.507457733154297,
49
+ "learning_rate": 1.5657057602180104e-06,
50
+ "loss": 0.3924,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.056451612903225805,
55
+ "grad_norm": 149.90699768066406,
56
+ "learning_rate": 1.5631557508365479e-06,
57
+ "loss": 8.8508,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.06451612903225806,
62
+ "grad_norm": 310.6943359375,
63
+ "learning_rate": 1.560605741455085e-06,
64
+ "loss": 19.5681,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.07258064516129033,
69
+ "grad_norm": 43.26463317871094,
70
+ "learning_rate": 1.5580557320736225e-06,
71
+ "loss": 0.3058,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.08064516129032258,
76
+ "grad_norm": 113.04293823242188,
77
+ "learning_rate": 1.5555057226921601e-06,
78
+ "loss": 4.2101,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.08870967741935484,
83
+ "grad_norm": 235.2570343017578,
84
+ "learning_rate": 1.5529557133106976e-06,
85
+ "loss": 16.8788,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.0967741935483871,
90
+ "grad_norm": 40.037757873535156,
91
+ "learning_rate": 1.5504057039292348e-06,
92
+ "loss": 0.8335,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.10483870967741936,
97
+ "grad_norm": 39.64387893676758,
98
+ "learning_rate": 1.5478556945477724e-06,
99
+ "loss": 0.2746,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.11290322580645161,
104
+ "grad_norm": 114.99188995361328,
105
+ "learning_rate": 1.5453056851663099e-06,
106
+ "loss": 2.96,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.12096774193548387,
111
+ "grad_norm": 21.276885986328125,
112
+ "learning_rate": 1.5427556757848473e-06,
113
+ "loss": 0.2243,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.12903225806451613,
118
+ "grad_norm": 160.51058959960938,
119
+ "learning_rate": 1.5402056664033847e-06,
120
+ "loss": 10.2011,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.13709677419354838,
125
+ "grad_norm": 18.14629364013672,
126
+ "learning_rate": 1.5376556570219222e-06,
127
+ "loss": 1.0228,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.14516129032258066,
132
+ "grad_norm": 91.3558349609375,
133
+ "learning_rate": 1.5351056476404596e-06,
134
+ "loss": 2.3441,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.1532258064516129,
139
+ "grad_norm": 43.8216667175293,
140
+ "learning_rate": 1.532555638258997e-06,
141
+ "loss": 0.8131,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.16129032258064516,
146
+ "grad_norm": 290.0969543457031,
147
+ "learning_rate": 1.5300056288775344e-06,
148
+ "loss": 10.4761,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.1693548387096774,
153
+ "grad_norm": 126.67745971679688,
154
+ "learning_rate": 1.527455619496072e-06,
155
+ "loss": 1.4179,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.1774193548387097,
160
+ "grad_norm": 204.8350830078125,
161
+ "learning_rate": 1.5249056101146093e-06,
162
+ "loss": 2.6962,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.18548387096774194,
167
+ "grad_norm": 127.1915054321289,
168
+ "learning_rate": 1.5223556007331467e-06,
169
+ "loss": 17.3755,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.1935483870967742,
174
+ "grad_norm": 250.90074157714844,
175
+ "learning_rate": 1.5198055913516844e-06,
176
+ "loss": 3.7896,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.20161290322580644,
181
+ "grad_norm": 92.21452331542969,
182
+ "learning_rate": 1.5172555819702218e-06,
183
+ "loss": 1.0904,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.20967741935483872,
188
+ "grad_norm": 114.67794036865234,
189
+ "learning_rate": 1.514705572588759e-06,
190
+ "loss": 1.1346,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.21774193548387097,
195
+ "grad_norm": 201.17628479003906,
196
+ "learning_rate": 1.5121555632072964e-06,
197
+ "loss": 2.9464,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.22580645161290322,
202
+ "grad_norm": 158.13563537597656,
203
+ "learning_rate": 1.509605553825834e-06,
204
+ "loss": 2.2773,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.23387096774193547,
209
+ "grad_norm": 86.0600357055664,
210
+ "learning_rate": 1.5070555444443715e-06,
211
+ "loss": 1.229,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.24193548387096775,
216
+ "grad_norm": 36.81131362915039,
217
+ "learning_rate": 1.504505535062909e-06,
218
+ "loss": 0.5345,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.25,
223
+ "grad_norm": 27.546449661254883,
224
+ "learning_rate": 1.5019555256814464e-06,
225
+ "loss": 0.9453,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.25806451612903225,
230
+ "grad_norm": 43.41412353515625,
231
+ "learning_rate": 1.4994055162999838e-06,
232
+ "loss": 0.9254,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.2661290322580645,
237
+ "grad_norm": 61.36131286621094,
238
+ "learning_rate": 1.4968555069185212e-06,
239
+ "loss": 0.9972,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.27419354838709675,
244
+ "grad_norm": 44.677345275878906,
245
+ "learning_rate": 1.4943054975370587e-06,
246
+ "loss": 1.4439,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.28225806451612906,
251
+ "grad_norm": 401.5712890625,
252
+ "learning_rate": 1.4917554881555963e-06,
253
+ "loss": 24.0851,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.2903225806451613,
258
+ "grad_norm": 40.864654541015625,
259
+ "learning_rate": 1.4892054787741335e-06,
260
+ "loss": 0.5807,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.29838709677419356,
265
+ "grad_norm": 95.5814208984375,
266
+ "learning_rate": 1.486655469392671e-06,
267
+ "loss": 1.1692,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.3064516129032258,
272
+ "grad_norm": 257.417236328125,
273
+ "learning_rate": 1.4841054600112084e-06,
274
+ "loss": 25.1592,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.31451612903225806,
279
+ "grad_norm": 54.77131652832031,
280
+ "learning_rate": 1.481555450629746e-06,
281
+ "loss": 0.5906,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.3225806451612903,
286
+ "grad_norm": 469.64788818359375,
287
+ "learning_rate": 1.4790054412482832e-06,
288
+ "loss": 32.1415,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.33064516129032256,
293
+ "grad_norm": 94.16633605957031,
294
+ "learning_rate": 1.4764554318668207e-06,
295
+ "loss": 1.4255,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.3387096774193548,
300
+ "grad_norm": 165.97323608398438,
301
+ "learning_rate": 1.4739054224853583e-06,
302
+ "loss": 18.915,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.3467741935483871,
307
+ "grad_norm": 165.71983337402344,
308
+ "learning_rate": 1.4713554131038957e-06,
309
+ "loss": 27.6679,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.3548387096774194,
314
+ "grad_norm": 11.058944702148438,
315
+ "learning_rate": 1.4688054037224332e-06,
316
+ "loss": 0.2848,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.3629032258064516,
321
+ "grad_norm": 55.48201370239258,
322
+ "learning_rate": 1.4662553943409704e-06,
323
+ "loss": 1.1594,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.3709677419354839,
328
+ "grad_norm": 54.40850830078125,
329
+ "learning_rate": 1.463705384959508e-06,
330
+ "loss": 2.2503,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.3790322580645161,
335
+ "grad_norm": 50.9710693359375,
336
+ "learning_rate": 1.4611553755780455e-06,
337
+ "loss": 2.6104,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.3870967741935484,
342
+ "grad_norm": 172.7392120361328,
343
+ "learning_rate": 1.4586053661965829e-06,
344
+ "loss": 2.2817,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.3951612903225806,
349
+ "grad_norm": 58.238197326660156,
350
+ "learning_rate": 1.4560553568151203e-06,
351
+ "loss": 0.9427,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.4032258064516129,
356
+ "grad_norm": 145.17626953125,
357
+ "learning_rate": 1.4535053474336577e-06,
358
+ "loss": 7.4593,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.4112903225806452,
363
+ "grad_norm": 143.7799835205078,
364
+ "learning_rate": 1.4509553380521952e-06,
365
+ "loss": 1.5799,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.41935483870967744,
370
+ "grad_norm": 13.08675479888916,
371
+ "learning_rate": 1.4484053286707326e-06,
372
+ "loss": 0.3258,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.4274193548387097,
377
+ "grad_norm": 321.4494934082031,
378
+ "learning_rate": 1.4458553192892702e-06,
379
+ "loss": 13.8006,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.43548387096774194,
384
+ "grad_norm": 72.99764251708984,
385
+ "learning_rate": 1.4433053099078077e-06,
386
+ "loss": 1.125,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.4435483870967742,
391
+ "grad_norm": 43.60110855102539,
392
+ "learning_rate": 1.4407553005263449e-06,
393
+ "loss": 4.6521,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.45161290322580644,
398
+ "grad_norm": 38.954627990722656,
399
+ "learning_rate": 1.4382052911448823e-06,
400
+ "loss": 0.8622,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.4596774193548387,
405
+ "grad_norm": 155.57080078125,
406
+ "learning_rate": 1.43565528176342e-06,
407
+ "loss": 1.2686,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.46774193548387094,
412
+ "grad_norm": 204.4175262451172,
413
+ "learning_rate": 1.4331052723819574e-06,
414
+ "loss": 7.6181,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.47580645161290325,
419
+ "grad_norm": 81.03907775878906,
420
+ "learning_rate": 1.4305552630004946e-06,
421
+ "loss": 0.7352,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.4838709677419355,
426
+ "grad_norm": 221.2462158203125,
427
+ "learning_rate": 1.4280052536190322e-06,
428
+ "loss": 14.4778,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.49193548387096775,
433
+ "grad_norm": 83.15576934814453,
434
+ "learning_rate": 1.4254552442375697e-06,
435
+ "loss": 0.6525,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.5,
440
+ "grad_norm": 185.10365295410156,
441
+ "learning_rate": 1.422905234856107e-06,
442
+ "loss": 23.472,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.5080645161290323,
447
+ "grad_norm": 116.64781188964844,
448
+ "learning_rate": 1.4203552254746445e-06,
449
+ "loss": 1.4442,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.5161290322580645,
454
+ "grad_norm": 50.69607925415039,
455
+ "learning_rate": 1.417805216093182e-06,
456
+ "loss": 2.6738,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.5241935483870968,
461
+ "grad_norm": 153.53695678710938,
462
+ "learning_rate": 1.4152552067117194e-06,
463
+ "loss": 14.6857,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.532258064516129,
468
+ "grad_norm": 59.4056396484375,
469
+ "learning_rate": 1.4127051973302568e-06,
470
+ "loss": 1.2974,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.5403225806451613,
475
+ "grad_norm": 71.67198944091797,
476
+ "learning_rate": 1.4101551879487942e-06,
477
+ "loss": 1.3162,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.5483870967741935,
482
+ "grad_norm": 126.54692077636719,
483
+ "learning_rate": 1.4076051785673319e-06,
484
+ "loss": 0.8612,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.5564516129032258,
489
+ "grad_norm": 99.8121337890625,
490
+ "learning_rate": 1.4050551691858691e-06,
491
+ "loss": 1.4869,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.5645161290322581,
496
+ "grad_norm": 56.35733413696289,
497
+ "learning_rate": 1.4025051598044065e-06,
498
+ "loss": 0.4881,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.5725806451612904,
503
+ "grad_norm": 84.86332702636719,
504
+ "learning_rate": 1.3999551504229442e-06,
505
+ "loss": 2.1241,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.5806451612903226,
510
+ "grad_norm": 251.727783203125,
511
+ "learning_rate": 1.3974051410414816e-06,
512
+ "loss": 16.5523,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.5887096774193549,
517
+ "grad_norm": 42.96049880981445,
518
+ "learning_rate": 1.3948551316600188e-06,
519
+ "loss": 0.7147,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.5967741935483871,
524
+ "grad_norm": 237.23057556152344,
525
+ "learning_rate": 1.3923051222785563e-06,
526
+ "loss": 2.0685,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.6048387096774194,
531
+ "grad_norm": 134.69094848632812,
532
+ "learning_rate": 1.3897551128970939e-06,
533
+ "loss": 16.2487,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.6129032258064516,
538
+ "grad_norm": 36.76881408691406,
539
+ "learning_rate": 1.3872051035156313e-06,
540
+ "loss": 0.2377,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.6209677419354839,
545
+ "grad_norm": 52.324462890625,
546
+ "learning_rate": 1.3846550941341688e-06,
547
+ "loss": 0.3357,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.6290322580645161,
552
+ "grad_norm": 128.68096923828125,
553
+ "learning_rate": 1.3821050847527062e-06,
554
+ "loss": 13.0358,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.6370967741935484,
559
+ "grad_norm": 215.3177490234375,
560
+ "learning_rate": 1.3795550753712436e-06,
561
+ "loss": 22.7613,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.6451612903225806,
566
+ "grad_norm": 54.24518966674805,
567
+ "learning_rate": 1.377005065989781e-06,
568
+ "loss": 1.0144,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.6532258064516129,
573
+ "grad_norm": 97.6909408569336,
574
+ "learning_rate": 1.3744550566083185e-06,
575
+ "loss": 0.7527,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.6612903225806451,
580
+ "grad_norm": 46.818058013916016,
581
+ "learning_rate": 1.3719050472268561e-06,
582
+ "loss": 2.2113,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.6693548387096774,
587
+ "grad_norm": 57.65644073486328,
588
+ "learning_rate": 1.3693550378453933e-06,
589
+ "loss": 0.2648,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.6774193548387096,
594
+ "grad_norm": 29.319469451904297,
595
+ "learning_rate": 1.3668050284639308e-06,
596
+ "loss": 1.0656,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.6854838709677419,
601
+ "grad_norm": 268.3933410644531,
602
+ "learning_rate": 1.3642550190824682e-06,
603
+ "loss": 18.3582,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.6935483870967742,
608
+ "grad_norm": 87.497314453125,
609
+ "learning_rate": 1.3617050097010058e-06,
610
+ "loss": 14.8085,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.7016129032258065,
615
+ "grad_norm": 93.05633544921875,
616
+ "learning_rate": 1.359155000319543e-06,
617
+ "loss": 2.0656,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.7096774193548387,
622
+ "grad_norm": 92.07386779785156,
623
+ "learning_rate": 1.3566049909380805e-06,
624
+ "loss": 0.9094,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.717741935483871,
629
+ "grad_norm": 119.07695007324219,
630
+ "learning_rate": 1.3540549815566181e-06,
631
+ "loss": 1.7242,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.7258064516129032,
636
+ "grad_norm": 143.6659698486328,
637
+ "learning_rate": 1.3515049721751555e-06,
638
+ "loss": 1.7576,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.7338709677419355,
643
+ "grad_norm": 609.6104736328125,
644
+ "learning_rate": 1.348954962793693e-06,
645
+ "loss": 19.9501,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.7419354838709677,
650
+ "grad_norm": 221.9893341064453,
651
+ "learning_rate": 1.3464049534122302e-06,
652
+ "loss": 7.2264,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.75,
657
+ "grad_norm": 24.371877670288086,
658
+ "learning_rate": 1.3438549440307678e-06,
659
+ "loss": 0.1521,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.7580645161290323,
664
+ "grad_norm": 110.1336669921875,
665
+ "learning_rate": 1.3413049346493053e-06,
666
+ "loss": 1.6734,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.7661290322580645,
671
+ "grad_norm": 194.0486602783203,
672
+ "learning_rate": 1.3387549252678427e-06,
673
+ "loss": 9.7306,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.7741935483870968,
678
+ "grad_norm": 110.2706298828125,
679
+ "learning_rate": 1.3362049158863801e-06,
680
+ "loss": 1.9304,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.782258064516129,
685
+ "grad_norm": 93.29808044433594,
686
+ "learning_rate": 1.3336549065049175e-06,
687
+ "loss": 0.9917,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.7903225806451613,
692
+ "grad_norm": 96.35302734375,
693
+ "learning_rate": 1.331104897123455e-06,
694
+ "loss": 1.2342,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.7983870967741935,
699
+ "grad_norm": 116.16810607910156,
700
+ "learning_rate": 1.3285548877419924e-06,
701
+ "loss": 1.5216,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.8064516129032258,
706
+ "grad_norm": 250.09878540039062,
707
+ "learning_rate": 1.32600487836053e-06,
708
+ "loss": 19.8972,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.8145161290322581,
713
+ "grad_norm": 37.814064025878906,
714
+ "learning_rate": 1.3234548689790673e-06,
715
+ "loss": 0.8875,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.8225806451612904,
720
+ "grad_norm": 72.8825454711914,
721
+ "learning_rate": 1.3209048595976047e-06,
722
+ "loss": 0.8971,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.8306451612903226,
727
+ "grad_norm": 114.7145004272461,
728
+ "learning_rate": 1.3183548502161421e-06,
729
+ "loss": 11.7635,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.8387096774193549,
734
+ "grad_norm": 22.45995330810547,
735
+ "learning_rate": 1.3158048408346798e-06,
736
+ "loss": 0.5075,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.8467741935483871,
741
+ "grad_norm": 79.19178771972656,
742
+ "learning_rate": 1.3132548314532172e-06,
743
+ "loss": 1.0136,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.8548387096774194,
748
+ "grad_norm": 129.6581268310547,
749
+ "learning_rate": 1.3107048220717544e-06,
750
+ "loss": 3.6172,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.8629032258064516,
755
+ "grad_norm": 97.2802734375,
756
+ "learning_rate": 1.308154812690292e-06,
757
+ "loss": 2.0067,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.8709677419354839,
762
+ "grad_norm": 49.05671310424805,
763
+ "learning_rate": 1.3056048033088295e-06,
764
+ "loss": 0.706,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.8790322580645161,
769
+ "grad_norm": 224.2288055419922,
770
+ "learning_rate": 1.303054793927367e-06,
771
+ "loss": 1.6531,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.8870967741935484,
776
+ "grad_norm": 18.25168800354004,
777
+ "learning_rate": 1.3005047845459043e-06,
778
+ "loss": 0.8678,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.8951612903225806,
783
+ "grad_norm": 247.68292236328125,
784
+ "learning_rate": 1.2979547751644418e-06,
785
+ "loss": 4.613,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.9032258064516129,
790
+ "grad_norm": 42.98966979980469,
791
+ "learning_rate": 1.2954047657829792e-06,
792
+ "loss": 1.1734,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.9112903225806451,
797
+ "grad_norm": 39.229766845703125,
798
+ "learning_rate": 1.2928547564015166e-06,
799
+ "loss": 0.3543,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.9193548387096774,
804
+ "grad_norm": 21.00164222717285,
805
+ "learning_rate": 1.290304747020054e-06,
806
+ "loss": 0.7768,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.9274193548387096,
811
+ "grad_norm": 179.72483825683594,
812
+ "learning_rate": 1.2877547376385917e-06,
813
+ "loss": 6.3658,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.9354838709677419,
818
+ "grad_norm": 92.51776885986328,
819
+ "learning_rate": 1.285204728257129e-06,
820
+ "loss": 1.692,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.9435483870967742,
825
+ "grad_norm": 76.68828582763672,
826
+ "learning_rate": 1.2826547188756663e-06,
827
+ "loss": 0.7759,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.9516129032258065,
832
+ "grad_norm": 146.17495727539062,
833
+ "learning_rate": 1.280104709494204e-06,
834
+ "loss": 14.4356,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.9596774193548387,
839
+ "grad_norm": 59.15159606933594,
840
+ "learning_rate": 1.2775547001127414e-06,
841
+ "loss": 1.419,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.967741935483871,
846
+ "grad_norm": 205.44473266601562,
847
+ "learning_rate": 1.2750046907312786e-06,
848
+ "loss": 1.3069,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.9758064516129032,
853
+ "grad_norm": 90.58881378173828,
854
+ "learning_rate": 1.272454681349816e-06,
855
+ "loss": 0.9966,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.9838709677419355,
860
+ "grad_norm": 19.661802291870117,
861
+ "learning_rate": 1.2699046719683537e-06,
862
+ "loss": 0.4803,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.9919354838709677,
867
+ "grad_norm": 48.44066619873047,
868
+ "learning_rate": 1.2673546625868911e-06,
869
+ "loss": 0.3767,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 1.0,
874
+ "grad_norm": 17.720691680908203,
875
+ "learning_rate": 1.2648046532054286e-06,
876
+ "loss": 0.0335,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 1.0,
881
+ "eval_loss": 3.730757236480713,
882
+ "eval_mae": 1.167590856552124,
883
+ "eval_mse": 3.730756998062134,
884
+ "eval_r2": -0.021416425704956055,
885
+ "eval_rmse": 1.9315167610098893,
886
+ "eval_runtime": 1.4549,
887
+ "eval_samples_per_second": 37.804,
888
+ "eval_smape": 46.43867909908295,
889
+ "eval_steps_per_second": 9.623,
890
+ "step": 124
891
+ },
892
+ {
893
+ "epoch": 1.0080645161290323,
894
+ "grad_norm": 44.09088897705078,
895
+ "learning_rate": 1.262254643823966e-06,
896
+ "loss": 0.8738,
897
+ "step": 125
898
+ },
899
+ {
900
+ "epoch": 1.0161290322580645,
901
+ "grad_norm": 26.77426528930664,
902
+ "learning_rate": 1.2597046344425034e-06,
903
+ "loss": 0.8001,
904
+ "step": 126
905
+ },
906
+ {
907
+ "epoch": 1.0241935483870968,
908
+ "grad_norm": 146.48257446289062,
909
+ "learning_rate": 1.2571546250610408e-06,
910
+ "loss": 6.5604,
911
+ "step": 127
912
+ },
913
+ {
914
+ "epoch": 1.032258064516129,
915
+ "grad_norm": 49.48170852661133,
916
+ "learning_rate": 1.2546046156795783e-06,
917
+ "loss": 0.6386,
918
+ "step": 128
919
+ },
920
+ {
921
+ "epoch": 1.0403225806451613,
922
+ "grad_norm": 64.10269165039062,
923
+ "learning_rate": 1.2520546062981157e-06,
924
+ "loss": 0.849,
925
+ "step": 129
926
+ },
927
+ {
928
+ "epoch": 1.0483870967741935,
929
+ "grad_norm": 43.6318244934082,
930
+ "learning_rate": 1.2495045969166531e-06,
931
+ "loss": 2.1214,
932
+ "step": 130
933
+ },
934
+ {
935
+ "epoch": 1.0564516129032258,
936
+ "grad_norm": 45.166866302490234,
937
+ "learning_rate": 1.2469545875351906e-06,
938
+ "loss": 0.2376,
939
+ "step": 131
940
+ },
941
+ {
942
+ "epoch": 1.064516129032258,
943
+ "grad_norm": 161.25682067871094,
944
+ "learning_rate": 1.244404578153728e-06,
945
+ "loss": 5.1018,
946
+ "step": 132
947
+ },
948
+ {
949
+ "epoch": 1.0725806451612903,
950
+ "grad_norm": 70.09097290039062,
951
+ "learning_rate": 1.2418545687722656e-06,
952
+ "loss": 0.7737,
953
+ "step": 133
954
+ },
955
+ {
956
+ "epoch": 1.0806451612903225,
957
+ "grad_norm": 65.81153869628906,
958
+ "learning_rate": 1.2393045593908029e-06,
959
+ "loss": 0.6568,
960
+ "step": 134
961
+ },
962
+ {
963
+ "epoch": 1.0887096774193548,
964
+ "grad_norm": 154.65948486328125,
965
+ "learning_rate": 1.2367545500093403e-06,
966
+ "loss": 7.4121,
967
+ "step": 135
968
+ },
969
+ {
970
+ "epoch": 1.096774193548387,
971
+ "grad_norm": 174.8935546875,
972
+ "learning_rate": 1.234204540627878e-06,
973
+ "loss": 16.1176,
974
+ "step": 136
975
+ },
976
+ {
977
+ "epoch": 1.1048387096774193,
978
+ "grad_norm": 271.92041015625,
979
+ "learning_rate": 1.2316545312464154e-06,
980
+ "loss": 10.3709,
981
+ "step": 137
982
+ },
983
+ {
984
+ "epoch": 1.1129032258064515,
985
+ "grad_norm": 29.178434371948242,
986
+ "learning_rate": 1.2291045218649528e-06,
987
+ "loss": 0.4082,
988
+ "step": 138
989
+ },
990
+ {
991
+ "epoch": 1.120967741935484,
992
+ "grad_norm": 404.24847412109375,
993
+ "learning_rate": 1.22655451248349e-06,
994
+ "loss": 45.207,
995
+ "step": 139
996
+ },
997
+ {
998
+ "epoch": 1.129032258064516,
999
+ "grad_norm": 23.773378372192383,
1000
+ "learning_rate": 1.2240045031020276e-06,
1001
+ "loss": 0.1074,
1002
+ "step": 140
1003
+ },
1004
+ {
1005
+ "epoch": 1.1370967741935485,
1006
+ "grad_norm": 238.07757568359375,
1007
+ "learning_rate": 1.221454493720565e-06,
1008
+ "loss": 1.2619,
1009
+ "step": 141
1010
+ },
1011
+ {
1012
+ "epoch": 1.1451612903225807,
1013
+ "grad_norm": 154.51072692871094,
1014
+ "learning_rate": 1.2189044843391025e-06,
1015
+ "loss": 1.0179,
1016
+ "step": 142
1017
+ },
1018
+ {
1019
+ "epoch": 1.153225806451613,
1020
+ "grad_norm": 33.342430114746094,
1021
+ "learning_rate": 1.21635447495764e-06,
1022
+ "loss": 0.2273,
1023
+ "step": 143
1024
+ },
1025
+ {
1026
+ "epoch": 1.1612903225806452,
1027
+ "grad_norm": 37.803348541259766,
1028
+ "learning_rate": 1.2138044655761774e-06,
1029
+ "loss": 0.2887,
1030
+ "step": 144
1031
+ },
1032
+ {
1033
+ "epoch": 1.1693548387096775,
1034
+ "grad_norm": 35.60560989379883,
1035
+ "learning_rate": 1.2112544561947148e-06,
1036
+ "loss": 0.7311,
1037
+ "step": 145
1038
+ },
1039
+ {
1040
+ "epoch": 1.1774193548387097,
1041
+ "grad_norm": 33.31494140625,
1042
+ "learning_rate": 1.2087044468132522e-06,
1043
+ "loss": 2.0214,
1044
+ "step": 146
1045
+ },
1046
+ {
1047
+ "epoch": 1.185483870967742,
1048
+ "grad_norm": 41.15532684326172,
1049
+ "learning_rate": 1.2061544374317896e-06,
1050
+ "loss": 0.4238,
1051
+ "step": 147
1052
+ },
1053
+ {
1054
+ "epoch": 1.1935483870967742,
1055
+ "grad_norm": 49.68434143066406,
1056
+ "learning_rate": 1.203604428050327e-06,
1057
+ "loss": 1.3357,
1058
+ "step": 148
1059
+ },
1060
+ {
1061
+ "epoch": 1.2016129032258065,
1062
+ "grad_norm": 325.6404724121094,
1063
+ "learning_rate": 1.2010544186688645e-06,
1064
+ "loss": 20.1013,
1065
+ "step": 149
1066
+ },
1067
+ {
1068
+ "epoch": 1.2096774193548387,
1069
+ "grad_norm": 388.65625,
1070
+ "learning_rate": 1.198504409287402e-06,
1071
+ "loss": 36.6465,
1072
+ "step": 150
1073
+ },
1074
+ {
1075
+ "epoch": 1.217741935483871,
1076
+ "grad_norm": 231.11331176757812,
1077
+ "learning_rate": 1.1959543999059396e-06,
1078
+ "loss": 17.261,
1079
+ "step": 151
1080
+ },
1081
+ {
1082
+ "epoch": 1.2258064516129032,
1083
+ "grad_norm": 112.41427612304688,
1084
+ "learning_rate": 1.193404390524477e-06,
1085
+ "loss": 1.7195,
1086
+ "step": 152
1087
+ },
1088
+ {
1089
+ "epoch": 1.2338709677419355,
1090
+ "grad_norm": 167.3661651611328,
1091
+ "learning_rate": 1.1908543811430142e-06,
1092
+ "loss": 10.1658,
1093
+ "step": 153
1094
+ },
1095
+ {
1096
+ "epoch": 1.2419354838709677,
1097
+ "grad_norm": 121.19483947753906,
1098
+ "learning_rate": 1.1883043717615519e-06,
1099
+ "loss": 12.4887,
1100
+ "step": 154
1101
+ },
1102
+ {
1103
+ "epoch": 1.25,
1104
+ "grad_norm": 59.64546203613281,
1105
+ "learning_rate": 1.1857543623800893e-06,
1106
+ "loss": 0.6071,
1107
+ "step": 155
1108
+ },
1109
+ {
1110
+ "epoch": 1.2580645161290323,
1111
+ "grad_norm": 75.89230346679688,
1112
+ "learning_rate": 1.1832043529986267e-06,
1113
+ "loss": 0.9969,
1114
+ "step": 156
1115
+ },
1116
+ {
1117
+ "epoch": 1.2661290322580645,
1118
+ "grad_norm": 126.55144500732422,
1119
+ "learning_rate": 1.1806543436171641e-06,
1120
+ "loss": 3.966,
1121
+ "step": 157
1122
+ },
1123
+ {
1124
+ "epoch": 1.2741935483870968,
1125
+ "grad_norm": 183.3189697265625,
1126
+ "learning_rate": 1.1781043342357016e-06,
1127
+ "loss": 13.638,
1128
+ "step": 158
1129
+ },
1130
+ {
1131
+ "epoch": 1.282258064516129,
1132
+ "grad_norm": 340.3012390136719,
1133
+ "learning_rate": 1.175554324854239e-06,
1134
+ "loss": 23.4134,
1135
+ "step": 159
1136
+ },
1137
+ {
1138
+ "epoch": 1.2903225806451613,
1139
+ "grad_norm": 147.78733825683594,
1140
+ "learning_rate": 1.1730043154727764e-06,
1141
+ "loss": 10.5212,
1142
+ "step": 160
1143
+ },
1144
+ {
1145
+ "epoch": 1.2983870967741935,
1146
+ "grad_norm": 409.5651550292969,
1147
+ "learning_rate": 1.1704543060913139e-06,
1148
+ "loss": 39.5865,
1149
+ "step": 161
1150
+ },
1151
+ {
1152
+ "epoch": 1.3064516129032258,
1153
+ "grad_norm": 66.4163589477539,
1154
+ "learning_rate": 1.1679042967098515e-06,
1155
+ "loss": 1.1522,
1156
+ "step": 162
1157
+ },
1158
+ {
1159
+ "epoch": 1.314516129032258,
1160
+ "grad_norm": 195.2645263671875,
1161
+ "learning_rate": 1.1653542873283887e-06,
1162
+ "loss": 0.9955,
1163
+ "step": 163
1164
+ },
1165
+ {
1166
+ "epoch": 1.3225806451612903,
1167
+ "grad_norm": 172.76194763183594,
1168
+ "learning_rate": 1.1628042779469261e-06,
1169
+ "loss": 2.4719,
1170
+ "step": 164
1171
+ },
1172
+ {
1173
+ "epoch": 1.3306451612903225,
1174
+ "grad_norm": 278.2360534667969,
1175
+ "learning_rate": 1.1602542685654636e-06,
1176
+ "loss": 3.492,
1177
+ "step": 165
1178
+ },
1179
+ {
1180
+ "epoch": 1.3387096774193548,
1181
+ "grad_norm": 178.6942901611328,
1182
+ "learning_rate": 1.1577042591840012e-06,
1183
+ "loss": 2.5787,
1184
+ "step": 166
1185
+ },
1186
+ {
1187
+ "epoch": 1.346774193548387,
1188
+ "grad_norm": 137.12525939941406,
1189
+ "learning_rate": 1.1551542498025384e-06,
1190
+ "loss": 2.5417,
1191
+ "step": 167
1192
+ },
1193
+ {
1194
+ "epoch": 1.3548387096774195,
1195
+ "grad_norm": 120.3616943359375,
1196
+ "learning_rate": 1.1526042404210759e-06,
1197
+ "loss": 0.9734,
1198
+ "step": 168
1199
+ },
1200
+ {
1201
+ "epoch": 1.3629032258064515,
1202
+ "grad_norm": 179.4165496826172,
1203
+ "learning_rate": 1.1500542310396135e-06,
1204
+ "loss": 2.6588,
1205
+ "step": 169
1206
+ },
1207
+ {
1208
+ "epoch": 1.370967741935484,
1209
+ "grad_norm": 473.0636901855469,
1210
+ "learning_rate": 1.147504221658151e-06,
1211
+ "loss": 127.6287,
1212
+ "step": 170
1213
+ },
1214
+ {
1215
+ "epoch": 1.379032258064516,
1216
+ "grad_norm": 83.53874206542969,
1217
+ "learning_rate": 1.1449542122766884e-06,
1218
+ "loss": 1.2538,
1219
+ "step": 171
1220
+ },
1221
+ {
1222
+ "epoch": 1.3870967741935485,
1223
+ "grad_norm": 312.453857421875,
1224
+ "learning_rate": 1.1424042028952258e-06,
1225
+ "loss": 7.3581,
1226
+ "step": 172
1227
+ },
1228
+ {
1229
+ "epoch": 1.3951612903225805,
1230
+ "grad_norm": 150.91253662109375,
1231
+ "learning_rate": 1.1398541935137632e-06,
1232
+ "loss": 2.3145,
1233
+ "step": 173
1234
+ },
1235
+ {
1236
+ "epoch": 1.403225806451613,
1237
+ "grad_norm": 119.02738189697266,
1238
+ "learning_rate": 1.1373041841323007e-06,
1239
+ "loss": 8.6639,
1240
+ "step": 174
1241
+ },
1242
+ {
1243
+ "epoch": 1.4112903225806452,
1244
+ "grad_norm": 70.6019058227539,
1245
+ "learning_rate": 1.134754174750838e-06,
1246
+ "loss": 0.5583,
1247
+ "step": 175
1248
+ },
1249
+ {
1250
+ "epoch": 1.4193548387096775,
1251
+ "grad_norm": 333.658447265625,
1252
+ "learning_rate": 1.1322041653693755e-06,
1253
+ "loss": 18.0422,
1254
+ "step": 176
1255
+ },
1256
+ {
1257
+ "epoch": 1.4274193548387097,
1258
+ "grad_norm": 123.302001953125,
1259
+ "learning_rate": 1.129654155987913e-06,
1260
+ "loss": 0.8602,
1261
+ "step": 177
1262
+ },
1263
+ {
1264
+ "epoch": 1.435483870967742,
1265
+ "grad_norm": 139.1856689453125,
1266
+ "learning_rate": 1.1271041466064504e-06,
1267
+ "loss": 1.7481,
1268
+ "step": 178
1269
+ },
1270
+ {
1271
+ "epoch": 1.4435483870967742,
1272
+ "grad_norm": 137.33912658691406,
1273
+ "learning_rate": 1.1245541372249878e-06,
1274
+ "loss": 2.8533,
1275
+ "step": 179
1276
+ },
1277
+ {
1278
+ "epoch": 1.4516129032258065,
1279
+ "grad_norm": 187.89089965820312,
1280
+ "learning_rate": 1.1220041278435254e-06,
1281
+ "loss": 1.8385,
1282
+ "step": 180
1283
+ },
1284
+ {
1285
+ "epoch": 1.4596774193548387,
1286
+ "grad_norm": 130.551513671875,
1287
+ "learning_rate": 1.1194541184620627e-06,
1288
+ "loss": 1.3891,
1289
+ "step": 181
1290
+ },
1291
+ {
1292
+ "epoch": 1.467741935483871,
1293
+ "grad_norm": 66.99901580810547,
1294
+ "learning_rate": 1.1169041090806e-06,
1295
+ "loss": 1.1031,
1296
+ "step": 182
1297
+ },
1298
+ {
1299
+ "epoch": 1.4758064516129032,
1300
+ "grad_norm": 89.26786804199219,
1301
+ "learning_rate": 1.1143540996991375e-06,
1302
+ "loss": 0.9218,
1303
+ "step": 183
1304
+ },
1305
+ {
1306
+ "epoch": 1.4838709677419355,
1307
+ "grad_norm": 101.36397552490234,
1308
+ "learning_rate": 1.1118040903176752e-06,
1309
+ "loss": 3.3235,
1310
+ "step": 184
1311
+ },
1312
+ {
1313
+ "epoch": 1.4919354838709677,
1314
+ "grad_norm": 115.42027282714844,
1315
+ "learning_rate": 1.1092540809362126e-06,
1316
+ "loss": 1.6457,
1317
+ "step": 185
1318
+ },
1319
+ {
1320
+ "epoch": 1.5,
1321
+ "grad_norm": 113.05801391601562,
1322
+ "learning_rate": 1.1067040715547498e-06,
1323
+ "loss": 0.7229,
1324
+ "step": 186
1325
+ },
1326
+ {
1327
+ "epoch": 1.5080645161290323,
1328
+ "grad_norm": 151.0050506591797,
1329
+ "learning_rate": 1.1041540621732874e-06,
1330
+ "loss": 4.1861,
1331
+ "step": 187
1332
+ },
1333
+ {
1334
+ "epoch": 1.5161290322580645,
1335
+ "grad_norm": 47.2061653137207,
1336
+ "learning_rate": 1.1016040527918249e-06,
1337
+ "loss": 1.9876,
1338
+ "step": 188
1339
+ },
1340
+ {
1341
+ "epoch": 1.5241935483870968,
1342
+ "grad_norm": 74.51702117919922,
1343
+ "learning_rate": 1.0990540434103623e-06,
1344
+ "loss": 1.3335,
1345
+ "step": 189
1346
+ },
1347
+ {
1348
+ "epoch": 1.532258064516129,
1349
+ "grad_norm": 90.15420532226562,
1350
+ "learning_rate": 1.0965040340288997e-06,
1351
+ "loss": 1.1029,
1352
+ "step": 190
1353
+ },
1354
+ {
1355
+ "epoch": 1.5403225806451613,
1356
+ "grad_norm": 41.844879150390625,
1357
+ "learning_rate": 1.0939540246474372e-06,
1358
+ "loss": 0.8871,
1359
+ "step": 191
1360
+ },
1361
+ {
1362
+ "epoch": 1.5483870967741935,
1363
+ "grad_norm": 34.21913528442383,
1364
+ "learning_rate": 1.0914040152659746e-06,
1365
+ "loss": 0.5249,
1366
+ "step": 192
1367
+ },
1368
+ {
1369
+ "epoch": 1.5564516129032258,
1370
+ "grad_norm": 39.128990173339844,
1371
+ "learning_rate": 1.088854005884512e-06,
1372
+ "loss": 0.7401,
1373
+ "step": 193
1374
+ },
1375
+ {
1376
+ "epoch": 1.564516129032258,
1377
+ "grad_norm": 65.44004821777344,
1378
+ "learning_rate": 1.0863039965030494e-06,
1379
+ "loss": 0.7682,
1380
+ "step": 194
1381
+ },
1382
+ {
1383
+ "epoch": 1.5725806451612905,
1384
+ "grad_norm": 58.968135833740234,
1385
+ "learning_rate": 1.0837539871215869e-06,
1386
+ "loss": 0.6109,
1387
+ "step": 195
1388
+ },
1389
+ {
1390
+ "epoch": 1.5806451612903225,
1391
+ "grad_norm": 233.93258666992188,
1392
+ "learning_rate": 1.0812039777401243e-06,
1393
+ "loss": 9.1722,
1394
+ "step": 196
1395
+ },
1396
+ {
1397
+ "epoch": 1.588709677419355,
1398
+ "grad_norm": 42.01539993286133,
1399
+ "learning_rate": 1.0786539683586617e-06,
1400
+ "loss": 1.6803,
1401
+ "step": 197
1402
+ },
1403
+ {
1404
+ "epoch": 1.596774193548387,
1405
+ "grad_norm": 36.658546447753906,
1406
+ "learning_rate": 1.0761039589771994e-06,
1407
+ "loss": 0.5086,
1408
+ "step": 198
1409
+ },
1410
+ {
1411
+ "epoch": 1.6048387096774195,
1412
+ "grad_norm": 61.448612213134766,
1413
+ "learning_rate": 1.0735539495957368e-06,
1414
+ "loss": 0.3062,
1415
+ "step": 199
1416
+ },
1417
+ {
1418
+ "epoch": 1.6129032258064515,
1419
+ "grad_norm": 100.31672668457031,
1420
+ "learning_rate": 1.071003940214274e-06,
1421
+ "loss": 0.7721,
1422
+ "step": 200
1423
+ },
1424
+ {
1425
+ "epoch": 1.620967741935484,
1426
+ "grad_norm": 7.334116458892822,
1427
+ "learning_rate": 1.0684539308328115e-06,
1428
+ "loss": 0.2008,
1429
+ "step": 201
1430
+ },
1431
+ {
1432
+ "epoch": 1.629032258064516,
1433
+ "grad_norm": 65.1685562133789,
1434
+ "learning_rate": 1.065903921451349e-06,
1435
+ "loss": 1.615,
1436
+ "step": 202
1437
+ },
1438
+ {
1439
+ "epoch": 1.6370967741935485,
1440
+ "grad_norm": 15.108301162719727,
1441
+ "learning_rate": 1.0633539120698865e-06,
1442
+ "loss": 0.123,
1443
+ "step": 203
1444
+ },
1445
+ {
1446
+ "epoch": 1.6451612903225805,
1447
+ "grad_norm": 58.7624626159668,
1448
+ "learning_rate": 1.060803902688424e-06,
1449
+ "loss": 0.663,
1450
+ "step": 204
1451
+ },
1452
+ {
1453
+ "epoch": 1.653225806451613,
1454
+ "grad_norm": 216.7863311767578,
1455
+ "learning_rate": 1.0582538933069614e-06,
1456
+ "loss": 16.203,
1457
+ "step": 205
1458
+ },
1459
+ {
1460
+ "epoch": 1.661290322580645,
1461
+ "grad_norm": 97.55227661132812,
1462
+ "learning_rate": 1.0557038839254988e-06,
1463
+ "loss": 0.9618,
1464
+ "step": 206
1465
+ },
1466
+ {
1467
+ "epoch": 1.6693548387096775,
1468
+ "grad_norm": 35.726932525634766,
1469
+ "learning_rate": 1.0531538745440362e-06,
1470
+ "loss": 0.727,
1471
+ "step": 207
1472
+ },
1473
+ {
1474
+ "epoch": 1.6774193548387095,
1475
+ "grad_norm": 31.328706741333008,
1476
+ "learning_rate": 1.0506038651625737e-06,
1477
+ "loss": 0.3076,
1478
+ "step": 208
1479
+ },
1480
+ {
1481
+ "epoch": 1.685483870967742,
1482
+ "grad_norm": 19.492021560668945,
1483
+ "learning_rate": 1.0480538557811113e-06,
1484
+ "loss": 0.3841,
1485
+ "step": 209
1486
+ },
1487
+ {
1488
+ "epoch": 1.6935483870967742,
1489
+ "grad_norm": 33.60645294189453,
1490
+ "learning_rate": 1.0455038463996485e-06,
1491
+ "loss": 0.2718,
1492
+ "step": 210
1493
+ },
1494
+ {
1495
+ "epoch": 1.7016129032258065,
1496
+ "grad_norm": 157.64622497558594,
1497
+ "learning_rate": 1.042953837018186e-06,
1498
+ "loss": 3.0856,
1499
+ "step": 211
1500
+ },
1501
+ {
1502
+ "epoch": 1.7096774193548387,
1503
+ "grad_norm": 42.80052185058594,
1504
+ "learning_rate": 1.0404038276367234e-06,
1505
+ "loss": 0.9122,
1506
+ "step": 212
1507
+ },
1508
+ {
1509
+ "epoch": 1.717741935483871,
1510
+ "grad_norm": 19.76460075378418,
1511
+ "learning_rate": 1.037853818255261e-06,
1512
+ "loss": 0.1621,
1513
+ "step": 213
1514
+ },
1515
+ {
1516
+ "epoch": 1.7258064516129032,
1517
+ "grad_norm": 286.8505859375,
1518
+ "learning_rate": 1.0353038088737982e-06,
1519
+ "loss": 25.7877,
1520
+ "step": 214
1521
+ },
1522
+ {
1523
+ "epoch": 1.7338709677419355,
1524
+ "grad_norm": 203.42111206054688,
1525
+ "learning_rate": 1.0327537994923357e-06,
1526
+ "loss": 16.5754,
1527
+ "step": 215
1528
+ },
1529
+ {
1530
+ "epoch": 1.7419354838709677,
1531
+ "grad_norm": 60.32425308227539,
1532
+ "learning_rate": 1.0302037901108733e-06,
1533
+ "loss": 1.282,
1534
+ "step": 216
1535
+ },
1536
+ {
1537
+ "epoch": 1.75,
1538
+ "grad_norm": 308.719970703125,
1539
+ "learning_rate": 1.0276537807294107e-06,
1540
+ "loss": 8.6198,
1541
+ "step": 217
1542
+ },
1543
+ {
1544
+ "epoch": 1.7580645161290323,
1545
+ "grad_norm": 14.96123218536377,
1546
+ "learning_rate": 1.0251037713479482e-06,
1547
+ "loss": 0.108,
1548
+ "step": 218
1549
+ },
1550
+ {
1551
+ "epoch": 1.7661290322580645,
1552
+ "grad_norm": 363.8222351074219,
1553
+ "learning_rate": 1.0225537619664854e-06,
1554
+ "loss": 17.4903,
1555
+ "step": 219
1556
+ },
1557
+ {
1558
+ "epoch": 1.7741935483870968,
1559
+ "grad_norm": 19.47015953063965,
1560
+ "learning_rate": 1.020003752585023e-06,
1561
+ "loss": 0.1757,
1562
+ "step": 220
1563
+ },
1564
+ {
1565
+ "epoch": 1.782258064516129,
1566
+ "grad_norm": 39.350669860839844,
1567
+ "learning_rate": 1.0174537432035605e-06,
1568
+ "loss": 0.2219,
1569
+ "step": 221
1570
+ },
1571
+ {
1572
+ "epoch": 1.7903225806451613,
1573
+ "grad_norm": 85.70793914794922,
1574
+ "learning_rate": 1.0149037338220979e-06,
1575
+ "loss": 0.6299,
1576
+ "step": 222
1577
+ },
1578
+ {
1579
+ "epoch": 1.7983870967741935,
1580
+ "grad_norm": 19.446735382080078,
1581
+ "learning_rate": 1.0123537244406353e-06,
1582
+ "loss": 0.1134,
1583
+ "step": 223
1584
+ },
1585
+ {
1586
+ "epoch": 1.8064516129032258,
1587
+ "grad_norm": 281.6679992675781,
1588
+ "learning_rate": 1.0098037150591727e-06,
1589
+ "loss": 19.1561,
1590
+ "step": 224
1591
+ },
1592
+ {
1593
+ "epoch": 1.814516129032258,
1594
+ "grad_norm": 294.34197998046875,
1595
+ "learning_rate": 1.0072537056777102e-06,
1596
+ "loss": 9.8189,
1597
+ "step": 225
1598
+ },
1599
+ {
1600
+ "epoch": 1.8225806451612905,
1601
+ "grad_norm": 32.63911437988281,
1602
+ "learning_rate": 1.0047036962962476e-06,
1603
+ "loss": 1.851,
1604
+ "step": 226
1605
+ },
1606
+ {
1607
+ "epoch": 1.8306451612903225,
1608
+ "grad_norm": 43.1181526184082,
1609
+ "learning_rate": 1.0021536869147852e-06,
1610
+ "loss": 0.4436,
1611
+ "step": 227
1612
+ },
1613
+ {
1614
+ "epoch": 1.838709677419355,
1615
+ "grad_norm": 57.709442138671875,
1616
+ "learning_rate": 9.996036775333225e-07,
1617
+ "loss": 0.8785,
1618
+ "step": 228
1619
+ },
1620
+ {
1621
+ "epoch": 1.846774193548387,
1622
+ "grad_norm": 57.25394821166992,
1623
+ "learning_rate": 9.970536681518599e-07,
1624
+ "loss": 0.5494,
1625
+ "step": 229
1626
+ },
1627
+ {
1628
+ "epoch": 1.8548387096774195,
1629
+ "grad_norm": 202.76467895507812,
1630
+ "learning_rate": 9.945036587703973e-07,
1631
+ "loss": 7.7274,
1632
+ "step": 230
1633
+ },
1634
+ {
1635
+ "epoch": 1.8629032258064515,
1636
+ "grad_norm": 103.01250457763672,
1637
+ "learning_rate": 9.91953649388935e-07,
1638
+ "loss": 1.1239,
1639
+ "step": 231
1640
+ },
1641
+ {
1642
+ "epoch": 1.870967741935484,
1643
+ "grad_norm": 32.47471618652344,
1644
+ "learning_rate": 9.894036400074724e-07,
1645
+ "loss": 0.8574,
1646
+ "step": 232
1647
+ },
1648
+ {
1649
+ "epoch": 1.879032258064516,
1650
+ "grad_norm": 58.068180084228516,
1651
+ "learning_rate": 9.868536306260096e-07,
1652
+ "loss": 0.6567,
1653
+ "step": 233
1654
+ },
1655
+ {
1656
+ "epoch": 1.8870967741935485,
1657
+ "grad_norm": 24.15645980834961,
1658
+ "learning_rate": 9.843036212445473e-07,
1659
+ "loss": 0.3775,
1660
+ "step": 234
1661
+ },
1662
+ {
1663
+ "epoch": 1.8951612903225805,
1664
+ "grad_norm": 28.187562942504883,
1665
+ "learning_rate": 9.817536118630847e-07,
1666
+ "loss": 0.2793,
1667
+ "step": 235
1668
+ },
1669
+ {
1670
+ "epoch": 1.903225806451613,
1671
+ "grad_norm": 238.00563049316406,
1672
+ "learning_rate": 9.792036024816221e-07,
1673
+ "loss": 10.1773,
1674
+ "step": 236
1675
+ },
1676
+ {
1677
+ "epoch": 1.911290322580645,
1678
+ "grad_norm": 79.65827941894531,
1679
+ "learning_rate": 9.766535931001595e-07,
1680
+ "loss": 6.1575,
1681
+ "step": 237
1682
+ },
1683
+ {
1684
+ "epoch": 1.9193548387096775,
1685
+ "grad_norm": 39.00093460083008,
1686
+ "learning_rate": 9.74103583718697e-07,
1687
+ "loss": 0.7295,
1688
+ "step": 238
1689
+ },
1690
+ {
1691
+ "epoch": 1.9274193548387095,
1692
+ "grad_norm": 44.81210708618164,
1693
+ "learning_rate": 9.715535743372344e-07,
1694
+ "loss": 0.6221,
1695
+ "step": 239
1696
+ },
1697
+ {
1698
+ "epoch": 1.935483870967742,
1699
+ "grad_norm": 109.5570068359375,
1700
+ "learning_rate": 9.690035649557718e-07,
1701
+ "loss": 2.6696,
1702
+ "step": 240
1703
+ },
1704
+ {
1705
+ "epoch": 1.9435483870967742,
1706
+ "grad_norm": 38.722591400146484,
1707
+ "learning_rate": 9.664535555743093e-07,
1708
+ "loss": 0.8759,
1709
+ "step": 241
1710
+ },
1711
+ {
1712
+ "epoch": 1.9516129032258065,
1713
+ "grad_norm": 42.692073822021484,
1714
+ "learning_rate": 9.639035461928467e-07,
1715
+ "loss": 1.5391,
1716
+ "step": 242
1717
+ },
1718
+ {
1719
+ "epoch": 1.9596774193548387,
1720
+ "grad_norm": 149.17979431152344,
1721
+ "learning_rate": 9.613535368113841e-07,
1722
+ "loss": 2.4909,
1723
+ "step": 243
1724
+ },
1725
+ {
1726
+ "epoch": 1.967741935483871,
1727
+ "grad_norm": 113.84545135498047,
1728
+ "learning_rate": 9.588035274299215e-07,
1729
+ "loss": 3.2092,
1730
+ "step": 244
1731
+ },
1732
+ {
1733
+ "epoch": 1.9758064516129032,
1734
+ "grad_norm": 73.42545318603516,
1735
+ "learning_rate": 9.562535180484592e-07,
1736
+ "loss": 0.8856,
1737
+ "step": 245
1738
+ },
1739
+ {
1740
+ "epoch": 1.9838709677419355,
1741
+ "grad_norm": 234.1503448486328,
1742
+ "learning_rate": 9.537035086669966e-07,
1743
+ "loss": 3.4449,
1744
+ "step": 246
1745
+ },
1746
+ {
1747
+ "epoch": 1.9919354838709677,
1748
+ "grad_norm": 163.1854705810547,
1749
+ "learning_rate": 9.511534992855338e-07,
1750
+ "loss": 1.3792,
1751
+ "step": 247
1752
+ },
1753
+ {
1754
+ "epoch": 2.0,
1755
+ "grad_norm": 199.5084228515625,
1756
+ "learning_rate": 9.486034899040714e-07,
1757
+ "loss": 3.0211,
1758
+ "step": 248
1759
+ },
1760
+ {
1761
+ "epoch": 2.0,
1762
+ "eval_loss": 3.5324182510375977,
1763
+ "eval_mae": 1.171051025390625,
1764
+ "eval_mse": 3.5324196815490723,
1765
+ "eval_r2": 0.032884955406188965,
1766
+ "eval_rmse": 1.8794732457657044,
1767
+ "eval_runtime": 1.4104,
1768
+ "eval_samples_per_second": 38.996,
1769
+ "eval_smape": 48.2732355594635,
1770
+ "eval_steps_per_second": 9.926,
1771
+ "step": 248
1772
+ }
1773
+ ],
1774
+ "logging_steps": 1,
1775
+ "max_steps": 620,
1776
+ "num_input_tokens_seen": 0,
1777
+ "num_train_epochs": 5,
1778
+ "save_steps": 500,
1779
+ "stateful_callbacks": {
1780
+ "TrainerControl": {
1781
+ "args": {
1782
+ "should_epoch_stop": false,
1783
+ "should_evaluate": false,
1784
+ "should_log": false,
1785
+ "should_save": true,
1786
+ "should_training_stop": false
1787
+ },
1788
+ "attributes": {}
1789
+ }
1790
+ },
1791
+ "total_flos": 1343939627507712.0,
1792
+ "train_batch_size": 4,
1793
+ "trial_name": null,
1794
+ "trial_params": {
1795
+ "learning_rate": 1.5810058165067856e-06
1796
+ }
1797
+ }
run-1/checkpoint-248/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c75f52d1f03874f3ea77755ee6f5645b32f15ed0540249d11f68829799fbf1ad
3
+ size 5432
run-2/checkpoint-496/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-base",
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": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0"
25
+ },
26
+ "initializer_cutoff_factor": 2.0,
27
+ "initializer_range": 0.02,
28
+ "intermediate_size": 1152,
29
+ "label2id": {
30
+ "LABEL_0": 0
31
+ },
32
+ "layer_norm_eps": 1e-05,
33
+ "local_attention": 128,
34
+ "local_rope_theta": 10000.0,
35
+ "max_position_embeddings": 8192,
36
+ "mlp_bias": false,
37
+ "mlp_dropout": 0.0,
38
+ "model_type": "modernbert",
39
+ "norm_bias": false,
40
+ "norm_eps": 1e-05,
41
+ "num_attention_heads": 12,
42
+ "num_hidden_layers": 22,
43
+ "pad_token_id": 50283,
44
+ "position_embedding_type": "absolute",
45
+ "problem_type": "regression",
46
+ "reference_compile": true,
47
+ "sep_token_id": 50282,
48
+ "sparse_pred_ignore_index": -100,
49
+ "sparse_prediction": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.48.0.dev0",
52
+ "vocab_size": 50368
53
+ }
run-2/checkpoint-496/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:700a9e0547bf342b129b703d14213c4398639a0838f36475c66313517dcc3900
3
+ size 598436708
run-2/checkpoint-496/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47905aff0b3011b97ca95ed35acb7858570262060dc0e83b3890e0078e1a1578
3
+ size 1196963194
run-2/checkpoint-496/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f88d13d80dce0f9736996fbfe6064188892cdba5964429232e474be0fcc2dcd6
3
+ size 14244
run-2/checkpoint-496/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0592d84c37a102269bbdd672f9b318fe845ae14aa7bc582aac054d41f127379
3
+ size 1064
run-2/checkpoint-496/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": true,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
run-2/checkpoint-496/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-2/checkpoint-496/tokenizer_config.json ADDED
@@ -0,0 +1,945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "|||IP_ADDRESS|||",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1": {
12
+ "content": "<|padding|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "50254": {
20
+ "content": " ",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": false
26
+ },
27
+ "50255": {
28
+ "content": " ",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": false
34
+ },
35
+ "50256": {
36
+ "content": " ",
37
+ "lstrip": false,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": false
42
+ },
43
+ "50257": {
44
+ "content": " ",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ },
51
+ "50258": {
52
+ "content": " ",
53
+ "lstrip": false,
54
+ "normalized": true,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": false
58
+ },
59
+ "50259": {
60
+ "content": " ",
61
+ "lstrip": false,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": false
66
+ },
67
+ "50260": {
68
+ "content": " ",
69
+ "lstrip": false,
70
+ "normalized": true,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": false
74
+ },
75
+ "50261": {
76
+ "content": " ",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "50262": {
84
+ "content": " ",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "50263": {
92
+ "content": " ",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "50264": {
100
+ "content": " ",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "50265": {
108
+ "content": " ",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "50266": {
116
+ "content": " ",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "50267": {
124
+ "content": " ",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "50268": {
132
+ "content": " ",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "50269": {
140
+ "content": " ",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "50270": {
148
+ "content": " ",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "50271": {
156
+ "content": " ",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "50272": {
164
+ "content": " ",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "50273": {
172
+ "content": " ",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "50274": {
180
+ "content": " ",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "50275": {
188
+ "content": " ",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "50276": {
196
+ "content": " ",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "50277": {
204
+ "content": "|||EMAIL_ADDRESS|||",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "50278": {
212
+ "content": "|||PHONE_NUMBER|||",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "50279": {
220
+ "content": "<|endoftext|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "50280": {
228
+ "content": "[UNK]",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "50281": {
236
+ "content": "[CLS]",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "50282": {
244
+ "content": "[SEP]",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "50283": {
252
+ "content": "[PAD]",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "50284": {
260
+ "content": "[MASK]",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "50285": {
268
+ "content": "[unused0]",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "50286": {
276
+ "content": "[unused1]",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "50287": {
284
+ "content": "[unused2]",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "50288": {
292
+ "content": "[unused3]",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "50289": {
300
+ "content": "[unused4]",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "50290": {
308
+ "content": "[unused5]",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "50291": {
316
+ "content": "[unused6]",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "50292": {
324
+ "content": "[unused7]",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "50293": {
332
+ "content": "[unused8]",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "50294": {
340
+ "content": "[unused9]",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "50295": {
348
+ "content": "[unused10]",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "50296": {
356
+ "content": "[unused11]",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "50297": {
364
+ "content": "[unused12]",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "50298": {
372
+ "content": "[unused13]",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "50299": {
380
+ "content": "[unused14]",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "50300": {
388
+ "content": "[unused15]",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "50301": {
396
+ "content": "[unused16]",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "50302": {
404
+ "content": "[unused17]",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "50303": {
412
+ "content": "[unused18]",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "50304": {
420
+ "content": "[unused19]",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "50305": {
428
+ "content": "[unused20]",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "50306": {
436
+ "content": "[unused21]",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "50307": {
444
+ "content": "[unused22]",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "50308": {
452
+ "content": "[unused23]",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "50309": {
460
+ "content": "[unused24]",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "50310": {
468
+ "content": "[unused25]",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "50311": {
476
+ "content": "[unused26]",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "50312": {
484
+ "content": "[unused27]",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "50313": {
492
+ "content": "[unused28]",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "50314": {
500
+ "content": "[unused29]",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "50315": {
508
+ "content": "[unused30]",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "50316": {
516
+ "content": "[unused31]",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "50317": {
524
+ "content": "[unused32]",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "50318": {
532
+ "content": "[unused33]",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "50319": {
540
+ "content": "[unused34]",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "50320": {
548
+ "content": "[unused35]",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "50321": {
556
+ "content": "[unused36]",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "50322": {
564
+ "content": "[unused37]",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "50323": {
572
+ "content": "[unused38]",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "50324": {
580
+ "content": "[unused39]",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "50325": {
588
+ "content": "[unused40]",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "50326": {
596
+ "content": "[unused41]",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "50327": {
604
+ "content": "[unused42]",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "50328": {
612
+ "content": "[unused43]",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "50329": {
620
+ "content": "[unused44]",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "50330": {
628
+ "content": "[unused45]",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "50331": {
636
+ "content": "[unused46]",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "50332": {
644
+ "content": "[unused47]",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "50333": {
652
+ "content": "[unused48]",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "50334": {
660
+ "content": "[unused49]",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "50335": {
668
+ "content": "[unused50]",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "50336": {
676
+ "content": "[unused51]",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "50337": {
684
+ "content": "[unused52]",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "50338": {
692
+ "content": "[unused53]",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "50339": {
700
+ "content": "[unused54]",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "50340": {
708
+ "content": "[unused55]",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "50341": {
716
+ "content": "[unused56]",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "50342": {
724
+ "content": "[unused57]",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "50343": {
732
+ "content": "[unused58]",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "50344": {
740
+ "content": "[unused59]",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "50345": {
748
+ "content": "[unused60]",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "50346": {
756
+ "content": "[unused61]",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "50347": {
764
+ "content": "[unused62]",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "50348": {
772
+ "content": "[unused63]",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "50349": {
780
+ "content": "[unused64]",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "50350": {
788
+ "content": "[unused65]",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "50351": {
796
+ "content": "[unused66]",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "50352": {
804
+ "content": "[unused67]",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "50353": {
812
+ "content": "[unused68]",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "50354": {
820
+ "content": "[unused69]",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "50355": {
828
+ "content": "[unused70]",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "50356": {
836
+ "content": "[unused71]",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "50357": {
844
+ "content": "[unused72]",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "50358": {
852
+ "content": "[unused73]",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "50359": {
860
+ "content": "[unused74]",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "50360": {
868
+ "content": "[unused75]",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "50361": {
876
+ "content": "[unused76]",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "50362": {
884
+ "content": "[unused77]",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "50363": {
892
+ "content": "[unused78]",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "50364": {
900
+ "content": "[unused79]",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "50365": {
908
+ "content": "[unused80]",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "50366": {
916
+ "content": "[unused81]",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "50367": {
924
+ "content": "[unused82]",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ }
931
+ },
932
+ "clean_up_tokenization_spaces": true,
933
+ "cls_token": "[CLS]",
934
+ "extra_special_tokens": {},
935
+ "mask_token": "[MASK]",
936
+ "model_input_names": [
937
+ "input_ids",
938
+ "attention_mask"
939
+ ],
940
+ "model_max_length": 2048,
941
+ "pad_token": "[PAD]",
942
+ "sep_token": "[SEP]",
943
+ "tokenizer_class": "PreTrainedTokenizerFast",
944
+ "unk_token": "[UNK]"
945
+ }
run-2/checkpoint-496/trainer_state.json ADDED
@@ -0,0 +1,3559 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.5342012108859804,
3
+ "best_model_checkpoint": "./modernBERT-content-regression/run-2/checkpoint-496",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 496,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.008064516129032258,
13
+ "grad_norm": 312.814453125,
14
+ "learning_rate": 2.4759427123128026e-05,
15
+ "loss": 21.3087,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.016129032258064516,
20
+ "grad_norm": 73.38264465332031,
21
+ "learning_rate": 2.4719428048615702e-05,
22
+ "loss": 0.7206,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.024193548387096774,
27
+ "grad_norm": 363.2410583496094,
28
+ "learning_rate": 2.467942897410338e-05,
29
+ "loss": 135.449,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.03225806451612903,
34
+ "grad_norm": 64.86761474609375,
35
+ "learning_rate": 2.4639429899591056e-05,
36
+ "loss": 12.5636,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.04032258064516129,
41
+ "grad_norm": 86.92879486083984,
42
+ "learning_rate": 2.4599430825078733e-05,
43
+ "loss": 2.4774,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.04838709677419355,
48
+ "grad_norm": 271.8641662597656,
49
+ "learning_rate": 2.455943175056641e-05,
50
+ "loss": 3.729,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.056451612903225805,
55
+ "grad_norm": 41.483768463134766,
56
+ "learning_rate": 2.4519432676054087e-05,
57
+ "loss": 5.4716,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.06451612903225806,
62
+ "grad_norm": 163.63929748535156,
63
+ "learning_rate": 2.4479433601541764e-05,
64
+ "loss": 11.4921,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.07258064516129033,
69
+ "grad_norm": 67.34258270263672,
70
+ "learning_rate": 2.443943452702944e-05,
71
+ "loss": 0.827,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.08064516129032258,
76
+ "grad_norm": 45.18867874145508,
77
+ "learning_rate": 2.4399435452517118e-05,
78
+ "loss": 2.3716,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.08870967741935484,
83
+ "grad_norm": 130.92532348632812,
84
+ "learning_rate": 2.4359436378004795e-05,
85
+ "loss": 10.3088,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.0967741935483871,
90
+ "grad_norm": 34.51155471801758,
91
+ "learning_rate": 2.4319437303492468e-05,
92
+ "loss": 0.7379,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.10483870967741936,
97
+ "grad_norm": 109.77779388427734,
98
+ "learning_rate": 2.427943822898015e-05,
99
+ "loss": 1.3991,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.11290322580645161,
104
+ "grad_norm": 33.590213775634766,
105
+ "learning_rate": 2.4239439154467826e-05,
106
+ "loss": 1.3007,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.12096774193548387,
111
+ "grad_norm": 34.60383605957031,
112
+ "learning_rate": 2.4199440079955502e-05,
113
+ "loss": 0.1627,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.12903225806451613,
118
+ "grad_norm": 183.3640899658203,
119
+ "learning_rate": 2.415944100544318e-05,
120
+ "loss": 9.0763,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.13709677419354838,
125
+ "grad_norm": 27.07301139831543,
126
+ "learning_rate": 2.4119441930930853e-05,
127
+ "loss": 1.9223,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.14516129032258066,
132
+ "grad_norm": 106.09356689453125,
133
+ "learning_rate": 2.4079442856418533e-05,
134
+ "loss": 2.1224,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.1532258064516129,
139
+ "grad_norm": 32.14213562011719,
140
+ "learning_rate": 2.403944378190621e-05,
141
+ "loss": 0.84,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.16129032258064516,
146
+ "grad_norm": 186.9098358154297,
147
+ "learning_rate": 2.3999444707393887e-05,
148
+ "loss": 6.1026,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.1693548387096774,
153
+ "grad_norm": 252.75877380371094,
154
+ "learning_rate": 2.3959445632881564e-05,
155
+ "loss": 4.5225,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.1774193548387097,
160
+ "grad_norm": 577.6574096679688,
161
+ "learning_rate": 2.3919446558369238e-05,
162
+ "loss": 13.0971,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.18548387096774194,
167
+ "grad_norm": 236.11961364746094,
168
+ "learning_rate": 2.3879447483856914e-05,
169
+ "loss": 12.425,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.1935483870967742,
174
+ "grad_norm": 329.27984619140625,
175
+ "learning_rate": 2.3839448409344595e-05,
176
+ "loss": 7.6775,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.20161290322580644,
181
+ "grad_norm": 100.30821990966797,
182
+ "learning_rate": 2.379944933483227e-05,
183
+ "loss": 1.4652,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.20967741935483872,
188
+ "grad_norm": 96.21698760986328,
189
+ "learning_rate": 2.3759450260319945e-05,
190
+ "loss": 1.4142,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.21774193548387097,
195
+ "grad_norm": 31.13923454284668,
196
+ "learning_rate": 2.3719451185807622e-05,
197
+ "loss": 1.5587,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.22580645161290322,
202
+ "grad_norm": 37.34580612182617,
203
+ "learning_rate": 2.36794521112953e-05,
204
+ "loss": 0.6072,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.23387096774193547,
209
+ "grad_norm": 50.031856536865234,
210
+ "learning_rate": 2.363945303678298e-05,
211
+ "loss": 1.2987,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.24193548387096775,
216
+ "grad_norm": 65.12112426757812,
217
+ "learning_rate": 2.3599453962270656e-05,
218
+ "loss": 1.5842,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.25,
223
+ "grad_norm": 64.7094497680664,
224
+ "learning_rate": 2.355945488775833e-05,
225
+ "loss": 2.2444,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.25806451612903225,
230
+ "grad_norm": 85.2149429321289,
231
+ "learning_rate": 2.3519455813246007e-05,
232
+ "loss": 2.9294,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.2661290322580645,
237
+ "grad_norm": 20.915496826171875,
238
+ "learning_rate": 2.3479456738733684e-05,
239
+ "loss": 0.6207,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.27419354838709675,
244
+ "grad_norm": 54.965335845947266,
245
+ "learning_rate": 2.3439457664221364e-05,
246
+ "loss": 1.9851,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.28225806451612906,
251
+ "grad_norm": 187.736328125,
252
+ "learning_rate": 2.339945858970904e-05,
253
+ "loss": 21.4765,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.2903225806451613,
258
+ "grad_norm": 65.61473846435547,
259
+ "learning_rate": 2.3359459515196714e-05,
260
+ "loss": 1.4577,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.29838709677419356,
265
+ "grad_norm": 32.118228912353516,
266
+ "learning_rate": 2.331946044068439e-05,
267
+ "loss": 0.4661,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.3064516129032258,
272
+ "grad_norm": 78.08635711669922,
273
+ "learning_rate": 2.3279461366172068e-05,
274
+ "loss": 19.0058,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.31451612903225806,
279
+ "grad_norm": 153.27804565429688,
280
+ "learning_rate": 2.3239462291659745e-05,
281
+ "loss": 5.0193,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.3225806451612903,
286
+ "grad_norm": 205.57205200195312,
287
+ "learning_rate": 2.3199463217147422e-05,
288
+ "loss": 17.3633,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.33064516129032256,
293
+ "grad_norm": 207.53005981445312,
294
+ "learning_rate": 2.31594641426351e-05,
295
+ "loss": 7.6631,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.3387096774193548,
300
+ "grad_norm": 85.1321029663086,
301
+ "learning_rate": 2.3119465068122776e-05,
302
+ "loss": 19.0139,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.3467741935483871,
307
+ "grad_norm": 133.19154357910156,
308
+ "learning_rate": 2.3079465993610453e-05,
309
+ "loss": 21.4597,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.3548387096774194,
314
+ "grad_norm": 116.96082305908203,
315
+ "learning_rate": 2.303946691909813e-05,
316
+ "loss": 3.4332,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.3629032258064516,
321
+ "grad_norm": 99.06409454345703,
322
+ "learning_rate": 2.2999467844585807e-05,
323
+ "loss": 3.114,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.3709677419354839,
328
+ "grad_norm": 58.03582763671875,
329
+ "learning_rate": 2.2959468770073484e-05,
330
+ "loss": 2.1934,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.3790322580645161,
335
+ "grad_norm": 66.2879409790039,
336
+ "learning_rate": 2.291946969556116e-05,
337
+ "loss": 3.4978,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.3870967741935484,
342
+ "grad_norm": 84.04444122314453,
343
+ "learning_rate": 2.2879470621048837e-05,
344
+ "loss": 2.5483,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.3951612903225806,
349
+ "grad_norm": 40.6508903503418,
350
+ "learning_rate": 2.2839471546536514e-05,
351
+ "loss": 0.8297,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.4032258064516129,
356
+ "grad_norm": 75.82645416259766,
357
+ "learning_rate": 2.279947247202419e-05,
358
+ "loss": 8.1934,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.4112903225806452,
363
+ "grad_norm": 24.05048179626465,
364
+ "learning_rate": 2.2759473397511868e-05,
365
+ "loss": 0.3819,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.41935483870967744,
370
+ "grad_norm": 39.44225311279297,
371
+ "learning_rate": 2.2719474322999545e-05,
372
+ "loss": 1.1629,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.4274193548387097,
377
+ "grad_norm": 195.91845703125,
378
+ "learning_rate": 2.2679475248487222e-05,
379
+ "loss": 19.0751,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.43548387096774194,
384
+ "grad_norm": 10.670635223388672,
385
+ "learning_rate": 2.26394761739749e-05,
386
+ "loss": 0.5451,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.4435483870967742,
391
+ "grad_norm": 45.76976013183594,
392
+ "learning_rate": 2.2599477099462576e-05,
393
+ "loss": 5.5405,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.45161290322580644,
398
+ "grad_norm": 31.259849548339844,
399
+ "learning_rate": 2.2559478024950253e-05,
400
+ "loss": 1.616,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.4596774193548387,
405
+ "grad_norm": 16.430635452270508,
406
+ "learning_rate": 2.251947895043793e-05,
407
+ "loss": 0.2542,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.46774193548387094,
412
+ "grad_norm": 83.4207992553711,
413
+ "learning_rate": 2.2479479875925607e-05,
414
+ "loss": 9.8378,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.47580645161290325,
419
+ "grad_norm": 18.212440490722656,
420
+ "learning_rate": 2.2439480801413284e-05,
421
+ "loss": 0.2447,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.4838709677419355,
426
+ "grad_norm": 107.16078186035156,
427
+ "learning_rate": 2.239948172690096e-05,
428
+ "loss": 14.317,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.49193548387096775,
433
+ "grad_norm": 32.10540771484375,
434
+ "learning_rate": 2.2359482652388637e-05,
435
+ "loss": 0.6299,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.5,
440
+ "grad_norm": 73.37329864501953,
441
+ "learning_rate": 2.2319483577876314e-05,
442
+ "loss": 21.8458,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.5080645161290323,
447
+ "grad_norm": 46.12656784057617,
448
+ "learning_rate": 2.227948450336399e-05,
449
+ "loss": 1.2991,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.5161290322580645,
454
+ "grad_norm": 18.842960357666016,
455
+ "learning_rate": 2.2239485428851665e-05,
456
+ "loss": 2.414,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.5241935483870968,
461
+ "grad_norm": 77.6194839477539,
462
+ "learning_rate": 2.2199486354339345e-05,
463
+ "loss": 13.3138,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.532258064516129,
468
+ "grad_norm": 66.23066711425781,
469
+ "learning_rate": 2.2159487279827022e-05,
470
+ "loss": 1.6535,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.5403225806451613,
475
+ "grad_norm": 37.77677917480469,
476
+ "learning_rate": 2.21194882053147e-05,
477
+ "loss": 1.8361,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.5483870967741935,
482
+ "grad_norm": 46.24187088012695,
483
+ "learning_rate": 2.2079489130802376e-05,
484
+ "loss": 1.4393,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.5564516129032258,
489
+ "grad_norm": 80.01097106933594,
490
+ "learning_rate": 2.203949005629005e-05,
491
+ "loss": 3.5612,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.5645161290322581,
496
+ "grad_norm": 7.726450443267822,
497
+ "learning_rate": 2.199949098177773e-05,
498
+ "loss": 0.3842,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.5725806451612904,
503
+ "grad_norm": 13.420072555541992,
504
+ "learning_rate": 2.1959491907265407e-05,
505
+ "loss": 2.0311,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.5806451612903226,
510
+ "grad_norm": 81.66263580322266,
511
+ "learning_rate": 2.1919492832753084e-05,
512
+ "loss": 15.1423,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.5887096774193549,
517
+ "grad_norm": 30.341995239257812,
518
+ "learning_rate": 2.1879493758240757e-05,
519
+ "loss": 0.4867,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.5967741935483871,
524
+ "grad_norm": 83.49819946289062,
525
+ "learning_rate": 2.1839494683728434e-05,
526
+ "loss": 3.1285,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.6048387096774194,
531
+ "grad_norm": 62.19274139404297,
532
+ "learning_rate": 2.1799495609216114e-05,
533
+ "loss": 6.7166,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.6129032258064516,
538
+ "grad_norm": 25.30666732788086,
539
+ "learning_rate": 2.175949653470379e-05,
540
+ "loss": 0.4676,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.6209677419354839,
545
+ "grad_norm": 33.63331604003906,
546
+ "learning_rate": 2.1719497460191468e-05,
547
+ "loss": 0.7665,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.6290322580645161,
552
+ "grad_norm": 42.89381408691406,
553
+ "learning_rate": 2.1679498385679142e-05,
554
+ "loss": 12.7414,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.6370967741935484,
559
+ "grad_norm": 87.96309661865234,
560
+ "learning_rate": 2.163949931116682e-05,
561
+ "loss": 23.3716,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.6451612903225806,
566
+ "grad_norm": 11.2321195602417,
567
+ "learning_rate": 2.1599500236654496e-05,
568
+ "loss": 1.0413,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.6532258064516129,
573
+ "grad_norm": 21.426607131958008,
574
+ "learning_rate": 2.1559501162142176e-05,
575
+ "loss": 0.4798,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.6612903225806451,
580
+ "grad_norm": 26.169578552246094,
581
+ "learning_rate": 2.1519502087629853e-05,
582
+ "loss": 1.9313,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.6693548387096774,
587
+ "grad_norm": 18.651702880859375,
588
+ "learning_rate": 2.1479503013117526e-05,
589
+ "loss": 0.3472,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.6774193548387096,
594
+ "grad_norm": 7.624361038208008,
595
+ "learning_rate": 2.1439503938605203e-05,
596
+ "loss": 0.6932,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.6854838709677419,
601
+ "grad_norm": 123.84193420410156,
602
+ "learning_rate": 2.139950486409288e-05,
603
+ "loss": 17.2375,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.6935483870967742,
608
+ "grad_norm": 39.71210861206055,
609
+ "learning_rate": 2.135950578958056e-05,
610
+ "loss": 14.5549,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.7016129032258065,
615
+ "grad_norm": 23.699077606201172,
616
+ "learning_rate": 2.1319506715068234e-05,
617
+ "loss": 1.417,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.7096774193548387,
622
+ "grad_norm": 49.412071228027344,
623
+ "learning_rate": 2.127950764055591e-05,
624
+ "loss": 1.2961,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.717741935483871,
629
+ "grad_norm": 60.343544006347656,
630
+ "learning_rate": 2.1239508566043588e-05,
631
+ "loss": 2.5778,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.7258064516129032,
636
+ "grad_norm": 62.71953582763672,
637
+ "learning_rate": 2.1199509491531265e-05,
638
+ "loss": 2.0629,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.7338709677419355,
643
+ "grad_norm": 120.01128387451172,
644
+ "learning_rate": 2.1159510417018945e-05,
645
+ "loss": 19.2536,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.7419354838709677,
650
+ "grad_norm": 59.287296295166016,
651
+ "learning_rate": 2.111951134250662e-05,
652
+ "loss": 5.4721,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.75,
657
+ "grad_norm": 11.04751968383789,
658
+ "learning_rate": 2.1079512267994296e-05,
659
+ "loss": 0.1843,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.7580645161290323,
664
+ "grad_norm": 101.02423858642578,
665
+ "learning_rate": 2.1039513193481973e-05,
666
+ "loss": 5.4431,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.7661290322580645,
671
+ "grad_norm": 72.17565155029297,
672
+ "learning_rate": 2.099951411896965e-05,
673
+ "loss": 9.7386,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.7741935483870968,
678
+ "grad_norm": 83.0063705444336,
679
+ "learning_rate": 2.0959515044457326e-05,
680
+ "loss": 3.1414,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.782258064516129,
685
+ "grad_norm": 47.760589599609375,
686
+ "learning_rate": 2.0919515969945003e-05,
687
+ "loss": 1.2319,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.7903225806451613,
692
+ "grad_norm": 43.6202507019043,
693
+ "learning_rate": 2.087951689543268e-05,
694
+ "loss": 1.2002,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.7983870967741935,
699
+ "grad_norm": 42.899200439453125,
700
+ "learning_rate": 2.0839517820920357e-05,
701
+ "loss": 1.1685,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.8064516129032258,
706
+ "grad_norm": 116.83399963378906,
707
+ "learning_rate": 2.0799518746408034e-05,
708
+ "loss": 19.3838,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.8145161290322581,
713
+ "grad_norm": 16.667503356933594,
714
+ "learning_rate": 2.075951967189571e-05,
715
+ "loss": 0.8709,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.8225806451612904,
720
+ "grad_norm": 13.620025634765625,
721
+ "learning_rate": 2.0719520597383388e-05,
722
+ "loss": 0.3991,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.8306451612903226,
727
+ "grad_norm": 71.85466766357422,
728
+ "learning_rate": 2.0679521522871065e-05,
729
+ "loss": 11.6404,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.8387096774193549,
734
+ "grad_norm": 11.629712104797363,
735
+ "learning_rate": 2.0639522448358742e-05,
736
+ "loss": 0.9058,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.8467741935483871,
741
+ "grad_norm": 33.48215866088867,
742
+ "learning_rate": 2.059952337384642e-05,
743
+ "loss": 0.6932,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.8548387096774194,
748
+ "grad_norm": 47.76714324951172,
749
+ "learning_rate": 2.0559524299334096e-05,
750
+ "loss": 2.1864,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.8629032258064516,
755
+ "grad_norm": 40.71377944946289,
756
+ "learning_rate": 2.0519525224821773e-05,
757
+ "loss": 2.2803,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.8709677419354839,
762
+ "grad_norm": 40.90349578857422,
763
+ "learning_rate": 2.047952615030945e-05,
764
+ "loss": 0.7261,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.8790322580645161,
769
+ "grad_norm": 61.77216720581055,
770
+ "learning_rate": 2.0439527075797126e-05,
771
+ "loss": 1.9658,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.8870967741935484,
776
+ "grad_norm": 15.168683052062988,
777
+ "learning_rate": 2.0399528001284803e-05,
778
+ "loss": 0.6779,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.8951612903225806,
783
+ "grad_norm": 77.2820816040039,
784
+ "learning_rate": 2.035952892677248e-05,
785
+ "loss": 4.008,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.9032258064516129,
790
+ "grad_norm": 9.020164489746094,
791
+ "learning_rate": 2.0319529852260157e-05,
792
+ "loss": 0.8001,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.9112903225806451,
797
+ "grad_norm": 11.099763870239258,
798
+ "learning_rate": 2.0279530777747834e-05,
799
+ "loss": 0.4358,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.9193548387096774,
804
+ "grad_norm": 39.36742401123047,
805
+ "learning_rate": 2.023953170323551e-05,
806
+ "loss": 1.39,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.9274193548387096,
811
+ "grad_norm": 66.89869689941406,
812
+ "learning_rate": 2.0199532628723188e-05,
813
+ "loss": 4.703,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.9354838709677419,
818
+ "grad_norm": 25.90433692932129,
819
+ "learning_rate": 2.0159533554210865e-05,
820
+ "loss": 1.4303,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.9435483870967742,
825
+ "grad_norm": 21.590999603271484,
826
+ "learning_rate": 2.0119534479698542e-05,
827
+ "loss": 0.3786,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.9516129032258065,
832
+ "grad_norm": 68.37796020507812,
833
+ "learning_rate": 2.007953540518622e-05,
834
+ "loss": 14.1292,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.9596774193548387,
839
+ "grad_norm": 18.72194480895996,
840
+ "learning_rate": 2.0039536330673896e-05,
841
+ "loss": 2.0311,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.967741935483871,
846
+ "grad_norm": 15.883110046386719,
847
+ "learning_rate": 1.999953725616157e-05,
848
+ "loss": 0.8801,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.9758064516129032,
853
+ "grad_norm": 18.568222045898438,
854
+ "learning_rate": 1.9959538181649246e-05,
855
+ "loss": 0.403,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.9838709677419355,
860
+ "grad_norm": 17.32382583618164,
861
+ "learning_rate": 1.9919539107136926e-05,
862
+ "loss": 0.3303,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.9919354838709677,
867
+ "grad_norm": 10.257896423339844,
868
+ "learning_rate": 1.9879540032624603e-05,
869
+ "loss": 0.2042,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 1.0,
874
+ "grad_norm": 25.707752227783203,
875
+ "learning_rate": 1.983954095811228e-05,
876
+ "loss": 0.3115,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 1.0,
881
+ "eval_loss": 4.090590000152588,
882
+ "eval_mae": 1.2299772500991821,
883
+ "eval_mse": 4.090590000152588,
884
+ "eval_r2": -0.11993241310119629,
885
+ "eval_rmse": 2.0225207045052933,
886
+ "eval_runtime": 1.3441,
887
+ "eval_samples_per_second": 40.92,
888
+ "eval_smape": 50.51450729370117,
889
+ "eval_steps_per_second": 10.416,
890
+ "step": 124
891
+ },
892
+ {
893
+ "epoch": 1.0080645161290323,
894
+ "grad_norm": 37.24449920654297,
895
+ "learning_rate": 1.9799541883599954e-05,
896
+ "loss": 0.6422,
897
+ "step": 125
898
+ },
899
+ {
900
+ "epoch": 1.0161290322580645,
901
+ "grad_norm": 32.547119140625,
902
+ "learning_rate": 1.975954280908763e-05,
903
+ "loss": 0.9869,
904
+ "step": 126
905
+ },
906
+ {
907
+ "epoch": 1.0241935483870968,
908
+ "grad_norm": 70.05367279052734,
909
+ "learning_rate": 1.971954373457531e-05,
910
+ "loss": 4.6137,
911
+ "step": 127
912
+ },
913
+ {
914
+ "epoch": 1.032258064516129,
915
+ "grad_norm": 18.101661682128906,
916
+ "learning_rate": 1.9679544660062988e-05,
917
+ "loss": 0.4156,
918
+ "step": 128
919
+ },
920
+ {
921
+ "epoch": 1.0403225806451613,
922
+ "grad_norm": 22.2724609375,
923
+ "learning_rate": 1.9639545585550665e-05,
924
+ "loss": 0.841,
925
+ "step": 129
926
+ },
927
+ {
928
+ "epoch": 1.0483870967741935,
929
+ "grad_norm": 13.428308486938477,
930
+ "learning_rate": 1.959954651103834e-05,
931
+ "loss": 2.2846,
932
+ "step": 130
933
+ },
934
+ {
935
+ "epoch": 1.0564516129032258,
936
+ "grad_norm": 35.4150505065918,
937
+ "learning_rate": 1.9559547436526015e-05,
938
+ "loss": 0.7014,
939
+ "step": 131
940
+ },
941
+ {
942
+ "epoch": 1.064516129032258,
943
+ "grad_norm": 31.687740325927734,
944
+ "learning_rate": 1.9519548362013696e-05,
945
+ "loss": 3.4412,
946
+ "step": 132
947
+ },
948
+ {
949
+ "epoch": 1.0725806451612903,
950
+ "grad_norm": 54.95043182373047,
951
+ "learning_rate": 1.9479549287501373e-05,
952
+ "loss": 1.5723,
953
+ "step": 133
954
+ },
955
+ {
956
+ "epoch": 1.0806451612903225,
957
+ "grad_norm": 43.67780303955078,
958
+ "learning_rate": 1.9439550212989046e-05,
959
+ "loss": 1.161,
960
+ "step": 134
961
+ },
962
+ {
963
+ "epoch": 1.0887096774193548,
964
+ "grad_norm": 47.5596809387207,
965
+ "learning_rate": 1.9399551138476723e-05,
966
+ "loss": 5.3313,
967
+ "step": 135
968
+ },
969
+ {
970
+ "epoch": 1.096774193548387,
971
+ "grad_norm": 71.37387084960938,
972
+ "learning_rate": 1.93595520639644e-05,
973
+ "loss": 12.8636,
974
+ "step": 136
975
+ },
976
+ {
977
+ "epoch": 1.1048387096774193,
978
+ "grad_norm": 80.62567901611328,
979
+ "learning_rate": 1.9319552989452077e-05,
980
+ "loss": 7.3218,
981
+ "step": 137
982
+ },
983
+ {
984
+ "epoch": 1.1129032258064515,
985
+ "grad_norm": 59.973960876464844,
986
+ "learning_rate": 1.9279553914939757e-05,
987
+ "loss": 1.3948,
988
+ "step": 138
989
+ },
990
+ {
991
+ "epoch": 1.120967741935484,
992
+ "grad_norm": 179.83071899414062,
993
+ "learning_rate": 1.923955484042743e-05,
994
+ "loss": 34.7136,
995
+ "step": 139
996
+ },
997
+ {
998
+ "epoch": 1.129032258064516,
999
+ "grad_norm": 84.4344482421875,
1000
+ "learning_rate": 1.9199555765915108e-05,
1001
+ "loss": 2.435,
1002
+ "step": 140
1003
+ },
1004
+ {
1005
+ "epoch": 1.1370967741935485,
1006
+ "grad_norm": 65.75166320800781,
1007
+ "learning_rate": 1.9159556691402785e-05,
1008
+ "loss": 2.0554,
1009
+ "step": 141
1010
+ },
1011
+ {
1012
+ "epoch": 1.1451612903225807,
1013
+ "grad_norm": 29.691560745239258,
1014
+ "learning_rate": 1.911955761689046e-05,
1015
+ "loss": 0.8479,
1016
+ "step": 142
1017
+ },
1018
+ {
1019
+ "epoch": 1.153225806451613,
1020
+ "grad_norm": 28.317535400390625,
1021
+ "learning_rate": 1.9079558542378142e-05,
1022
+ "loss": 0.4651,
1023
+ "step": 143
1024
+ },
1025
+ {
1026
+ "epoch": 1.1612903225806452,
1027
+ "grad_norm": 28.46364402770996,
1028
+ "learning_rate": 1.9039559467865815e-05,
1029
+ "loss": 0.5143,
1030
+ "step": 144
1031
+ },
1032
+ {
1033
+ "epoch": 1.1693548387096775,
1034
+ "grad_norm": 14.189237594604492,
1035
+ "learning_rate": 1.8999560393353492e-05,
1036
+ "loss": 0.2308,
1037
+ "step": 145
1038
+ },
1039
+ {
1040
+ "epoch": 1.1774193548387097,
1041
+ "grad_norm": 16.224409103393555,
1042
+ "learning_rate": 1.895956131884117e-05,
1043
+ "loss": 1.8195,
1044
+ "step": 146
1045
+ },
1046
+ {
1047
+ "epoch": 1.185483870967742,
1048
+ "grad_norm": 25.36056900024414,
1049
+ "learning_rate": 1.8919562244328846e-05,
1050
+ "loss": 0.6262,
1051
+ "step": 147
1052
+ },
1053
+ {
1054
+ "epoch": 1.1935483870967742,
1055
+ "grad_norm": 20.145959854125977,
1056
+ "learning_rate": 1.8879563169816523e-05,
1057
+ "loss": 0.7281,
1058
+ "step": 148
1059
+ },
1060
+ {
1061
+ "epoch": 1.2016129032258065,
1062
+ "grad_norm": 159.1549072265625,
1063
+ "learning_rate": 1.88395640953042e-05,
1064
+ "loss": 11.0343,
1065
+ "step": 149
1066
+ },
1067
+ {
1068
+ "epoch": 1.2096774193548387,
1069
+ "grad_norm": 162.7301788330078,
1070
+ "learning_rate": 1.8799565020791877e-05,
1071
+ "loss": 34.9038,
1072
+ "step": 150
1073
+ },
1074
+ {
1075
+ "epoch": 1.217741935483871,
1076
+ "grad_norm": 86.77201080322266,
1077
+ "learning_rate": 1.8759565946279554e-05,
1078
+ "loss": 16.0319,
1079
+ "step": 151
1080
+ },
1081
+ {
1082
+ "epoch": 1.2258064516129032,
1083
+ "grad_norm": 48.960357666015625,
1084
+ "learning_rate": 1.871956687176723e-05,
1085
+ "loss": 1.4731,
1086
+ "step": 152
1087
+ },
1088
+ {
1089
+ "epoch": 1.2338709677419355,
1090
+ "grad_norm": 82.25511932373047,
1091
+ "learning_rate": 1.8679567797254908e-05,
1092
+ "loss": 7.8374,
1093
+ "step": 153
1094
+ },
1095
+ {
1096
+ "epoch": 1.2419354838709677,
1097
+ "grad_norm": 60.74580764770508,
1098
+ "learning_rate": 1.8639568722742585e-05,
1099
+ "loss": 10.0002,
1100
+ "step": 154
1101
+ },
1102
+ {
1103
+ "epoch": 1.25,
1104
+ "grad_norm": 34.497222900390625,
1105
+ "learning_rate": 1.859956964823026e-05,
1106
+ "loss": 0.6464,
1107
+ "step": 155
1108
+ },
1109
+ {
1110
+ "epoch": 1.2580645161290323,
1111
+ "grad_norm": 38.092491149902344,
1112
+ "learning_rate": 1.855957057371794e-05,
1113
+ "loss": 1.5106,
1114
+ "step": 156
1115
+ },
1116
+ {
1117
+ "epoch": 1.2661290322580645,
1118
+ "grad_norm": 19.8438663482666,
1119
+ "learning_rate": 1.8519571499205615e-05,
1120
+ "loss": 1.5936,
1121
+ "step": 157
1122
+ },
1123
+ {
1124
+ "epoch": 1.2741935483870968,
1125
+ "grad_norm": 69.82159423828125,
1126
+ "learning_rate": 1.8479572424693292e-05,
1127
+ "loss": 10.4774,
1128
+ "step": 158
1129
+ },
1130
+ {
1131
+ "epoch": 1.282258064516129,
1132
+ "grad_norm": 125.10799407958984,
1133
+ "learning_rate": 1.843957335018097e-05,
1134
+ "loss": 15.5645,
1135
+ "step": 159
1136
+ },
1137
+ {
1138
+ "epoch": 1.2903225806451613,
1139
+ "grad_norm": 49.95842361450195,
1140
+ "learning_rate": 1.8399574275668646e-05,
1141
+ "loss": 7.8997,
1142
+ "step": 160
1143
+ },
1144
+ {
1145
+ "epoch": 1.2983870967741935,
1146
+ "grad_norm": 187.53903198242188,
1147
+ "learning_rate": 1.8359575201156323e-05,
1148
+ "loss": 19.5171,
1149
+ "step": 161
1150
+ },
1151
+ {
1152
+ "epoch": 1.3064516129032258,
1153
+ "grad_norm": 137.54644775390625,
1154
+ "learning_rate": 1.8319576126644e-05,
1155
+ "loss": 4.1746,
1156
+ "step": 162
1157
+ },
1158
+ {
1159
+ "epoch": 1.314516129032258,
1160
+ "grad_norm": 116.5649185180664,
1161
+ "learning_rate": 1.8279577052131677e-05,
1162
+ "loss": 2.607,
1163
+ "step": 163
1164
+ },
1165
+ {
1166
+ "epoch": 1.3225806451612903,
1167
+ "grad_norm": 95.47283935546875,
1168
+ "learning_rate": 1.8239577977619354e-05,
1169
+ "loss": 5.3841,
1170
+ "step": 164
1171
+ },
1172
+ {
1173
+ "epoch": 1.3306451612903225,
1174
+ "grad_norm": 354.6362609863281,
1175
+ "learning_rate": 1.819957890310703e-05,
1176
+ "loss": 13.093,
1177
+ "step": 165
1178
+ },
1179
+ {
1180
+ "epoch": 1.3387096774193548,
1181
+ "grad_norm": 140.9376220703125,
1182
+ "learning_rate": 1.8159579828594708e-05,
1183
+ "loss": 4.0244,
1184
+ "step": 166
1185
+ },
1186
+ {
1187
+ "epoch": 1.346774193548387,
1188
+ "grad_norm": 94.6503677368164,
1189
+ "learning_rate": 1.811958075408238e-05,
1190
+ "loss": 1.2288,
1191
+ "step": 167
1192
+ },
1193
+ {
1194
+ "epoch": 1.3548387096774195,
1195
+ "grad_norm": 25.267553329467773,
1196
+ "learning_rate": 1.807958167957006e-05,
1197
+ "loss": 0.5603,
1198
+ "step": 168
1199
+ },
1200
+ {
1201
+ "epoch": 1.3629032258064515,
1202
+ "grad_norm": 49.03358840942383,
1203
+ "learning_rate": 1.803958260505774e-05,
1204
+ "loss": 1.6284,
1205
+ "step": 169
1206
+ },
1207
+ {
1208
+ "epoch": 1.370967741935484,
1209
+ "grad_norm": 369.3719787597656,
1210
+ "learning_rate": 1.7999583530545415e-05,
1211
+ "loss": 127.8047,
1212
+ "step": 170
1213
+ },
1214
+ {
1215
+ "epoch": 1.379032258064516,
1216
+ "grad_norm": 30.973657608032227,
1217
+ "learning_rate": 1.7959584456033092e-05,
1218
+ "loss": 0.6907,
1219
+ "step": 171
1220
+ },
1221
+ {
1222
+ "epoch": 1.3870967741935485,
1223
+ "grad_norm": 163.9529266357422,
1224
+ "learning_rate": 1.7919585381520766e-05,
1225
+ "loss": 15.7879,
1226
+ "step": 172
1227
+ },
1228
+ {
1229
+ "epoch": 1.3951612903225805,
1230
+ "grad_norm": 13.607197761535645,
1231
+ "learning_rate": 1.7879586307008446e-05,
1232
+ "loss": 0.3138,
1233
+ "step": 173
1234
+ },
1235
+ {
1236
+ "epoch": 1.403225806451613,
1237
+ "grad_norm": 86.09737396240234,
1238
+ "learning_rate": 1.7839587232496123e-05,
1239
+ "loss": 9.6102,
1240
+ "step": 174
1241
+ },
1242
+ {
1243
+ "epoch": 1.4112903225806452,
1244
+ "grad_norm": 54.644554138183594,
1245
+ "learning_rate": 1.77995881579838e-05,
1246
+ "loss": 1.5776,
1247
+ "step": 175
1248
+ },
1249
+ {
1250
+ "epoch": 1.4193548387096775,
1251
+ "grad_norm": 190.9008331298828,
1252
+ "learning_rate": 1.7759589083471477e-05,
1253
+ "loss": 22.1159,
1254
+ "step": 176
1255
+ },
1256
+ {
1257
+ "epoch": 1.4274193548387097,
1258
+ "grad_norm": 50.92055130004883,
1259
+ "learning_rate": 1.771959000895915e-05,
1260
+ "loss": 2.3412,
1261
+ "step": 177
1262
+ },
1263
+ {
1264
+ "epoch": 1.435483870967742,
1265
+ "grad_norm": 38.6739387512207,
1266
+ "learning_rate": 1.7679590934446827e-05,
1267
+ "loss": 0.7342,
1268
+ "step": 178
1269
+ },
1270
+ {
1271
+ "epoch": 1.4435483870967742,
1272
+ "grad_norm": 28.455352783203125,
1273
+ "learning_rate": 1.7639591859934508e-05,
1274
+ "loss": 1.7024,
1275
+ "step": 179
1276
+ },
1277
+ {
1278
+ "epoch": 1.4516129032258065,
1279
+ "grad_norm": 29.965150833129883,
1280
+ "learning_rate": 1.7599592785422185e-05,
1281
+ "loss": 0.6395,
1282
+ "step": 180
1283
+ },
1284
+ {
1285
+ "epoch": 1.4596774193548387,
1286
+ "grad_norm": 46.12895202636719,
1287
+ "learning_rate": 1.7559593710909858e-05,
1288
+ "loss": 0.8689,
1289
+ "step": 181
1290
+ },
1291
+ {
1292
+ "epoch": 1.467741935483871,
1293
+ "grad_norm": 21.45783805847168,
1294
+ "learning_rate": 1.7519594636397535e-05,
1295
+ "loss": 0.8418,
1296
+ "step": 182
1297
+ },
1298
+ {
1299
+ "epoch": 1.4758064516129032,
1300
+ "grad_norm": 24.071561813354492,
1301
+ "learning_rate": 1.7479595561885212e-05,
1302
+ "loss": 1.8008,
1303
+ "step": 183
1304
+ },
1305
+ {
1306
+ "epoch": 1.4838709677419355,
1307
+ "grad_norm": 35.17493438720703,
1308
+ "learning_rate": 1.7439596487372892e-05,
1309
+ "loss": 3.9543,
1310
+ "step": 184
1311
+ },
1312
+ {
1313
+ "epoch": 1.4919354838709677,
1314
+ "grad_norm": 28.088899612426758,
1315
+ "learning_rate": 1.739959741286057e-05,
1316
+ "loss": 1.3278,
1317
+ "step": 185
1318
+ },
1319
+ {
1320
+ "epoch": 1.5,
1321
+ "grad_norm": 15.711658477783203,
1322
+ "learning_rate": 1.7359598338348243e-05,
1323
+ "loss": 0.3969,
1324
+ "step": 186
1325
+ },
1326
+ {
1327
+ "epoch": 1.5080645161290323,
1328
+ "grad_norm": 37.35240173339844,
1329
+ "learning_rate": 1.731959926383592e-05,
1330
+ "loss": 4.7095,
1331
+ "step": 187
1332
+ },
1333
+ {
1334
+ "epoch": 1.5161290322580645,
1335
+ "grad_norm": 24.711170196533203,
1336
+ "learning_rate": 1.7279600189323597e-05,
1337
+ "loss": 1.3376,
1338
+ "step": 188
1339
+ },
1340
+ {
1341
+ "epoch": 1.5241935483870968,
1342
+ "grad_norm": 40.604549407958984,
1343
+ "learning_rate": 1.7239601114811277e-05,
1344
+ "loss": 1.7629,
1345
+ "step": 189
1346
+ },
1347
+ {
1348
+ "epoch": 1.532258064516129,
1349
+ "grad_norm": 25.869022369384766,
1350
+ "learning_rate": 1.7199602040298954e-05,
1351
+ "loss": 0.8421,
1352
+ "step": 190
1353
+ },
1354
+ {
1355
+ "epoch": 1.5403225806451613,
1356
+ "grad_norm": 6.451545715332031,
1357
+ "learning_rate": 1.7159602965786627e-05,
1358
+ "loss": 0.2054,
1359
+ "step": 191
1360
+ },
1361
+ {
1362
+ "epoch": 1.5483870967741935,
1363
+ "grad_norm": 8.05915641784668,
1364
+ "learning_rate": 1.7119603891274304e-05,
1365
+ "loss": 0.5445,
1366
+ "step": 192
1367
+ },
1368
+ {
1369
+ "epoch": 1.5564516129032258,
1370
+ "grad_norm": 23.62920570373535,
1371
+ "learning_rate": 1.707960481676198e-05,
1372
+ "loss": 0.6784,
1373
+ "step": 193
1374
+ },
1375
+ {
1376
+ "epoch": 1.564516129032258,
1377
+ "grad_norm": 19.833181381225586,
1378
+ "learning_rate": 1.7039605742249658e-05,
1379
+ "loss": 0.5534,
1380
+ "step": 194
1381
+ },
1382
+ {
1383
+ "epoch": 1.5725806451612905,
1384
+ "grad_norm": 14.539905548095703,
1385
+ "learning_rate": 1.6999606667737335e-05,
1386
+ "loss": 0.4673,
1387
+ "step": 195
1388
+ },
1389
+ {
1390
+ "epoch": 1.5806451612903225,
1391
+ "grad_norm": 103.28005981445312,
1392
+ "learning_rate": 1.6959607593225012e-05,
1393
+ "loss": 6.914,
1394
+ "step": 196
1395
+ },
1396
+ {
1397
+ "epoch": 1.588709677419355,
1398
+ "grad_norm": 39.585941314697266,
1399
+ "learning_rate": 1.691960851871269e-05,
1400
+ "loss": 1.0972,
1401
+ "step": 197
1402
+ },
1403
+ {
1404
+ "epoch": 1.596774193548387,
1405
+ "grad_norm": 15.518098831176758,
1406
+ "learning_rate": 1.6879609444200366e-05,
1407
+ "loss": 0.4614,
1408
+ "step": 198
1409
+ },
1410
+ {
1411
+ "epoch": 1.6048387096774195,
1412
+ "grad_norm": 4.58119535446167,
1413
+ "learning_rate": 1.6839610369688043e-05,
1414
+ "loss": 0.1082,
1415
+ "step": 199
1416
+ },
1417
+ {
1418
+ "epoch": 1.6129032258064515,
1419
+ "grad_norm": 25.006546020507812,
1420
+ "learning_rate": 1.679961129517572e-05,
1421
+ "loss": 0.4672,
1422
+ "step": 200
1423
+ },
1424
+ {
1425
+ "epoch": 1.620967741935484,
1426
+ "grad_norm": 28.234201431274414,
1427
+ "learning_rate": 1.6759612220663397e-05,
1428
+ "loss": 0.5142,
1429
+ "step": 201
1430
+ },
1431
+ {
1432
+ "epoch": 1.629032258064516,
1433
+ "grad_norm": 24.016407012939453,
1434
+ "learning_rate": 1.6719613146151073e-05,
1435
+ "loss": 0.9564,
1436
+ "step": 202
1437
+ },
1438
+ {
1439
+ "epoch": 1.6370967741935485,
1440
+ "grad_norm": 18.26400375366211,
1441
+ "learning_rate": 1.667961407163875e-05,
1442
+ "loss": 0.3671,
1443
+ "step": 203
1444
+ },
1445
+ {
1446
+ "epoch": 1.6451612903225805,
1447
+ "grad_norm": 12.657865524291992,
1448
+ "learning_rate": 1.6639614997126427e-05,
1449
+ "loss": 0.6099,
1450
+ "step": 204
1451
+ },
1452
+ {
1453
+ "epoch": 1.653225806451613,
1454
+ "grad_norm": 117.20157623291016,
1455
+ "learning_rate": 1.6599615922614104e-05,
1456
+ "loss": 14.7071,
1457
+ "step": 205
1458
+ },
1459
+ {
1460
+ "epoch": 1.661290322580645,
1461
+ "grad_norm": 5.277425289154053,
1462
+ "learning_rate": 1.655961684810178e-05,
1463
+ "loss": 0.0183,
1464
+ "step": 206
1465
+ },
1466
+ {
1467
+ "epoch": 1.6693548387096775,
1468
+ "grad_norm": 8.111205101013184,
1469
+ "learning_rate": 1.6519617773589458e-05,
1470
+ "loss": 0.2201,
1471
+ "step": 207
1472
+ },
1473
+ {
1474
+ "epoch": 1.6774193548387095,
1475
+ "grad_norm": 15.38776683807373,
1476
+ "learning_rate": 1.6479618699077135e-05,
1477
+ "loss": 0.3226,
1478
+ "step": 208
1479
+ },
1480
+ {
1481
+ "epoch": 1.685483870967742,
1482
+ "grad_norm": 6.60953950881958,
1483
+ "learning_rate": 1.6439619624564812e-05,
1484
+ "loss": 0.3358,
1485
+ "step": 209
1486
+ },
1487
+ {
1488
+ "epoch": 1.6935483870967742,
1489
+ "grad_norm": 17.016639709472656,
1490
+ "learning_rate": 1.639962055005249e-05,
1491
+ "loss": 0.2497,
1492
+ "step": 210
1493
+ },
1494
+ {
1495
+ "epoch": 1.7016129032258065,
1496
+ "grad_norm": 54.830169677734375,
1497
+ "learning_rate": 1.6359621475540166e-05,
1498
+ "loss": 1.7451,
1499
+ "step": 211
1500
+ },
1501
+ {
1502
+ "epoch": 1.7096774193548387,
1503
+ "grad_norm": 14.282036781311035,
1504
+ "learning_rate": 1.6319622401027843e-05,
1505
+ "loss": 1.1429,
1506
+ "step": 212
1507
+ },
1508
+ {
1509
+ "epoch": 1.717741935483871,
1510
+ "grad_norm": 4.297520160675049,
1511
+ "learning_rate": 1.627962332651552e-05,
1512
+ "loss": 0.1797,
1513
+ "step": 213
1514
+ },
1515
+ {
1516
+ "epoch": 1.7258064516129032,
1517
+ "grad_norm": 142.38172912597656,
1518
+ "learning_rate": 1.6239624252003196e-05,
1519
+ "loss": 21.8816,
1520
+ "step": 214
1521
+ },
1522
+ {
1523
+ "epoch": 1.7338709677419355,
1524
+ "grad_norm": 88.52197265625,
1525
+ "learning_rate": 1.6199625177490873e-05,
1526
+ "loss": 15.5891,
1527
+ "step": 215
1528
+ },
1529
+ {
1530
+ "epoch": 1.7419354838709677,
1531
+ "grad_norm": 15.318389892578125,
1532
+ "learning_rate": 1.615962610297855e-05,
1533
+ "loss": 0.4315,
1534
+ "step": 216
1535
+ },
1536
+ {
1537
+ "epoch": 1.75,
1538
+ "grad_norm": 102.21027374267578,
1539
+ "learning_rate": 1.6119627028466227e-05,
1540
+ "loss": 6.8871,
1541
+ "step": 217
1542
+ },
1543
+ {
1544
+ "epoch": 1.7580645161290323,
1545
+ "grad_norm": 12.100990295410156,
1546
+ "learning_rate": 1.6079627953953904e-05,
1547
+ "loss": 0.1098,
1548
+ "step": 218
1549
+ },
1550
+ {
1551
+ "epoch": 1.7661290322580645,
1552
+ "grad_norm": 137.26162719726562,
1553
+ "learning_rate": 1.6039628879441578e-05,
1554
+ "loss": 12.6184,
1555
+ "step": 219
1556
+ },
1557
+ {
1558
+ "epoch": 1.7741935483870968,
1559
+ "grad_norm": 16.48599624633789,
1560
+ "learning_rate": 1.5999629804929258e-05,
1561
+ "loss": 0.3973,
1562
+ "step": 220
1563
+ },
1564
+ {
1565
+ "epoch": 1.782258064516129,
1566
+ "grad_norm": 23.34356689453125,
1567
+ "learning_rate": 1.5959630730416935e-05,
1568
+ "loss": 0.3448,
1569
+ "step": 221
1570
+ },
1571
+ {
1572
+ "epoch": 1.7903225806451613,
1573
+ "grad_norm": 7.864492893218994,
1574
+ "learning_rate": 1.5919631655904612e-05,
1575
+ "loss": 0.502,
1576
+ "step": 222
1577
+ },
1578
+ {
1579
+ "epoch": 1.7983870967741935,
1580
+ "grad_norm": 3.356924295425415,
1581
+ "learning_rate": 1.587963258139229e-05,
1582
+ "loss": 0.0997,
1583
+ "step": 223
1584
+ },
1585
+ {
1586
+ "epoch": 1.8064516129032258,
1587
+ "grad_norm": 125.84968566894531,
1588
+ "learning_rate": 1.5839633506879962e-05,
1589
+ "loss": 13.3317,
1590
+ "step": 224
1591
+ },
1592
+ {
1593
+ "epoch": 1.814516129032258,
1594
+ "grad_norm": 77.49720764160156,
1595
+ "learning_rate": 1.5799634432367643e-05,
1596
+ "loss": 4.3421,
1597
+ "step": 225
1598
+ },
1599
+ {
1600
+ "epoch": 1.8225806451612905,
1601
+ "grad_norm": 24.029203414916992,
1602
+ "learning_rate": 1.575963535785532e-05,
1603
+ "loss": 1.8343,
1604
+ "step": 226
1605
+ },
1606
+ {
1607
+ "epoch": 1.8306451612903225,
1608
+ "grad_norm": 70.15037536621094,
1609
+ "learning_rate": 1.5719636283342996e-05,
1610
+ "loss": 1.469,
1611
+ "step": 227
1612
+ },
1613
+ {
1614
+ "epoch": 1.838709677419355,
1615
+ "grad_norm": 69.79930877685547,
1616
+ "learning_rate": 1.567963720883067e-05,
1617
+ "loss": 1.8992,
1618
+ "step": 228
1619
+ },
1620
+ {
1621
+ "epoch": 1.846774193548387,
1622
+ "grad_norm": 63.507442474365234,
1623
+ "learning_rate": 1.5639638134318347e-05,
1624
+ "loss": 1.2837,
1625
+ "step": 229
1626
+ },
1627
+ {
1628
+ "epoch": 1.8548387096774195,
1629
+ "grad_norm": 54.0720100402832,
1630
+ "learning_rate": 1.5599639059806027e-05,
1631
+ "loss": 3.8239,
1632
+ "step": 230
1633
+ },
1634
+ {
1635
+ "epoch": 1.8629032258064515,
1636
+ "grad_norm": 71.55313873291016,
1637
+ "learning_rate": 1.5559639985293704e-05,
1638
+ "loss": 2.0285,
1639
+ "step": 231
1640
+ },
1641
+ {
1642
+ "epoch": 1.870967741935484,
1643
+ "grad_norm": 9.896157264709473,
1644
+ "learning_rate": 1.551964091078138e-05,
1645
+ "loss": 0.329,
1646
+ "step": 232
1647
+ },
1648
+ {
1649
+ "epoch": 1.879032258064516,
1650
+ "grad_norm": 19.860387802124023,
1651
+ "learning_rate": 1.5479641836269055e-05,
1652
+ "loss": 0.6743,
1653
+ "step": 233
1654
+ },
1655
+ {
1656
+ "epoch": 1.8870967741935485,
1657
+ "grad_norm": 5.622892379760742,
1658
+ "learning_rate": 1.543964276175673e-05,
1659
+ "loss": 0.1553,
1660
+ "step": 234
1661
+ },
1662
+ {
1663
+ "epoch": 1.8951612903225805,
1664
+ "grad_norm": 12.858086585998535,
1665
+ "learning_rate": 1.539964368724441e-05,
1666
+ "loss": 0.2809,
1667
+ "step": 235
1668
+ },
1669
+ {
1670
+ "epoch": 1.903225806451613,
1671
+ "grad_norm": 108.66515350341797,
1672
+ "learning_rate": 1.535964461273209e-05,
1673
+ "loss": 5.8853,
1674
+ "step": 236
1675
+ },
1676
+ {
1677
+ "epoch": 1.911290322580645,
1678
+ "grad_norm": 47.20570755004883,
1679
+ "learning_rate": 1.5319645538219766e-05,
1680
+ "loss": 4.3275,
1681
+ "step": 237
1682
+ },
1683
+ {
1684
+ "epoch": 1.9193548387096775,
1685
+ "grad_norm": 43.51235580444336,
1686
+ "learning_rate": 1.527964646370744e-05,
1687
+ "loss": 0.7748,
1688
+ "step": 238
1689
+ },
1690
+ {
1691
+ "epoch": 1.9274193548387095,
1692
+ "grad_norm": 27.735137939453125,
1693
+ "learning_rate": 1.5239647389195118e-05,
1694
+ "loss": 0.5191,
1695
+ "step": 239
1696
+ },
1697
+ {
1698
+ "epoch": 1.935483870967742,
1699
+ "grad_norm": 27.011123657226562,
1700
+ "learning_rate": 1.5199648314682795e-05,
1701
+ "loss": 0.9489,
1702
+ "step": 240
1703
+ },
1704
+ {
1705
+ "epoch": 1.9435483870967742,
1706
+ "grad_norm": 10.22940731048584,
1707
+ "learning_rate": 1.5159649240170472e-05,
1708
+ "loss": 0.3328,
1709
+ "step": 241
1710
+ },
1711
+ {
1712
+ "epoch": 1.9516129032258065,
1713
+ "grad_norm": 17.85344886779785,
1714
+ "learning_rate": 1.5119650165658147e-05,
1715
+ "loss": 0.3374,
1716
+ "step": 242
1717
+ },
1718
+ {
1719
+ "epoch": 1.9596774193548387,
1720
+ "grad_norm": 35.01970291137695,
1721
+ "learning_rate": 1.5079651091145824e-05,
1722
+ "loss": 0.7993,
1723
+ "step": 243
1724
+ },
1725
+ {
1726
+ "epoch": 1.967741935483871,
1727
+ "grad_norm": 259.75616455078125,
1728
+ "learning_rate": 1.50396520166335e-05,
1729
+ "loss": 7.7043,
1730
+ "step": 244
1731
+ },
1732
+ {
1733
+ "epoch": 1.9758064516129032,
1734
+ "grad_norm": 75.41777038574219,
1735
+ "learning_rate": 1.499965294212118e-05,
1736
+ "loss": 1.8485,
1737
+ "step": 245
1738
+ },
1739
+ {
1740
+ "epoch": 1.9838709677419355,
1741
+ "grad_norm": 401.5027770996094,
1742
+ "learning_rate": 1.4959653867608856e-05,
1743
+ "loss": 19.4312,
1744
+ "step": 246
1745
+ },
1746
+ {
1747
+ "epoch": 1.9919354838709677,
1748
+ "grad_norm": 46.57631301879883,
1749
+ "learning_rate": 1.4919654793096532e-05,
1750
+ "loss": 1.3285,
1751
+ "step": 247
1752
+ },
1753
+ {
1754
+ "epoch": 2.0,
1755
+ "grad_norm": 110.93121337890625,
1756
+ "learning_rate": 1.4879655718584208e-05,
1757
+ "loss": 2.6528,
1758
+ "step": 248
1759
+ },
1760
+ {
1761
+ "epoch": 2.0,
1762
+ "eval_loss": 4.852317810058594,
1763
+ "eval_mae": 1.4657387733459473,
1764
+ "eval_mse": 4.85231876373291,
1765
+ "eval_r2": -0.32848048210144043,
1766
+ "eval_rmse": 2.2027979398330912,
1767
+ "eval_runtime": 1.3879,
1768
+ "eval_samples_per_second": 39.629,
1769
+ "eval_smape": 54.71565127372742,
1770
+ "eval_steps_per_second": 10.087,
1771
+ "step": 248
1772
+ },
1773
+ {
1774
+ "epoch": 2.0080645161290325,
1775
+ "grad_norm": 154.8404998779297,
1776
+ "learning_rate": 1.4839656644071885e-05,
1777
+ "loss": 3.6054,
1778
+ "step": 249
1779
+ },
1780
+ {
1781
+ "epoch": 2.0161290322580645,
1782
+ "grad_norm": 62.42089080810547,
1783
+ "learning_rate": 1.4799657569559564e-05,
1784
+ "loss": 1.8889,
1785
+ "step": 250
1786
+ },
1787
+ {
1788
+ "epoch": 2.024193548387097,
1789
+ "grad_norm": 70.86661529541016,
1790
+ "learning_rate": 1.4759658495047241e-05,
1791
+ "loss": 2.4493,
1792
+ "step": 251
1793
+ },
1794
+ {
1795
+ "epoch": 2.032258064516129,
1796
+ "grad_norm": 51.09244155883789,
1797
+ "learning_rate": 1.4719659420534916e-05,
1798
+ "loss": 1.168,
1799
+ "step": 252
1800
+ },
1801
+ {
1802
+ "epoch": 2.0403225806451615,
1803
+ "grad_norm": 18.558488845825195,
1804
+ "learning_rate": 1.4679660346022593e-05,
1805
+ "loss": 0.2762,
1806
+ "step": 253
1807
+ },
1808
+ {
1809
+ "epoch": 2.0483870967741935,
1810
+ "grad_norm": 168.22433471679688,
1811
+ "learning_rate": 1.463966127151027e-05,
1812
+ "loss": 1.9643,
1813
+ "step": 254
1814
+ },
1815
+ {
1816
+ "epoch": 2.056451612903226,
1817
+ "grad_norm": 40.86513900756836,
1818
+ "learning_rate": 1.4599662196997949e-05,
1819
+ "loss": 0.8871,
1820
+ "step": 255
1821
+ },
1822
+ {
1823
+ "epoch": 2.064516129032258,
1824
+ "grad_norm": 169.68312072753906,
1825
+ "learning_rate": 1.4559663122485622e-05,
1826
+ "loss": 10.1988,
1827
+ "step": 256
1828
+ },
1829
+ {
1830
+ "epoch": 2.0725806451612905,
1831
+ "grad_norm": 6.839456081390381,
1832
+ "learning_rate": 1.45196640479733e-05,
1833
+ "loss": 0.195,
1834
+ "step": 257
1835
+ },
1836
+ {
1837
+ "epoch": 2.0806451612903225,
1838
+ "grad_norm": 29.532468795776367,
1839
+ "learning_rate": 1.4479664973460978e-05,
1840
+ "loss": 0.4214,
1841
+ "step": 258
1842
+ },
1843
+ {
1844
+ "epoch": 2.088709677419355,
1845
+ "grad_norm": 134.2998504638672,
1846
+ "learning_rate": 1.4439665898948655e-05,
1847
+ "loss": 22.2033,
1848
+ "step": 259
1849
+ },
1850
+ {
1851
+ "epoch": 2.096774193548387,
1852
+ "grad_norm": 229.64031982421875,
1853
+ "learning_rate": 1.4399666824436333e-05,
1854
+ "loss": 0.8783,
1855
+ "step": 260
1856
+ },
1857
+ {
1858
+ "epoch": 2.1048387096774195,
1859
+ "grad_norm": 150.3468017578125,
1860
+ "learning_rate": 1.4359667749924007e-05,
1861
+ "loss": 22.9731,
1862
+ "step": 261
1863
+ },
1864
+ {
1865
+ "epoch": 2.1129032258064515,
1866
+ "grad_norm": 104.23390197753906,
1867
+ "learning_rate": 1.4319668675411685e-05,
1868
+ "loss": 1.326,
1869
+ "step": 262
1870
+ },
1871
+ {
1872
+ "epoch": 2.120967741935484,
1873
+ "grad_norm": 114.47550964355469,
1874
+ "learning_rate": 1.4279669600899362e-05,
1875
+ "loss": 1.3029,
1876
+ "step": 263
1877
+ },
1878
+ {
1879
+ "epoch": 2.129032258064516,
1880
+ "grad_norm": 15.87612533569336,
1881
+ "learning_rate": 1.423967052638704e-05,
1882
+ "loss": 0.4699,
1883
+ "step": 264
1884
+ },
1885
+ {
1886
+ "epoch": 2.1370967741935485,
1887
+ "grad_norm": 46.12168884277344,
1888
+ "learning_rate": 1.4199671451874716e-05,
1889
+ "loss": 0.7016,
1890
+ "step": 265
1891
+ },
1892
+ {
1893
+ "epoch": 2.1451612903225805,
1894
+ "grad_norm": 279.55279541015625,
1895
+ "learning_rate": 1.4159672377362391e-05,
1896
+ "loss": 3.2752,
1897
+ "step": 266
1898
+ },
1899
+ {
1900
+ "epoch": 2.153225806451613,
1901
+ "grad_norm": 20.628719329833984,
1902
+ "learning_rate": 1.4119673302850068e-05,
1903
+ "loss": 0.2388,
1904
+ "step": 267
1905
+ },
1906
+ {
1907
+ "epoch": 2.161290322580645,
1908
+ "grad_norm": 22.857067108154297,
1909
+ "learning_rate": 1.4079674228337747e-05,
1910
+ "loss": 0.5555,
1911
+ "step": 268
1912
+ },
1913
+ {
1914
+ "epoch": 2.1693548387096775,
1915
+ "grad_norm": 47.45686340332031,
1916
+ "learning_rate": 1.4039675153825424e-05,
1917
+ "loss": 2.8467,
1918
+ "step": 269
1919
+ },
1920
+ {
1921
+ "epoch": 2.1774193548387095,
1922
+ "grad_norm": 22.756681442260742,
1923
+ "learning_rate": 1.39996760793131e-05,
1924
+ "loss": 0.7615,
1925
+ "step": 270
1926
+ },
1927
+ {
1928
+ "epoch": 2.185483870967742,
1929
+ "grad_norm": 80.15179443359375,
1930
+ "learning_rate": 1.3959677004800776e-05,
1931
+ "loss": 8.9323,
1932
+ "step": 271
1933
+ },
1934
+ {
1935
+ "epoch": 2.193548387096774,
1936
+ "grad_norm": 100.86170196533203,
1937
+ "learning_rate": 1.3919677930288453e-05,
1938
+ "loss": 12.4829,
1939
+ "step": 272
1940
+ },
1941
+ {
1942
+ "epoch": 2.2016129032258065,
1943
+ "grad_norm": 72.2647705078125,
1944
+ "learning_rate": 1.3879678855776132e-05,
1945
+ "loss": 9.6687,
1946
+ "step": 273
1947
+ },
1948
+ {
1949
+ "epoch": 2.2096774193548385,
1950
+ "grad_norm": 30.535030364990234,
1951
+ "learning_rate": 1.3839679781263808e-05,
1952
+ "loss": 0.4308,
1953
+ "step": 274
1954
+ },
1955
+ {
1956
+ "epoch": 2.217741935483871,
1957
+ "grad_norm": 54.71157455444336,
1958
+ "learning_rate": 1.3799680706751484e-05,
1959
+ "loss": 1.1555,
1960
+ "step": 275
1961
+ },
1962
+ {
1963
+ "epoch": 2.225806451612903,
1964
+ "grad_norm": 39.96649932861328,
1965
+ "learning_rate": 1.375968163223916e-05,
1966
+ "loss": 0.4534,
1967
+ "step": 276
1968
+ },
1969
+ {
1970
+ "epoch": 2.2338709677419355,
1971
+ "grad_norm": 55.768898010253906,
1972
+ "learning_rate": 1.3719682557726838e-05,
1973
+ "loss": 0.9371,
1974
+ "step": 277
1975
+ },
1976
+ {
1977
+ "epoch": 2.241935483870968,
1978
+ "grad_norm": 83.68522644042969,
1979
+ "learning_rate": 1.3679683483214516e-05,
1980
+ "loss": 2.4787,
1981
+ "step": 278
1982
+ },
1983
+ {
1984
+ "epoch": 2.25,
1985
+ "grad_norm": 29.500429153442383,
1986
+ "learning_rate": 1.3639684408702193e-05,
1987
+ "loss": 1.0634,
1988
+ "step": 279
1989
+ },
1990
+ {
1991
+ "epoch": 2.258064516129032,
1992
+ "grad_norm": 100.61076354980469,
1993
+ "learning_rate": 1.3599685334189868e-05,
1994
+ "loss": 10.4759,
1995
+ "step": 280
1996
+ },
1997
+ {
1998
+ "epoch": 2.2661290322580645,
1999
+ "grad_norm": 182.71800231933594,
2000
+ "learning_rate": 1.3559686259677545e-05,
2001
+ "loss": 1.687,
2002
+ "step": 281
2003
+ },
2004
+ {
2005
+ "epoch": 2.274193548387097,
2006
+ "grad_norm": 187.59808349609375,
2007
+ "learning_rate": 1.3519687185165222e-05,
2008
+ "loss": 15.9992,
2009
+ "step": 282
2010
+ },
2011
+ {
2012
+ "epoch": 2.282258064516129,
2013
+ "grad_norm": 105.1393051147461,
2014
+ "learning_rate": 1.3479688110652899e-05,
2015
+ "loss": 4.9415,
2016
+ "step": 283
2017
+ },
2018
+ {
2019
+ "epoch": 2.2903225806451615,
2020
+ "grad_norm": 115.4957504272461,
2021
+ "learning_rate": 1.3439689036140578e-05,
2022
+ "loss": 7.5135,
2023
+ "step": 284
2024
+ },
2025
+ {
2026
+ "epoch": 2.2983870967741935,
2027
+ "grad_norm": 42.67521286010742,
2028
+ "learning_rate": 1.3399689961628251e-05,
2029
+ "loss": 0.9061,
2030
+ "step": 285
2031
+ },
2032
+ {
2033
+ "epoch": 2.306451612903226,
2034
+ "grad_norm": 31.427526473999023,
2035
+ "learning_rate": 1.335969088711593e-05,
2036
+ "loss": 0.4332,
2037
+ "step": 286
2038
+ },
2039
+ {
2040
+ "epoch": 2.314516129032258,
2041
+ "grad_norm": 47.62785339355469,
2042
+ "learning_rate": 1.3319691812603607e-05,
2043
+ "loss": 0.9642,
2044
+ "step": 287
2045
+ },
2046
+ {
2047
+ "epoch": 2.3225806451612905,
2048
+ "grad_norm": 15.01259708404541,
2049
+ "learning_rate": 1.3279692738091284e-05,
2050
+ "loss": 0.2899,
2051
+ "step": 288
2052
+ },
2053
+ {
2054
+ "epoch": 2.3306451612903225,
2055
+ "grad_norm": 37.724735260009766,
2056
+ "learning_rate": 1.3239693663578959e-05,
2057
+ "loss": 1.7805,
2058
+ "step": 289
2059
+ },
2060
+ {
2061
+ "epoch": 2.338709677419355,
2062
+ "grad_norm": 79.78099060058594,
2063
+ "learning_rate": 1.3199694589066636e-05,
2064
+ "loss": 8.5522,
2065
+ "step": 290
2066
+ },
2067
+ {
2068
+ "epoch": 2.346774193548387,
2069
+ "grad_norm": 20.72849464416504,
2070
+ "learning_rate": 1.3159695514554314e-05,
2071
+ "loss": 0.4305,
2072
+ "step": 291
2073
+ },
2074
+ {
2075
+ "epoch": 2.3548387096774195,
2076
+ "grad_norm": 137.8477325439453,
2077
+ "learning_rate": 1.3119696440041991e-05,
2078
+ "loss": 5.5653,
2079
+ "step": 292
2080
+ },
2081
+ {
2082
+ "epoch": 2.3629032258064515,
2083
+ "grad_norm": 49.036468505859375,
2084
+ "learning_rate": 1.3079697365529668e-05,
2085
+ "loss": 0.3396,
2086
+ "step": 293
2087
+ },
2088
+ {
2089
+ "epoch": 2.370967741935484,
2090
+ "grad_norm": 77.85248565673828,
2091
+ "learning_rate": 1.3039698291017344e-05,
2092
+ "loss": 12.0845,
2093
+ "step": 294
2094
+ },
2095
+ {
2096
+ "epoch": 2.379032258064516,
2097
+ "grad_norm": 84.66986846923828,
2098
+ "learning_rate": 1.299969921650502e-05,
2099
+ "loss": 13.1287,
2100
+ "step": 295
2101
+ },
2102
+ {
2103
+ "epoch": 2.3870967741935485,
2104
+ "grad_norm": 19.693037033081055,
2105
+ "learning_rate": 1.2959700141992699e-05,
2106
+ "loss": 0.4963,
2107
+ "step": 296
2108
+ },
2109
+ {
2110
+ "epoch": 2.3951612903225805,
2111
+ "grad_norm": 10.413263320922852,
2112
+ "learning_rate": 1.2919701067480376e-05,
2113
+ "loss": 0.2287,
2114
+ "step": 297
2115
+ },
2116
+ {
2117
+ "epoch": 2.403225806451613,
2118
+ "grad_norm": 64.92196655273438,
2119
+ "learning_rate": 1.2879701992968053e-05,
2120
+ "loss": 1.6285,
2121
+ "step": 298
2122
+ },
2123
+ {
2124
+ "epoch": 2.411290322580645,
2125
+ "grad_norm": 36.941349029541016,
2126
+ "learning_rate": 1.2839702918455728e-05,
2127
+ "loss": 0.8435,
2128
+ "step": 299
2129
+ },
2130
+ {
2131
+ "epoch": 2.4193548387096775,
2132
+ "grad_norm": 149.9400177001953,
2133
+ "learning_rate": 1.2799703843943405e-05,
2134
+ "loss": 4.2246,
2135
+ "step": 300
2136
+ },
2137
+ {
2138
+ "epoch": 2.4274193548387095,
2139
+ "grad_norm": 40.55129623413086,
2140
+ "learning_rate": 1.2759704769431082e-05,
2141
+ "loss": 0.628,
2142
+ "step": 301
2143
+ },
2144
+ {
2145
+ "epoch": 2.435483870967742,
2146
+ "grad_norm": 39.12997817993164,
2147
+ "learning_rate": 1.271970569491876e-05,
2148
+ "loss": 0.4721,
2149
+ "step": 302
2150
+ },
2151
+ {
2152
+ "epoch": 2.443548387096774,
2153
+ "grad_norm": 87.25220489501953,
2154
+ "learning_rate": 1.2679706620406434e-05,
2155
+ "loss": 6.3045,
2156
+ "step": 303
2157
+ },
2158
+ {
2159
+ "epoch": 2.4516129032258065,
2160
+ "grad_norm": 47.91136169433594,
2161
+ "learning_rate": 1.2639707545894113e-05,
2162
+ "loss": 0.6415,
2163
+ "step": 304
2164
+ },
2165
+ {
2166
+ "epoch": 2.4596774193548385,
2167
+ "grad_norm": 103.10691833496094,
2168
+ "learning_rate": 1.259970847138179e-05,
2169
+ "loss": 10.6521,
2170
+ "step": 305
2171
+ },
2172
+ {
2173
+ "epoch": 2.467741935483871,
2174
+ "grad_norm": 16.19340705871582,
2175
+ "learning_rate": 1.2559709396869467e-05,
2176
+ "loss": 0.2191,
2177
+ "step": 306
2178
+ },
2179
+ {
2180
+ "epoch": 2.475806451612903,
2181
+ "grad_norm": 18.110265731811523,
2182
+ "learning_rate": 1.2519710322357145e-05,
2183
+ "loss": 0.2425,
2184
+ "step": 307
2185
+ },
2186
+ {
2187
+ "epoch": 2.4838709677419355,
2188
+ "grad_norm": 156.45272827148438,
2189
+ "learning_rate": 1.2479711247844819e-05,
2190
+ "loss": 4.5986,
2191
+ "step": 308
2192
+ },
2193
+ {
2194
+ "epoch": 2.491935483870968,
2195
+ "grad_norm": 16.38495635986328,
2196
+ "learning_rate": 1.2439712173332497e-05,
2197
+ "loss": 0.3502,
2198
+ "step": 309
2199
+ },
2200
+ {
2201
+ "epoch": 2.5,
2202
+ "grad_norm": 24.752309799194336,
2203
+ "learning_rate": 1.2399713098820174e-05,
2204
+ "loss": 0.4216,
2205
+ "step": 310
2206
+ },
2207
+ {
2208
+ "epoch": 2.508064516129032,
2209
+ "grad_norm": 22.331012725830078,
2210
+ "learning_rate": 1.2359714024307851e-05,
2211
+ "loss": 0.3915,
2212
+ "step": 311
2213
+ },
2214
+ {
2215
+ "epoch": 2.5161290322580645,
2216
+ "grad_norm": 4.984405994415283,
2217
+ "learning_rate": 1.2319714949795528e-05,
2218
+ "loss": 0.1369,
2219
+ "step": 312
2220
+ },
2221
+ {
2222
+ "epoch": 2.524193548387097,
2223
+ "grad_norm": 4.796787261962891,
2224
+ "learning_rate": 1.2279715875283205e-05,
2225
+ "loss": 0.1145,
2226
+ "step": 313
2227
+ },
2228
+ {
2229
+ "epoch": 2.532258064516129,
2230
+ "grad_norm": 11.541577339172363,
2231
+ "learning_rate": 1.2239716800770882e-05,
2232
+ "loss": 0.4872,
2233
+ "step": 314
2234
+ },
2235
+ {
2236
+ "epoch": 2.540322580645161,
2237
+ "grad_norm": 117.4135971069336,
2238
+ "learning_rate": 1.2199717726258559e-05,
2239
+ "loss": 8.6798,
2240
+ "step": 315
2241
+ },
2242
+ {
2243
+ "epoch": 2.5483870967741935,
2244
+ "grad_norm": 11.267465591430664,
2245
+ "learning_rate": 1.2159718651746234e-05,
2246
+ "loss": 0.2144,
2247
+ "step": 316
2248
+ },
2249
+ {
2250
+ "epoch": 2.556451612903226,
2251
+ "grad_norm": 17.15199851989746,
2252
+ "learning_rate": 1.2119719577233913e-05,
2253
+ "loss": 0.4501,
2254
+ "step": 317
2255
+ },
2256
+ {
2257
+ "epoch": 2.564516129032258,
2258
+ "grad_norm": 14.21686840057373,
2259
+ "learning_rate": 1.207972050272159e-05,
2260
+ "loss": 0.4759,
2261
+ "step": 318
2262
+ },
2263
+ {
2264
+ "epoch": 2.5725806451612905,
2265
+ "grad_norm": 52.95610809326172,
2266
+ "learning_rate": 1.2039721428209267e-05,
2267
+ "loss": 0.792,
2268
+ "step": 319
2269
+ },
2270
+ {
2271
+ "epoch": 2.5806451612903225,
2272
+ "grad_norm": 18.222946166992188,
2273
+ "learning_rate": 1.1999722353696944e-05,
2274
+ "loss": 0.9593,
2275
+ "step": 320
2276
+ },
2277
+ {
2278
+ "epoch": 2.588709677419355,
2279
+ "grad_norm": 112.54412078857422,
2280
+ "learning_rate": 1.1959723279184619e-05,
2281
+ "loss": 5.1768,
2282
+ "step": 321
2283
+ },
2284
+ {
2285
+ "epoch": 2.596774193548387,
2286
+ "grad_norm": 5.891172885894775,
2287
+ "learning_rate": 1.1919724204672297e-05,
2288
+ "loss": 0.336,
2289
+ "step": 322
2290
+ },
2291
+ {
2292
+ "epoch": 2.6048387096774195,
2293
+ "grad_norm": 140.28492736816406,
2294
+ "learning_rate": 1.1879725130159973e-05,
2295
+ "loss": 11.1532,
2296
+ "step": 323
2297
+ },
2298
+ {
2299
+ "epoch": 2.6129032258064515,
2300
+ "grad_norm": 27.81146812438965,
2301
+ "learning_rate": 1.183972605564765e-05,
2302
+ "loss": 0.4171,
2303
+ "step": 324
2304
+ },
2305
+ {
2306
+ "epoch": 2.620967741935484,
2307
+ "grad_norm": 13.444504737854004,
2308
+ "learning_rate": 1.1799726981135328e-05,
2309
+ "loss": 0.181,
2310
+ "step": 325
2311
+ },
2312
+ {
2313
+ "epoch": 2.629032258064516,
2314
+ "grad_norm": 19.829408645629883,
2315
+ "learning_rate": 1.1759727906623003e-05,
2316
+ "loss": 0.6601,
2317
+ "step": 326
2318
+ },
2319
+ {
2320
+ "epoch": 2.6370967741935485,
2321
+ "grad_norm": 13.109299659729004,
2322
+ "learning_rate": 1.1719728832110682e-05,
2323
+ "loss": 0.8257,
2324
+ "step": 327
2325
+ },
2326
+ {
2327
+ "epoch": 2.6451612903225805,
2328
+ "grad_norm": 37.20515823364258,
2329
+ "learning_rate": 1.1679729757598357e-05,
2330
+ "loss": 0.604,
2331
+ "step": 328
2332
+ },
2333
+ {
2334
+ "epoch": 2.653225806451613,
2335
+ "grad_norm": 401.8742980957031,
2336
+ "learning_rate": 1.1639730683086034e-05,
2337
+ "loss": 123.0045,
2338
+ "step": 329
2339
+ },
2340
+ {
2341
+ "epoch": 2.661290322580645,
2342
+ "grad_norm": 41.35789489746094,
2343
+ "learning_rate": 1.1599731608573711e-05,
2344
+ "loss": 0.8282,
2345
+ "step": 330
2346
+ },
2347
+ {
2348
+ "epoch": 2.6693548387096775,
2349
+ "grad_norm": 36.715728759765625,
2350
+ "learning_rate": 1.1559732534061388e-05,
2351
+ "loss": 0.9222,
2352
+ "step": 331
2353
+ },
2354
+ {
2355
+ "epoch": 2.6774193548387095,
2356
+ "grad_norm": 236.11050415039062,
2357
+ "learning_rate": 1.1519733459549065e-05,
2358
+ "loss": 2.1798,
2359
+ "step": 332
2360
+ },
2361
+ {
2362
+ "epoch": 2.685483870967742,
2363
+ "grad_norm": 9.052068710327148,
2364
+ "learning_rate": 1.1479734385036742e-05,
2365
+ "loss": 0.3699,
2366
+ "step": 333
2367
+ },
2368
+ {
2369
+ "epoch": 2.693548387096774,
2370
+ "grad_norm": 120.23004150390625,
2371
+ "learning_rate": 1.1439735310524419e-05,
2372
+ "loss": 4.8277,
2373
+ "step": 334
2374
+ },
2375
+ {
2376
+ "epoch": 2.7016129032258065,
2377
+ "grad_norm": 86.39726257324219,
2378
+ "learning_rate": 1.1399736236012096e-05,
2379
+ "loss": 0.7964,
2380
+ "step": 335
2381
+ },
2382
+ {
2383
+ "epoch": 2.709677419354839,
2384
+ "grad_norm": 43.46727752685547,
2385
+ "learning_rate": 1.1359737161499773e-05,
2386
+ "loss": 0.4678,
2387
+ "step": 336
2388
+ },
2389
+ {
2390
+ "epoch": 2.717741935483871,
2391
+ "grad_norm": 92.9560546875,
2392
+ "learning_rate": 1.131973808698745e-05,
2393
+ "loss": 1.9784,
2394
+ "step": 337
2395
+ },
2396
+ {
2397
+ "epoch": 2.725806451612903,
2398
+ "grad_norm": 72.29792785644531,
2399
+ "learning_rate": 1.1279739012475126e-05,
2400
+ "loss": 1.1618,
2401
+ "step": 338
2402
+ },
2403
+ {
2404
+ "epoch": 2.7338709677419355,
2405
+ "grad_norm": 108.34342193603516,
2406
+ "learning_rate": 1.1239739937962803e-05,
2407
+ "loss": 8.9934,
2408
+ "step": 339
2409
+ },
2410
+ {
2411
+ "epoch": 2.741935483870968,
2412
+ "grad_norm": 44.143707275390625,
2413
+ "learning_rate": 1.119974086345048e-05,
2414
+ "loss": 0.7319,
2415
+ "step": 340
2416
+ },
2417
+ {
2418
+ "epoch": 2.75,
2419
+ "grad_norm": 14.93342113494873,
2420
+ "learning_rate": 1.1159741788938157e-05,
2421
+ "loss": 0.2592,
2422
+ "step": 341
2423
+ },
2424
+ {
2425
+ "epoch": 2.758064516129032,
2426
+ "grad_norm": 213.9510040283203,
2427
+ "learning_rate": 1.1119742714425832e-05,
2428
+ "loss": 12.5033,
2429
+ "step": 342
2430
+ },
2431
+ {
2432
+ "epoch": 2.7661290322580645,
2433
+ "grad_norm": 114.5474624633789,
2434
+ "learning_rate": 1.1079743639913511e-05,
2435
+ "loss": 1.5119,
2436
+ "step": 343
2437
+ },
2438
+ {
2439
+ "epoch": 2.774193548387097,
2440
+ "grad_norm": 22.315725326538086,
2441
+ "learning_rate": 1.1039744565401188e-05,
2442
+ "loss": 0.6701,
2443
+ "step": 344
2444
+ },
2445
+ {
2446
+ "epoch": 2.782258064516129,
2447
+ "grad_norm": 22.660259246826172,
2448
+ "learning_rate": 1.0999745490888865e-05,
2449
+ "loss": 0.7027,
2450
+ "step": 345
2451
+ },
2452
+ {
2453
+ "epoch": 2.790322580645161,
2454
+ "grad_norm": 14.709903717041016,
2455
+ "learning_rate": 1.0959746416376542e-05,
2456
+ "loss": 0.213,
2457
+ "step": 346
2458
+ },
2459
+ {
2460
+ "epoch": 2.7983870967741935,
2461
+ "grad_norm": 89.29466247558594,
2462
+ "learning_rate": 1.0919747341864217e-05,
2463
+ "loss": 1.5922,
2464
+ "step": 347
2465
+ },
2466
+ {
2467
+ "epoch": 2.806451612903226,
2468
+ "grad_norm": 13.790899276733398,
2469
+ "learning_rate": 1.0879748267351896e-05,
2470
+ "loss": 0.4771,
2471
+ "step": 348
2472
+ },
2473
+ {
2474
+ "epoch": 2.814516129032258,
2475
+ "grad_norm": 67.785888671875,
2476
+ "learning_rate": 1.0839749192839571e-05,
2477
+ "loss": 5.3096,
2478
+ "step": 349
2479
+ },
2480
+ {
2481
+ "epoch": 2.8225806451612905,
2482
+ "grad_norm": 70.6512222290039,
2483
+ "learning_rate": 1.0799750118327248e-05,
2484
+ "loss": 0.8779,
2485
+ "step": 350
2486
+ },
2487
+ {
2488
+ "epoch": 2.8306451612903225,
2489
+ "grad_norm": 51.96946334838867,
2490
+ "learning_rate": 1.0759751043814926e-05,
2491
+ "loss": 1.0843,
2492
+ "step": 351
2493
+ },
2494
+ {
2495
+ "epoch": 2.838709677419355,
2496
+ "grad_norm": 112.01322937011719,
2497
+ "learning_rate": 1.0719751969302602e-05,
2498
+ "loss": 4.3431,
2499
+ "step": 352
2500
+ },
2501
+ {
2502
+ "epoch": 2.846774193548387,
2503
+ "grad_norm": 30.747026443481445,
2504
+ "learning_rate": 1.067975289479028e-05,
2505
+ "loss": 0.6655,
2506
+ "step": 353
2507
+ },
2508
+ {
2509
+ "epoch": 2.8548387096774195,
2510
+ "grad_norm": 152.95237731933594,
2511
+ "learning_rate": 1.0639753820277956e-05,
2512
+ "loss": 7.7018,
2513
+ "step": 354
2514
+ },
2515
+ {
2516
+ "epoch": 2.8629032258064515,
2517
+ "grad_norm": 97.97068786621094,
2518
+ "learning_rate": 1.0599754745765632e-05,
2519
+ "loss": 6.0524,
2520
+ "step": 355
2521
+ },
2522
+ {
2523
+ "epoch": 2.870967741935484,
2524
+ "grad_norm": 24.1806583404541,
2525
+ "learning_rate": 1.055975567125331e-05,
2526
+ "loss": 0.4385,
2527
+ "step": 356
2528
+ },
2529
+ {
2530
+ "epoch": 2.879032258064516,
2531
+ "grad_norm": 125.32524871826172,
2532
+ "learning_rate": 1.0519756596740986e-05,
2533
+ "loss": 15.97,
2534
+ "step": 357
2535
+ },
2536
+ {
2537
+ "epoch": 2.8870967741935485,
2538
+ "grad_norm": 13.418540000915527,
2539
+ "learning_rate": 1.0479757522228663e-05,
2540
+ "loss": 0.1098,
2541
+ "step": 358
2542
+ },
2543
+ {
2544
+ "epoch": 2.8951612903225805,
2545
+ "grad_norm": 18.43660545349121,
2546
+ "learning_rate": 1.043975844771634e-05,
2547
+ "loss": 0.6929,
2548
+ "step": 359
2549
+ },
2550
+ {
2551
+ "epoch": 2.903225806451613,
2552
+ "grad_norm": 14.304896354675293,
2553
+ "learning_rate": 1.0399759373204017e-05,
2554
+ "loss": 0.6683,
2555
+ "step": 360
2556
+ },
2557
+ {
2558
+ "epoch": 2.911290322580645,
2559
+ "grad_norm": 124.25679016113281,
2560
+ "learning_rate": 1.0359760298691694e-05,
2561
+ "loss": 14.7953,
2562
+ "step": 361
2563
+ },
2564
+ {
2565
+ "epoch": 2.9193548387096775,
2566
+ "grad_norm": 27.41398048400879,
2567
+ "learning_rate": 1.0319761224179371e-05,
2568
+ "loss": 0.4546,
2569
+ "step": 362
2570
+ },
2571
+ {
2572
+ "epoch": 2.9274193548387095,
2573
+ "grad_norm": 17.463279724121094,
2574
+ "learning_rate": 1.0279762149667048e-05,
2575
+ "loss": 0.2985,
2576
+ "step": 363
2577
+ },
2578
+ {
2579
+ "epoch": 2.935483870967742,
2580
+ "grad_norm": 36.6607666015625,
2581
+ "learning_rate": 1.0239763075154725e-05,
2582
+ "loss": 0.6629,
2583
+ "step": 364
2584
+ },
2585
+ {
2586
+ "epoch": 2.943548387096774,
2587
+ "grad_norm": 36.471588134765625,
2588
+ "learning_rate": 1.0199764000642402e-05,
2589
+ "loss": 0.5639,
2590
+ "step": 365
2591
+ },
2592
+ {
2593
+ "epoch": 2.9516129032258065,
2594
+ "grad_norm": 17.774343490600586,
2595
+ "learning_rate": 1.0159764926130079e-05,
2596
+ "loss": 0.3925,
2597
+ "step": 366
2598
+ },
2599
+ {
2600
+ "epoch": 2.959677419354839,
2601
+ "grad_norm": 32.37789535522461,
2602
+ "learning_rate": 1.0119765851617756e-05,
2603
+ "loss": 0.5822,
2604
+ "step": 367
2605
+ },
2606
+ {
2607
+ "epoch": 2.967741935483871,
2608
+ "grad_norm": 45.42538070678711,
2609
+ "learning_rate": 1.0079766777105432e-05,
2610
+ "loss": 0.7533,
2611
+ "step": 368
2612
+ },
2613
+ {
2614
+ "epoch": 2.975806451612903,
2615
+ "grad_norm": 26.0896053314209,
2616
+ "learning_rate": 1.003976770259311e-05,
2617
+ "loss": 0.3197,
2618
+ "step": 369
2619
+ },
2620
+ {
2621
+ "epoch": 2.9838709677419355,
2622
+ "grad_norm": 5.230876445770264,
2623
+ "learning_rate": 9.999768628080785e-06,
2624
+ "loss": 0.2765,
2625
+ "step": 370
2626
+ },
2627
+ {
2628
+ "epoch": 2.991935483870968,
2629
+ "grad_norm": 50.81292724609375,
2630
+ "learning_rate": 9.959769553568463e-06,
2631
+ "loss": 2.345,
2632
+ "step": 371
2633
+ },
2634
+ {
2635
+ "epoch": 3.0,
2636
+ "grad_norm": 310.2722473144531,
2637
+ "learning_rate": 9.91977047905614e-06,
2638
+ "loss": 24.6264,
2639
+ "step": 372
2640
+ },
2641
+ {
2642
+ "epoch": 3.0,
2643
+ "eval_loss": 3.3316104412078857,
2644
+ "eval_mae": 1.1439374685287476,
2645
+ "eval_mse": 3.3316097259521484,
2646
+ "eval_r2": 0.08786314725875854,
2647
+ "eval_rmse": 1.8252697679938021,
2648
+ "eval_runtime": 1.3357,
2649
+ "eval_samples_per_second": 41.176,
2650
+ "eval_smape": 51.89841985702515,
2651
+ "eval_steps_per_second": 10.481,
2652
+ "step": 372
2653
+ },
2654
+ {
2655
+ "epoch": 3.0080645161290325,
2656
+ "grad_norm": 41.96479415893555,
2657
+ "learning_rate": 9.879771404543815e-06,
2658
+ "loss": 1.2152,
2659
+ "step": 373
2660
+ },
2661
+ {
2662
+ "epoch": 3.0161290322580645,
2663
+ "grad_norm": 136.83741760253906,
2664
+ "learning_rate": 9.839772330031494e-06,
2665
+ "loss": 3.896,
2666
+ "step": 374
2667
+ },
2668
+ {
2669
+ "epoch": 3.024193548387097,
2670
+ "grad_norm": 22.60567283630371,
2671
+ "learning_rate": 9.79977325551917e-06,
2672
+ "loss": 0.3975,
2673
+ "step": 375
2674
+ },
2675
+ {
2676
+ "epoch": 3.032258064516129,
2677
+ "grad_norm": 39.60881805419922,
2678
+ "learning_rate": 9.759774181006848e-06,
2679
+ "loss": 1.4481,
2680
+ "step": 376
2681
+ },
2682
+ {
2683
+ "epoch": 3.0403225806451615,
2684
+ "grad_norm": 7.505491256713867,
2685
+ "learning_rate": 9.719775106494523e-06,
2686
+ "loss": 0.0605,
2687
+ "step": 377
2688
+ },
2689
+ {
2690
+ "epoch": 3.0483870967741935,
2691
+ "grad_norm": 53.053001403808594,
2692
+ "learning_rate": 9.6797760319822e-06,
2693
+ "loss": 1.7315,
2694
+ "step": 378
2695
+ },
2696
+ {
2697
+ "epoch": 3.056451612903226,
2698
+ "grad_norm": 33.12321853637695,
2699
+ "learning_rate": 9.639776957469879e-06,
2700
+ "loss": 0.5669,
2701
+ "step": 379
2702
+ },
2703
+ {
2704
+ "epoch": 3.064516129032258,
2705
+ "grad_norm": 13.300590515136719,
2706
+ "learning_rate": 9.599777882957554e-06,
2707
+ "loss": 0.4431,
2708
+ "step": 380
2709
+ },
2710
+ {
2711
+ "epoch": 3.0725806451612905,
2712
+ "grad_norm": 104.96222686767578,
2713
+ "learning_rate": 9.55977880844523e-06,
2714
+ "loss": 5.6698,
2715
+ "step": 381
2716
+ },
2717
+ {
2718
+ "epoch": 3.0806451612903225,
2719
+ "grad_norm": 49.31301498413086,
2720
+ "learning_rate": 9.519779733932908e-06,
2721
+ "loss": 1.3213,
2722
+ "step": 382
2723
+ },
2724
+ {
2725
+ "epoch": 3.088709677419355,
2726
+ "grad_norm": 16.676259994506836,
2727
+ "learning_rate": 9.479780659420585e-06,
2728
+ "loss": 0.266,
2729
+ "step": 383
2730
+ },
2731
+ {
2732
+ "epoch": 3.096774193548387,
2733
+ "grad_norm": 34.20507049560547,
2734
+ "learning_rate": 9.439781584908261e-06,
2735
+ "loss": 1.5853,
2736
+ "step": 384
2737
+ },
2738
+ {
2739
+ "epoch": 3.1048387096774195,
2740
+ "grad_norm": 39.18606948852539,
2741
+ "learning_rate": 9.399782510395938e-06,
2742
+ "loss": 1.0645,
2743
+ "step": 385
2744
+ },
2745
+ {
2746
+ "epoch": 3.1129032258064515,
2747
+ "grad_norm": 5.354783535003662,
2748
+ "learning_rate": 9.359783435883615e-06,
2749
+ "loss": 0.1131,
2750
+ "step": 386
2751
+ },
2752
+ {
2753
+ "epoch": 3.120967741935484,
2754
+ "grad_norm": 41.82234573364258,
2755
+ "learning_rate": 9.319784361371292e-06,
2756
+ "loss": 0.5984,
2757
+ "step": 387
2758
+ },
2759
+ {
2760
+ "epoch": 3.129032258064516,
2761
+ "grad_norm": 16.759111404418945,
2762
+ "learning_rate": 9.27978528685897e-06,
2763
+ "loss": 0.4225,
2764
+ "step": 388
2765
+ },
2766
+ {
2767
+ "epoch": 3.1370967741935485,
2768
+ "grad_norm": 24.134700775146484,
2769
+ "learning_rate": 9.239786212346646e-06,
2770
+ "loss": 0.434,
2771
+ "step": 389
2772
+ },
2773
+ {
2774
+ "epoch": 3.1451612903225805,
2775
+ "grad_norm": 13.111348152160645,
2776
+ "learning_rate": 9.199787137834323e-06,
2777
+ "loss": 0.2413,
2778
+ "step": 390
2779
+ },
2780
+ {
2781
+ "epoch": 3.153225806451613,
2782
+ "grad_norm": 115.44222259521484,
2783
+ "learning_rate": 9.159788063322e-06,
2784
+ "loss": 8.9723,
2785
+ "step": 391
2786
+ },
2787
+ {
2788
+ "epoch": 3.161290322580645,
2789
+ "grad_norm": 13.346231460571289,
2790
+ "learning_rate": 9.119788988809677e-06,
2791
+ "loss": 0.8709,
2792
+ "step": 392
2793
+ },
2794
+ {
2795
+ "epoch": 3.1693548387096775,
2796
+ "grad_norm": 1.1584579944610596,
2797
+ "learning_rate": 9.079789914297354e-06,
2798
+ "loss": 0.0055,
2799
+ "step": 393
2800
+ },
2801
+ {
2802
+ "epoch": 3.1774193548387095,
2803
+ "grad_norm": 70.0774154663086,
2804
+ "learning_rate": 9.03979083978503e-06,
2805
+ "loss": 0.9703,
2806
+ "step": 394
2807
+ },
2808
+ {
2809
+ "epoch": 3.185483870967742,
2810
+ "grad_norm": 14.28967571258545,
2811
+ "learning_rate": 8.999791765272708e-06,
2812
+ "loss": 0.2817,
2813
+ "step": 395
2814
+ },
2815
+ {
2816
+ "epoch": 3.193548387096774,
2817
+ "grad_norm": 4.5178303718566895,
2818
+ "learning_rate": 8.959792690760383e-06,
2819
+ "loss": 0.0217,
2820
+ "step": 396
2821
+ },
2822
+ {
2823
+ "epoch": 3.2016129032258065,
2824
+ "grad_norm": 9.761589050292969,
2825
+ "learning_rate": 8.919793616248061e-06,
2826
+ "loss": 0.4557,
2827
+ "step": 397
2828
+ },
2829
+ {
2830
+ "epoch": 3.2096774193548385,
2831
+ "grad_norm": 53.35697555541992,
2832
+ "learning_rate": 8.879794541735738e-06,
2833
+ "loss": 0.8752,
2834
+ "step": 398
2835
+ },
2836
+ {
2837
+ "epoch": 3.217741935483871,
2838
+ "grad_norm": 165.54457092285156,
2839
+ "learning_rate": 8.839795467223414e-06,
2840
+ "loss": 14.6604,
2841
+ "step": 399
2842
+ },
2843
+ {
2844
+ "epoch": 3.225806451612903,
2845
+ "grad_norm": 247.7133331298828,
2846
+ "learning_rate": 8.799796392711092e-06,
2847
+ "loss": 9.448,
2848
+ "step": 400
2849
+ },
2850
+ {
2851
+ "epoch": 3.2338709677419355,
2852
+ "grad_norm": 53.525306701660156,
2853
+ "learning_rate": 8.759797318198767e-06,
2854
+ "loss": 0.724,
2855
+ "step": 401
2856
+ },
2857
+ {
2858
+ "epoch": 3.241935483870968,
2859
+ "grad_norm": 172.49952697753906,
2860
+ "learning_rate": 8.719798243686446e-06,
2861
+ "loss": 6.2532,
2862
+ "step": 402
2863
+ },
2864
+ {
2865
+ "epoch": 3.25,
2866
+ "grad_norm": 99.93900299072266,
2867
+ "learning_rate": 8.679799169174121e-06,
2868
+ "loss": 2.2522,
2869
+ "step": 403
2870
+ },
2871
+ {
2872
+ "epoch": 3.258064516129032,
2873
+ "grad_norm": 15.439353942871094,
2874
+ "learning_rate": 8.639800094661798e-06,
2875
+ "loss": 0.1913,
2876
+ "step": 404
2877
+ },
2878
+ {
2879
+ "epoch": 3.2661290322580645,
2880
+ "grad_norm": 49.3162956237793,
2881
+ "learning_rate": 8.599801020149477e-06,
2882
+ "loss": 0.4811,
2883
+ "step": 405
2884
+ },
2885
+ {
2886
+ "epoch": 3.274193548387097,
2887
+ "grad_norm": 184.40086364746094,
2888
+ "learning_rate": 8.559801945637152e-06,
2889
+ "loss": 7.606,
2890
+ "step": 406
2891
+ },
2892
+ {
2893
+ "epoch": 3.282258064516129,
2894
+ "grad_norm": 132.2401580810547,
2895
+ "learning_rate": 8.519802871124829e-06,
2896
+ "loss": 3.4354,
2897
+ "step": 407
2898
+ },
2899
+ {
2900
+ "epoch": 3.2903225806451615,
2901
+ "grad_norm": 64.83243560791016,
2902
+ "learning_rate": 8.479803796612506e-06,
2903
+ "loss": 2.5171,
2904
+ "step": 408
2905
+ },
2906
+ {
2907
+ "epoch": 3.2983870967741935,
2908
+ "grad_norm": 158.75596618652344,
2909
+ "learning_rate": 8.439804722100183e-06,
2910
+ "loss": 2.1743,
2911
+ "step": 409
2912
+ },
2913
+ {
2914
+ "epoch": 3.306451612903226,
2915
+ "grad_norm": 103.98823547363281,
2916
+ "learning_rate": 8.39980564758786e-06,
2917
+ "loss": 1.4439,
2918
+ "step": 410
2919
+ },
2920
+ {
2921
+ "epoch": 3.314516129032258,
2922
+ "grad_norm": 43.638450622558594,
2923
+ "learning_rate": 8.359806573075537e-06,
2924
+ "loss": 0.573,
2925
+ "step": 411
2926
+ },
2927
+ {
2928
+ "epoch": 3.3225806451612905,
2929
+ "grad_norm": 41.889686584472656,
2930
+ "learning_rate": 8.319807498563214e-06,
2931
+ "loss": 1.4203,
2932
+ "step": 412
2933
+ },
2934
+ {
2935
+ "epoch": 3.3306451612903225,
2936
+ "grad_norm": 6.998268127441406,
2937
+ "learning_rate": 8.27980842405089e-06,
2938
+ "loss": 0.0949,
2939
+ "step": 413
2940
+ },
2941
+ {
2942
+ "epoch": 3.338709677419355,
2943
+ "grad_norm": 167.7267303466797,
2944
+ "learning_rate": 8.239809349538567e-06,
2945
+ "loss": 3.6027,
2946
+ "step": 414
2947
+ },
2948
+ {
2949
+ "epoch": 3.346774193548387,
2950
+ "grad_norm": 58.6676025390625,
2951
+ "learning_rate": 8.199810275026244e-06,
2952
+ "loss": 1.0112,
2953
+ "step": 415
2954
+ },
2955
+ {
2956
+ "epoch": 3.3548387096774195,
2957
+ "grad_norm": 19.077991485595703,
2958
+ "learning_rate": 8.159811200513921e-06,
2959
+ "loss": 0.1047,
2960
+ "step": 416
2961
+ },
2962
+ {
2963
+ "epoch": 3.3629032258064515,
2964
+ "grad_norm": 61.30634307861328,
2965
+ "learning_rate": 8.119812126001598e-06,
2966
+ "loss": 0.3403,
2967
+ "step": 417
2968
+ },
2969
+ {
2970
+ "epoch": 3.370967741935484,
2971
+ "grad_norm": 40.68547058105469,
2972
+ "learning_rate": 8.079813051489275e-06,
2973
+ "loss": 0.3131,
2974
+ "step": 418
2975
+ },
2976
+ {
2977
+ "epoch": 3.379032258064516,
2978
+ "grad_norm": 39.6519660949707,
2979
+ "learning_rate": 8.039813976976952e-06,
2980
+ "loss": 0.0895,
2981
+ "step": 419
2982
+ },
2983
+ {
2984
+ "epoch": 3.3870967741935485,
2985
+ "grad_norm": 54.81489181518555,
2986
+ "learning_rate": 7.999814902464629e-06,
2987
+ "loss": 0.9378,
2988
+ "step": 420
2989
+ },
2990
+ {
2991
+ "epoch": 3.3951612903225805,
2992
+ "grad_norm": 43.20634460449219,
2993
+ "learning_rate": 7.959815827952306e-06,
2994
+ "loss": 0.5461,
2995
+ "step": 421
2996
+ },
2997
+ {
2998
+ "epoch": 3.403225806451613,
2999
+ "grad_norm": 35.164852142333984,
3000
+ "learning_rate": 7.919816753439981e-06,
3001
+ "loss": 1.0319,
3002
+ "step": 422
3003
+ },
3004
+ {
3005
+ "epoch": 3.411290322580645,
3006
+ "grad_norm": 283.366455078125,
3007
+ "learning_rate": 7.87981767892766e-06,
3008
+ "loss": 8.4101,
3009
+ "step": 423
3010
+ },
3011
+ {
3012
+ "epoch": 3.4193548387096775,
3013
+ "grad_norm": 507.85150146484375,
3014
+ "learning_rate": 7.839818604415335e-06,
3015
+ "loss": 17.8365,
3016
+ "step": 424
3017
+ },
3018
+ {
3019
+ "epoch": 3.4274193548387095,
3020
+ "grad_norm": 8.4439697265625,
3021
+ "learning_rate": 7.799819529903014e-06,
3022
+ "loss": 0.2679,
3023
+ "step": 425
3024
+ },
3025
+ {
3026
+ "epoch": 3.435483870967742,
3027
+ "grad_norm": 40.22681427001953,
3028
+ "learning_rate": 7.75982045539069e-06,
3029
+ "loss": 0.3714,
3030
+ "step": 426
3031
+ },
3032
+ {
3033
+ "epoch": 3.443548387096774,
3034
+ "grad_norm": 32.92212677001953,
3035
+ "learning_rate": 7.719821380878366e-06,
3036
+ "loss": 0.4915,
3037
+ "step": 427
3038
+ },
3039
+ {
3040
+ "epoch": 3.4516129032258065,
3041
+ "grad_norm": 26.323545455932617,
3042
+ "learning_rate": 7.679822306366044e-06,
3043
+ "loss": 0.4087,
3044
+ "step": 428
3045
+ },
3046
+ {
3047
+ "epoch": 3.4596774193548385,
3048
+ "grad_norm": 146.9882049560547,
3049
+ "learning_rate": 7.63982323185372e-06,
3050
+ "loss": 0.2726,
3051
+ "step": 429
3052
+ },
3053
+ {
3054
+ "epoch": 3.467741935483871,
3055
+ "grad_norm": 4.548903465270996,
3056
+ "learning_rate": 7.599824157341397e-06,
3057
+ "loss": 0.1744,
3058
+ "step": 430
3059
+ },
3060
+ {
3061
+ "epoch": 3.475806451612903,
3062
+ "grad_norm": 228.15928649902344,
3063
+ "learning_rate": 7.5598250828290735e-06,
3064
+ "loss": 10.8116,
3065
+ "step": 431
3066
+ },
3067
+ {
3068
+ "epoch": 3.4838709677419355,
3069
+ "grad_norm": 60.10261154174805,
3070
+ "learning_rate": 7.51982600831675e-06,
3071
+ "loss": 0.6974,
3072
+ "step": 432
3073
+ },
3074
+ {
3075
+ "epoch": 3.491935483870968,
3076
+ "grad_norm": 13.38789176940918,
3077
+ "learning_rate": 7.479826933804428e-06,
3078
+ "loss": 0.4782,
3079
+ "step": 433
3080
+ },
3081
+ {
3082
+ "epoch": 3.5,
3083
+ "grad_norm": 68.60205841064453,
3084
+ "learning_rate": 7.439827859292104e-06,
3085
+ "loss": 1.7533,
3086
+ "step": 434
3087
+ },
3088
+ {
3089
+ "epoch": 3.508064516129032,
3090
+ "grad_norm": 6.820003986358643,
3091
+ "learning_rate": 7.399828784779782e-06,
3092
+ "loss": 0.0396,
3093
+ "step": 435
3094
+ },
3095
+ {
3096
+ "epoch": 3.5161290322580645,
3097
+ "grad_norm": 6.984181880950928,
3098
+ "learning_rate": 7.359829710267458e-06,
3099
+ "loss": 0.0715,
3100
+ "step": 436
3101
+ },
3102
+ {
3103
+ "epoch": 3.524193548387097,
3104
+ "grad_norm": 26.827526092529297,
3105
+ "learning_rate": 7.319830635755135e-06,
3106
+ "loss": 0.1687,
3107
+ "step": 437
3108
+ },
3109
+ {
3110
+ "epoch": 3.532258064516129,
3111
+ "grad_norm": 8.610593795776367,
3112
+ "learning_rate": 7.279831561242811e-06,
3113
+ "loss": 0.134,
3114
+ "step": 438
3115
+ },
3116
+ {
3117
+ "epoch": 3.540322580645161,
3118
+ "grad_norm": 39.13728332519531,
3119
+ "learning_rate": 7.239832486730489e-06,
3120
+ "loss": 0.44,
3121
+ "step": 439
3122
+ },
3123
+ {
3124
+ "epoch": 3.5483870967741935,
3125
+ "grad_norm": 56.36937713623047,
3126
+ "learning_rate": 7.199833412218167e-06,
3127
+ "loss": 3.6674,
3128
+ "step": 440
3129
+ },
3130
+ {
3131
+ "epoch": 3.556451612903226,
3132
+ "grad_norm": 38.50959396362305,
3133
+ "learning_rate": 7.159834337705843e-06,
3134
+ "loss": 0.7563,
3135
+ "step": 441
3136
+ },
3137
+ {
3138
+ "epoch": 3.564516129032258,
3139
+ "grad_norm": 44.96533203125,
3140
+ "learning_rate": 7.11983526319352e-06,
3141
+ "loss": 0.7596,
3142
+ "step": 442
3143
+ },
3144
+ {
3145
+ "epoch": 3.5725806451612905,
3146
+ "grad_norm": 59.28168487548828,
3147
+ "learning_rate": 7.079836188681196e-06,
3148
+ "loss": 0.5658,
3149
+ "step": 443
3150
+ },
3151
+ {
3152
+ "epoch": 3.5806451612903225,
3153
+ "grad_norm": 24.149091720581055,
3154
+ "learning_rate": 7.0398371141688735e-06,
3155
+ "loss": 0.3909,
3156
+ "step": 444
3157
+ },
3158
+ {
3159
+ "epoch": 3.588709677419355,
3160
+ "grad_norm": 44.97443389892578,
3161
+ "learning_rate": 6.99983803965655e-06,
3162
+ "loss": 0.8401,
3163
+ "step": 445
3164
+ },
3165
+ {
3166
+ "epoch": 3.596774193548387,
3167
+ "grad_norm": 210.90919494628906,
3168
+ "learning_rate": 6.9598389651442265e-06,
3169
+ "loss": 8.9542,
3170
+ "step": 446
3171
+ },
3172
+ {
3173
+ "epoch": 3.6048387096774195,
3174
+ "grad_norm": 78.13578796386719,
3175
+ "learning_rate": 6.919839890631904e-06,
3176
+ "loss": 7.4033,
3177
+ "step": 447
3178
+ },
3179
+ {
3180
+ "epoch": 3.6129032258064515,
3181
+ "grad_norm": 54.309566497802734,
3182
+ "learning_rate": 6.87984081611958e-06,
3183
+ "loss": 0.8781,
3184
+ "step": 448
3185
+ },
3186
+ {
3187
+ "epoch": 3.620967741935484,
3188
+ "grad_norm": 8.429338455200195,
3189
+ "learning_rate": 6.839841741607258e-06,
3190
+ "loss": 0.2823,
3191
+ "step": 449
3192
+ },
3193
+ {
3194
+ "epoch": 3.629032258064516,
3195
+ "grad_norm": 32.88882827758789,
3196
+ "learning_rate": 6.799842667094934e-06,
3197
+ "loss": 1.153,
3198
+ "step": 450
3199
+ },
3200
+ {
3201
+ "epoch": 3.6370967741935485,
3202
+ "grad_norm": 3.0679757595062256,
3203
+ "learning_rate": 6.759843592582611e-06,
3204
+ "loss": 0.0494,
3205
+ "step": 451
3206
+ },
3207
+ {
3208
+ "epoch": 3.6451612903225805,
3209
+ "grad_norm": 17.686809539794922,
3210
+ "learning_rate": 6.719844518070289e-06,
3211
+ "loss": 0.1084,
3212
+ "step": 452
3213
+ },
3214
+ {
3215
+ "epoch": 3.653225806451613,
3216
+ "grad_norm": 48.20186996459961,
3217
+ "learning_rate": 6.679845443557965e-06,
3218
+ "loss": 0.7689,
3219
+ "step": 453
3220
+ },
3221
+ {
3222
+ "epoch": 3.661290322580645,
3223
+ "grad_norm": 134.13829040527344,
3224
+ "learning_rate": 6.639846369045642e-06,
3225
+ "loss": 12.3841,
3226
+ "step": 454
3227
+ },
3228
+ {
3229
+ "epoch": 3.6693548387096775,
3230
+ "grad_norm": 244.57977294921875,
3231
+ "learning_rate": 6.599847294533318e-06,
3232
+ "loss": 8.8699,
3233
+ "step": 455
3234
+ },
3235
+ {
3236
+ "epoch": 3.6774193548387095,
3237
+ "grad_norm": 218.47779846191406,
3238
+ "learning_rate": 6.559848220020996e-06,
3239
+ "loss": 4.3449,
3240
+ "step": 456
3241
+ },
3242
+ {
3243
+ "epoch": 3.685483870967742,
3244
+ "grad_norm": 4.526096820831299,
3245
+ "learning_rate": 6.519849145508672e-06,
3246
+ "loss": 0.0972,
3247
+ "step": 457
3248
+ },
3249
+ {
3250
+ "epoch": 3.693548387096774,
3251
+ "grad_norm": 134.6807861328125,
3252
+ "learning_rate": 6.4798500709963495e-06,
3253
+ "loss": 5.5296,
3254
+ "step": 458
3255
+ },
3256
+ {
3257
+ "epoch": 3.7016129032258065,
3258
+ "grad_norm": 16.304237365722656,
3259
+ "learning_rate": 6.4398509964840265e-06,
3260
+ "loss": 0.2073,
3261
+ "step": 459
3262
+ },
3263
+ {
3264
+ "epoch": 3.709677419354839,
3265
+ "grad_norm": 98.8707275390625,
3266
+ "learning_rate": 6.3998519219717025e-06,
3267
+ "loss": 0.9317,
3268
+ "step": 460
3269
+ },
3270
+ {
3271
+ "epoch": 3.717741935483871,
3272
+ "grad_norm": 38.19011688232422,
3273
+ "learning_rate": 6.35985284745938e-06,
3274
+ "loss": 0.5302,
3275
+ "step": 461
3276
+ },
3277
+ {
3278
+ "epoch": 3.725806451612903,
3279
+ "grad_norm": 56.57841110229492,
3280
+ "learning_rate": 6.319853772947056e-06,
3281
+ "loss": 1.2759,
3282
+ "step": 462
3283
+ },
3284
+ {
3285
+ "epoch": 3.7338709677419355,
3286
+ "grad_norm": 24.514759063720703,
3287
+ "learning_rate": 6.279854698434733e-06,
3288
+ "loss": 0.2036,
3289
+ "step": 463
3290
+ },
3291
+ {
3292
+ "epoch": 3.741935483870968,
3293
+ "grad_norm": 191.78668212890625,
3294
+ "learning_rate": 6.239855623922409e-06,
3295
+ "loss": 8.38,
3296
+ "step": 464
3297
+ },
3298
+ {
3299
+ "epoch": 3.75,
3300
+ "grad_norm": 7.4319376945495605,
3301
+ "learning_rate": 6.199856549410087e-06,
3302
+ "loss": 0.3825,
3303
+ "step": 465
3304
+ },
3305
+ {
3306
+ "epoch": 3.758064516129032,
3307
+ "grad_norm": 520.401123046875,
3308
+ "learning_rate": 6.159857474897764e-06,
3309
+ "loss": 115.6645,
3310
+ "step": 466
3311
+ },
3312
+ {
3313
+ "epoch": 3.7661290322580645,
3314
+ "grad_norm": 46.42934036254883,
3315
+ "learning_rate": 6.119858400385441e-06,
3316
+ "loss": 0.5399,
3317
+ "step": 467
3318
+ },
3319
+ {
3320
+ "epoch": 3.774193548387097,
3321
+ "grad_norm": 190.69224548339844,
3322
+ "learning_rate": 6.079859325873117e-06,
3323
+ "loss": 4.1707,
3324
+ "step": 468
3325
+ },
3326
+ {
3327
+ "epoch": 3.782258064516129,
3328
+ "grad_norm": 23.724159240722656,
3329
+ "learning_rate": 6.039860251360795e-06,
3330
+ "loss": 0.4132,
3331
+ "step": 469
3332
+ },
3333
+ {
3334
+ "epoch": 3.790322580645161,
3335
+ "grad_norm": 13.517512321472168,
3336
+ "learning_rate": 5.999861176848472e-06,
3337
+ "loss": 0.0815,
3338
+ "step": 470
3339
+ },
3340
+ {
3341
+ "epoch": 3.7983870967741935,
3342
+ "grad_norm": 70.30924987792969,
3343
+ "learning_rate": 5.959862102336149e-06,
3344
+ "loss": 2.7397,
3345
+ "step": 471
3346
+ },
3347
+ {
3348
+ "epoch": 3.806451612903226,
3349
+ "grad_norm": 16.972890853881836,
3350
+ "learning_rate": 5.919863027823825e-06,
3351
+ "loss": 0.0473,
3352
+ "step": 472
3353
+ },
3354
+ {
3355
+ "epoch": 3.814516129032258,
3356
+ "grad_norm": 56.665103912353516,
3357
+ "learning_rate": 5.879863953311502e-06,
3358
+ "loss": 0.5666,
3359
+ "step": 473
3360
+ },
3361
+ {
3362
+ "epoch": 3.8225806451612905,
3363
+ "grad_norm": 38.07454299926758,
3364
+ "learning_rate": 5.839864878799179e-06,
3365
+ "loss": 0.3797,
3366
+ "step": 474
3367
+ },
3368
+ {
3369
+ "epoch": 3.8306451612903225,
3370
+ "grad_norm": 63.56098937988281,
3371
+ "learning_rate": 5.7998658042868555e-06,
3372
+ "loss": 1.4492,
3373
+ "step": 475
3374
+ },
3375
+ {
3376
+ "epoch": 3.838709677419355,
3377
+ "grad_norm": 841.0067138671875,
3378
+ "learning_rate": 5.7598667297745325e-06,
3379
+ "loss": 8.2394,
3380
+ "step": 476
3381
+ },
3382
+ {
3383
+ "epoch": 3.846774193548387,
3384
+ "grad_norm": 77.66548919677734,
3385
+ "learning_rate": 5.719867655262209e-06,
3386
+ "loss": 1.4433,
3387
+ "step": 477
3388
+ },
3389
+ {
3390
+ "epoch": 3.8548387096774195,
3391
+ "grad_norm": 130.56881713867188,
3392
+ "learning_rate": 5.679868580749886e-06,
3393
+ "loss": 2.8763,
3394
+ "step": 478
3395
+ },
3396
+ {
3397
+ "epoch": 3.8629032258064515,
3398
+ "grad_norm": 5.675972938537598,
3399
+ "learning_rate": 5.639869506237563e-06,
3400
+ "loss": 0.0891,
3401
+ "step": 479
3402
+ },
3403
+ {
3404
+ "epoch": 3.870967741935484,
3405
+ "grad_norm": 69.9577407836914,
3406
+ "learning_rate": 5.59987043172524e-06,
3407
+ "loss": 2.6558,
3408
+ "step": 480
3409
+ },
3410
+ {
3411
+ "epoch": 3.879032258064516,
3412
+ "grad_norm": 14.687697410583496,
3413
+ "learning_rate": 5.559871357212916e-06,
3414
+ "loss": 0.1182,
3415
+ "step": 481
3416
+ },
3417
+ {
3418
+ "epoch": 3.8870967741935485,
3419
+ "grad_norm": 155.2462921142578,
3420
+ "learning_rate": 5.519872282700594e-06,
3421
+ "loss": 2.598,
3422
+ "step": 482
3423
+ },
3424
+ {
3425
+ "epoch": 3.8951612903225805,
3426
+ "grad_norm": 31.2712345123291,
3427
+ "learning_rate": 5.479873208188271e-06,
3428
+ "loss": 0.7228,
3429
+ "step": 483
3430
+ },
3431
+ {
3432
+ "epoch": 3.903225806451613,
3433
+ "grad_norm": 201.28076171875,
3434
+ "learning_rate": 5.439874133675948e-06,
3435
+ "loss": 12.3541,
3436
+ "step": 484
3437
+ },
3438
+ {
3439
+ "epoch": 3.911290322580645,
3440
+ "grad_norm": 70.37089538574219,
3441
+ "learning_rate": 5.399875059163624e-06,
3442
+ "loss": 0.7182,
3443
+ "step": 485
3444
+ },
3445
+ {
3446
+ "epoch": 3.9193548387096775,
3447
+ "grad_norm": 29.314970016479492,
3448
+ "learning_rate": 5.359875984651301e-06,
3449
+ "loss": 0.1898,
3450
+ "step": 486
3451
+ },
3452
+ {
3453
+ "epoch": 3.9274193548387095,
3454
+ "grad_norm": 25.780534744262695,
3455
+ "learning_rate": 5.319876910138978e-06,
3456
+ "loss": 0.3334,
3457
+ "step": 487
3458
+ },
3459
+ {
3460
+ "epoch": 3.935483870967742,
3461
+ "grad_norm": 8.816149711608887,
3462
+ "learning_rate": 5.279877835626655e-06,
3463
+ "loss": 0.3082,
3464
+ "step": 488
3465
+ },
3466
+ {
3467
+ "epoch": 3.943548387096774,
3468
+ "grad_norm": 348.0280456542969,
3469
+ "learning_rate": 5.239878761114332e-06,
3470
+ "loss": 4.8654,
3471
+ "step": 489
3472
+ },
3473
+ {
3474
+ "epoch": 3.9516129032258065,
3475
+ "grad_norm": 42.4134407043457,
3476
+ "learning_rate": 5.1998796866020085e-06,
3477
+ "loss": 1.2419,
3478
+ "step": 490
3479
+ },
3480
+ {
3481
+ "epoch": 3.959677419354839,
3482
+ "grad_norm": 31.053295135498047,
3483
+ "learning_rate": 5.1598806120896854e-06,
3484
+ "loss": 0.6271,
3485
+ "step": 491
3486
+ },
3487
+ {
3488
+ "epoch": 3.967741935483871,
3489
+ "grad_norm": 70.98059844970703,
3490
+ "learning_rate": 5.119881537577362e-06,
3491
+ "loss": 0.5582,
3492
+ "step": 492
3493
+ },
3494
+ {
3495
+ "epoch": 3.975806451612903,
3496
+ "grad_norm": 118.1554946899414,
3497
+ "learning_rate": 5.079882463065039e-06,
3498
+ "loss": 1.5444,
3499
+ "step": 493
3500
+ },
3501
+ {
3502
+ "epoch": 3.9838709677419355,
3503
+ "grad_norm": 288.1326599121094,
3504
+ "learning_rate": 5.039883388552716e-06,
3505
+ "loss": 7.396,
3506
+ "step": 494
3507
+ },
3508
+ {
3509
+ "epoch": 3.991935483870968,
3510
+ "grad_norm": 19.853105545043945,
3511
+ "learning_rate": 4.999884314040392e-06,
3512
+ "loss": 0.9112,
3513
+ "step": 495
3514
+ },
3515
+ {
3516
+ "epoch": 4.0,
3517
+ "grad_norm": 32.062965393066406,
3518
+ "learning_rate": 4.95988523952807e-06,
3519
+ "loss": 0.2896,
3520
+ "step": 496
3521
+ },
3522
+ {
3523
+ "epoch": 4.0,
3524
+ "eval_loss": 2.353773355484009,
3525
+ "eval_mae": 1.0791250467300415,
3526
+ "eval_mse": 2.353773355484009,
3527
+ "eval_r2": 0.3555777668952942,
3528
+ "eval_rmse": 1.5342012108859804,
3529
+ "eval_runtime": 1.3428,
3530
+ "eval_samples_per_second": 40.961,
3531
+ "eval_smape": 55.77985644340515,
3532
+ "eval_steps_per_second": 10.426,
3533
+ "step": 496
3534
+ }
3535
+ ],
3536
+ "logging_steps": 1,
3537
+ "max_steps": 620,
3538
+ "num_input_tokens_seen": 0,
3539
+ "num_train_epochs": 5,
3540
+ "save_steps": 500,
3541
+ "stateful_callbacks": {
3542
+ "TrainerControl": {
3543
+ "args": {
3544
+ "should_epoch_stop": false,
3545
+ "should_evaluate": false,
3546
+ "should_log": false,
3547
+ "should_save": true,
3548
+ "should_training_stop": false
3549
+ },
3550
+ "attributes": {}
3551
+ }
3552
+ },
3553
+ "total_flos": 2687879255015424.0,
3554
+ "train_batch_size": 4,
3555
+ "trial_name": null,
3556
+ "trial_params": {
3557
+ "learning_rate": 2.479942619764035e-05
3558
+ }
3559
+ }
run-2/checkpoint-496/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ec9e62cbb89b4c2bb9799864adb5aae418df3bf704aaa8403d55649b2c20b42
3
+ size 5432
run-3/checkpoint-248/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-base",
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": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0"
25
+ },
26
+ "initializer_cutoff_factor": 2.0,
27
+ "initializer_range": 0.02,
28
+ "intermediate_size": 1152,
29
+ "label2id": {
30
+ "LABEL_0": 0
31
+ },
32
+ "layer_norm_eps": 1e-05,
33
+ "local_attention": 128,
34
+ "local_rope_theta": 10000.0,
35
+ "max_position_embeddings": 8192,
36
+ "mlp_bias": false,
37
+ "mlp_dropout": 0.0,
38
+ "model_type": "modernbert",
39
+ "norm_bias": false,
40
+ "norm_eps": 1e-05,
41
+ "num_attention_heads": 12,
42
+ "num_hidden_layers": 22,
43
+ "pad_token_id": 50283,
44
+ "position_embedding_type": "absolute",
45
+ "problem_type": "regression",
46
+ "reference_compile": true,
47
+ "sep_token_id": 50282,
48
+ "sparse_pred_ignore_index": -100,
49
+ "sparse_prediction": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.48.0.dev0",
52
+ "vocab_size": 50368
53
+ }
run-3/checkpoint-248/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abae6b5d2928c34ede0bd61ff113f96870236bd353afc3e29e8461749ff35c5c
3
+ size 598436708
run-3/checkpoint-248/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f42e910d4f3668e15b74703716752175f1fb809efa9a74adcc4fef208fe2ac9
3
+ size 1196963194
run-3/checkpoint-248/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c61318d640fc0fcb7f8a9868b31a885fae09243869e3f7810bd29eb4e61b970c
3
+ size 14244
run-3/checkpoint-248/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4531be4f685b4399f473ce9d4f81c5fe03babe483c2340c0581a9372a0ee9f06
3
+ size 1064
run-3/checkpoint-248/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": true,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
run-3/checkpoint-248/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-3/checkpoint-248/tokenizer_config.json ADDED
@@ -0,0 +1,945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "|||IP_ADDRESS|||",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1": {
12
+ "content": "<|padding|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "50254": {
20
+ "content": " ",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": false
26
+ },
27
+ "50255": {
28
+ "content": " ",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": false
34
+ },
35
+ "50256": {
36
+ "content": " ",
37
+ "lstrip": false,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": false
42
+ },
43
+ "50257": {
44
+ "content": " ",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ },
51
+ "50258": {
52
+ "content": " ",
53
+ "lstrip": false,
54
+ "normalized": true,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": false
58
+ },
59
+ "50259": {
60
+ "content": " ",
61
+ "lstrip": false,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": false
66
+ },
67
+ "50260": {
68
+ "content": " ",
69
+ "lstrip": false,
70
+ "normalized": true,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": false
74
+ },
75
+ "50261": {
76
+ "content": " ",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "50262": {
84
+ "content": " ",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "50263": {
92
+ "content": " ",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "50264": {
100
+ "content": " ",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "50265": {
108
+ "content": " ",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "50266": {
116
+ "content": " ",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "50267": {
124
+ "content": " ",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "50268": {
132
+ "content": " ",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "50269": {
140
+ "content": " ",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "50270": {
148
+ "content": " ",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "50271": {
156
+ "content": " ",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "50272": {
164
+ "content": " ",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "50273": {
172
+ "content": " ",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "50274": {
180
+ "content": " ",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "50275": {
188
+ "content": " ",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "50276": {
196
+ "content": " ",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "50277": {
204
+ "content": "|||EMAIL_ADDRESS|||",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "50278": {
212
+ "content": "|||PHONE_NUMBER|||",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "50279": {
220
+ "content": "<|endoftext|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "50280": {
228
+ "content": "[UNK]",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "50281": {
236
+ "content": "[CLS]",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "50282": {
244
+ "content": "[SEP]",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "50283": {
252
+ "content": "[PAD]",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "50284": {
260
+ "content": "[MASK]",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "50285": {
268
+ "content": "[unused0]",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "50286": {
276
+ "content": "[unused1]",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "50287": {
284
+ "content": "[unused2]",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "50288": {
292
+ "content": "[unused3]",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "50289": {
300
+ "content": "[unused4]",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "50290": {
308
+ "content": "[unused5]",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "50291": {
316
+ "content": "[unused6]",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "50292": {
324
+ "content": "[unused7]",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "50293": {
332
+ "content": "[unused8]",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "50294": {
340
+ "content": "[unused9]",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "50295": {
348
+ "content": "[unused10]",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "50296": {
356
+ "content": "[unused11]",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "50297": {
364
+ "content": "[unused12]",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "50298": {
372
+ "content": "[unused13]",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "50299": {
380
+ "content": "[unused14]",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "50300": {
388
+ "content": "[unused15]",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "50301": {
396
+ "content": "[unused16]",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "50302": {
404
+ "content": "[unused17]",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "50303": {
412
+ "content": "[unused18]",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "50304": {
420
+ "content": "[unused19]",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "50305": {
428
+ "content": "[unused20]",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "50306": {
436
+ "content": "[unused21]",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "50307": {
444
+ "content": "[unused22]",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "50308": {
452
+ "content": "[unused23]",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "50309": {
460
+ "content": "[unused24]",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "50310": {
468
+ "content": "[unused25]",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "50311": {
476
+ "content": "[unused26]",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "50312": {
484
+ "content": "[unused27]",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "50313": {
492
+ "content": "[unused28]",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "50314": {
500
+ "content": "[unused29]",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "50315": {
508
+ "content": "[unused30]",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "50316": {
516
+ "content": "[unused31]",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "50317": {
524
+ "content": "[unused32]",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "50318": {
532
+ "content": "[unused33]",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "50319": {
540
+ "content": "[unused34]",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "50320": {
548
+ "content": "[unused35]",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "50321": {
556
+ "content": "[unused36]",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "50322": {
564
+ "content": "[unused37]",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "50323": {
572
+ "content": "[unused38]",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "50324": {
580
+ "content": "[unused39]",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "50325": {
588
+ "content": "[unused40]",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "50326": {
596
+ "content": "[unused41]",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "50327": {
604
+ "content": "[unused42]",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "50328": {
612
+ "content": "[unused43]",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "50329": {
620
+ "content": "[unused44]",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "50330": {
628
+ "content": "[unused45]",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "50331": {
636
+ "content": "[unused46]",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "50332": {
644
+ "content": "[unused47]",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "50333": {
652
+ "content": "[unused48]",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "50334": {
660
+ "content": "[unused49]",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "50335": {
668
+ "content": "[unused50]",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "50336": {
676
+ "content": "[unused51]",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "50337": {
684
+ "content": "[unused52]",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "50338": {
692
+ "content": "[unused53]",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "50339": {
700
+ "content": "[unused54]",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "50340": {
708
+ "content": "[unused55]",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "50341": {
716
+ "content": "[unused56]",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "50342": {
724
+ "content": "[unused57]",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "50343": {
732
+ "content": "[unused58]",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "50344": {
740
+ "content": "[unused59]",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "50345": {
748
+ "content": "[unused60]",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "50346": {
756
+ "content": "[unused61]",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "50347": {
764
+ "content": "[unused62]",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "50348": {
772
+ "content": "[unused63]",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "50349": {
780
+ "content": "[unused64]",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "50350": {
788
+ "content": "[unused65]",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "50351": {
796
+ "content": "[unused66]",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "50352": {
804
+ "content": "[unused67]",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "50353": {
812
+ "content": "[unused68]",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "50354": {
820
+ "content": "[unused69]",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "50355": {
828
+ "content": "[unused70]",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "50356": {
836
+ "content": "[unused71]",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "50357": {
844
+ "content": "[unused72]",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "50358": {
852
+ "content": "[unused73]",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "50359": {
860
+ "content": "[unused74]",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "50360": {
868
+ "content": "[unused75]",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "50361": {
876
+ "content": "[unused76]",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "50362": {
884
+ "content": "[unused77]",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "50363": {
892
+ "content": "[unused78]",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "50364": {
900
+ "content": "[unused79]",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "50365": {
908
+ "content": "[unused80]",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "50366": {
916
+ "content": "[unused81]",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "50367": {
924
+ "content": "[unused82]",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ }
931
+ },
932
+ "clean_up_tokenization_spaces": true,
933
+ "cls_token": "[CLS]",
934
+ "extra_special_tokens": {},
935
+ "mask_token": "[MASK]",
936
+ "model_input_names": [
937
+ "input_ids",
938
+ "attention_mask"
939
+ ],
940
+ "model_max_length": 2048,
941
+ "pad_token": "[PAD]",
942
+ "sep_token": "[SEP]",
943
+ "tokenizer_class": "PreTrainedTokenizerFast",
944
+ "unk_token": "[UNK]"
945
+ }
run-3/checkpoint-248/trainer_state.json ADDED
@@ -0,0 +1,1797 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.8645277672129885,
3
+ "best_model_checkpoint": "./modernBERT-content-regression/run-3/checkpoint-248",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 248,
7
+ "is_hyper_param_search": true,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.008064516129032258,
13
+ "grad_norm": 312.8093566894531,
14
+ "learning_rate": 1.1731316602713056e-06,
15
+ "loss": 21.3087,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.016129032258064516,
20
+ "grad_norm": 97.11482238769531,
21
+ "learning_rate": 1.1712364556505119e-06,
22
+ "loss": 1.9024,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.024193548387096774,
27
+ "grad_norm": 608.1284790039062,
28
+ "learning_rate": 1.1693412510297182e-06,
29
+ "loss": 154.2055,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.03225806451612903,
34
+ "grad_norm": 205.3589630126953,
35
+ "learning_rate": 1.1674460464089244e-06,
36
+ "loss": 18.409,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.04032258064516129,
41
+ "grad_norm": 116.6544189453125,
42
+ "learning_rate": 1.165550841788131e-06,
43
+ "loss": 4.2643,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.04838709677419355,
48
+ "grad_norm": 34.426883697509766,
49
+ "learning_rate": 1.1636556371673372e-06,
50
+ "loss": 0.5243,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.056451612903225805,
55
+ "grad_norm": 161.2743682861328,
56
+ "learning_rate": 1.1617604325465434e-06,
57
+ "loss": 9.653,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.06451612903225806,
62
+ "grad_norm": 316.244384765625,
63
+ "learning_rate": 1.1598652279257497e-06,
64
+ "loss": 21.3918,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.07258064516129033,
69
+ "grad_norm": 64.28992462158203,
70
+ "learning_rate": 1.157970023304956e-06,
71
+ "loss": 0.6204,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.08064516129032258,
76
+ "grad_norm": 130.6162109375,
77
+ "learning_rate": 1.1560748186841622e-06,
78
+ "loss": 5.0316,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.08870967741935484,
83
+ "grad_norm": 340.1315612792969,
84
+ "learning_rate": 1.1541796140633685e-06,
85
+ "loss": 18.7971,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.0967741935483871,
90
+ "grad_norm": 63.929317474365234,
91
+ "learning_rate": 1.1522844094425748e-06,
92
+ "loss": 1.2454,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.10483870967741936,
97
+ "grad_norm": 5.4183030128479,
98
+ "learning_rate": 1.150389204821781e-06,
99
+ "loss": 0.0884,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.11290322580645161,
104
+ "grad_norm": 137.7165985107422,
105
+ "learning_rate": 1.1484940002009875e-06,
106
+ "loss": 4.2256,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.12096774193548387,
111
+ "grad_norm": 31.921754837036133,
112
+ "learning_rate": 1.1465987955801938e-06,
113
+ "loss": 0.3319,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.12903225806451613,
118
+ "grad_norm": 200.37228393554688,
119
+ "learning_rate": 1.1447035909594e-06,
120
+ "loss": 11.9662,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.13709677419354838,
125
+ "grad_norm": 54.43822479248047,
126
+ "learning_rate": 1.1428083863386063e-06,
127
+ "loss": 1.2817,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.14516129032258066,
132
+ "grad_norm": 122.57898712158203,
133
+ "learning_rate": 1.1409131817178126e-06,
134
+ "loss": 3.3717,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.1532258064516129,
139
+ "grad_norm": 24.417898178100586,
140
+ "learning_rate": 1.1390179770970188e-06,
141
+ "loss": 0.5961,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.16129032258064516,
146
+ "grad_norm": 280.314453125,
147
+ "learning_rate": 1.1371227724762251e-06,
148
+ "loss": 12.7171,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.1693548387096774,
153
+ "grad_norm": 74.58489227294922,
154
+ "learning_rate": 1.1352275678554316e-06,
155
+ "loss": 0.6804,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.1774193548387097,
160
+ "grad_norm": 134.18431091308594,
161
+ "learning_rate": 1.1333323632346376e-06,
162
+ "loss": 1.4807,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.18548387096774194,
167
+ "grad_norm": 127.26359558105469,
168
+ "learning_rate": 1.1314371586138441e-06,
169
+ "loss": 18.1665,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.1935483870967742,
174
+ "grad_norm": 196.29579162597656,
175
+ "learning_rate": 1.1295419539930504e-06,
176
+ "loss": 2.6718,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.20161290322580644,
181
+ "grad_norm": 59.49917984008789,
182
+ "learning_rate": 1.1276467493722567e-06,
183
+ "loss": 0.7888,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.20967741935483872,
188
+ "grad_norm": 160.57215881347656,
189
+ "learning_rate": 1.125751544751463e-06,
190
+ "loss": 0.7413,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.21774193548387097,
195
+ "grad_norm": 63.3122444152832,
196
+ "learning_rate": 1.1238563401306692e-06,
197
+ "loss": 3.04,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.22580645161290322,
202
+ "grad_norm": 132.4581756591797,
203
+ "learning_rate": 1.1219611355098755e-06,
204
+ "loss": 1.7609,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.23387096774193547,
209
+ "grad_norm": 69.83406829833984,
210
+ "learning_rate": 1.1200659308890817e-06,
211
+ "loss": 1.1596,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.24193548387096775,
216
+ "grad_norm": 34.646114349365234,
217
+ "learning_rate": 1.1181707262682882e-06,
218
+ "loss": 0.5233,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.25,
223
+ "grad_norm": 28.260757446289062,
224
+ "learning_rate": 1.1162755216474943e-06,
225
+ "loss": 1.0132,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.25806451612903225,
230
+ "grad_norm": 44.67536544799805,
231
+ "learning_rate": 1.1143803170267007e-06,
232
+ "loss": 0.9895,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.2661290322580645,
237
+ "grad_norm": 54.21400451660156,
238
+ "learning_rate": 1.112485112405907e-06,
239
+ "loss": 1.1101,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.27419354838709675,
244
+ "grad_norm": 36.84498977661133,
245
+ "learning_rate": 1.1105899077851133e-06,
246
+ "loss": 1.3959,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.28225806451612906,
251
+ "grad_norm": 369.2140808105469,
252
+ "learning_rate": 1.1086947031643195e-06,
253
+ "loss": 23.5792,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.2903225806451613,
258
+ "grad_norm": 51.993858337402344,
259
+ "learning_rate": 1.1067994985435258e-06,
260
+ "loss": 0.6532,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.29838709677419356,
265
+ "grad_norm": 87.89725494384766,
266
+ "learning_rate": 1.104904293922732e-06,
267
+ "loss": 1.0406,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.3064516129032258,
272
+ "grad_norm": 272.9493713378906,
273
+ "learning_rate": 1.1030090893019383e-06,
274
+ "loss": 24.8131,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.31451612903225806,
279
+ "grad_norm": 96.62998962402344,
280
+ "learning_rate": 1.1011138846811448e-06,
281
+ "loss": 0.7572,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.3225806451612903,
286
+ "grad_norm": 538.8275756835938,
287
+ "learning_rate": 1.0992186800603509e-06,
288
+ "loss": 30.8125,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.33064516129032256,
293
+ "grad_norm": 285.717041015625,
294
+ "learning_rate": 1.0973234754395574e-06,
295
+ "loss": 1.837,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.3387096774193548,
300
+ "grad_norm": 215.8197021484375,
301
+ "learning_rate": 1.0954282708187636e-06,
302
+ "loss": 18.2501,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.3467741935483871,
307
+ "grad_norm": 163.01434326171875,
308
+ "learning_rate": 1.0935330661979699e-06,
309
+ "loss": 27.9919,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.3548387096774194,
314
+ "grad_norm": 24.41200828552246,
315
+ "learning_rate": 1.0916378615771762e-06,
316
+ "loss": 0.3323,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.3629032258064516,
321
+ "grad_norm": 39.837669372558594,
322
+ "learning_rate": 1.0897426569563824e-06,
323
+ "loss": 1.1347,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.3709677419354839,
328
+ "grad_norm": 47.366092681884766,
329
+ "learning_rate": 1.0878474523355887e-06,
330
+ "loss": 2.3667,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.3790322580645161,
335
+ "grad_norm": 85.20360565185547,
336
+ "learning_rate": 1.085952247714795e-06,
337
+ "loss": 2.8234,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.3870967741935484,
342
+ "grad_norm": 209.84487915039062,
343
+ "learning_rate": 1.0840570430940014e-06,
344
+ "loss": 2.7326,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.3951612903225806,
349
+ "grad_norm": 61.69172286987305,
350
+ "learning_rate": 1.0821618384732077e-06,
351
+ "loss": 0.9325,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.4032258064516129,
356
+ "grad_norm": 149.46607971191406,
357
+ "learning_rate": 1.080266633852414e-06,
358
+ "loss": 7.6034,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.4112903225806452,
363
+ "grad_norm": 163.2115478515625,
364
+ "learning_rate": 1.0783714292316202e-06,
365
+ "loss": 1.8005,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.41935483870967744,
370
+ "grad_norm": 42.81549072265625,
371
+ "learning_rate": 1.0764762246108265e-06,
372
+ "loss": 0.3282,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.4274193548387097,
377
+ "grad_norm": 516.490966796875,
378
+ "learning_rate": 1.0745810199900328e-06,
379
+ "loss": 13.6021,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.43548387096774194,
384
+ "grad_norm": 83.2822265625,
385
+ "learning_rate": 1.072685815369239e-06,
386
+ "loss": 1.2216,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.4435483870967742,
391
+ "grad_norm": 86.0993881225586,
392
+ "learning_rate": 1.0707906107484453e-06,
393
+ "loss": 4.7606,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.45161290322580644,
398
+ "grad_norm": 44.0859375,
399
+ "learning_rate": 1.0688954061276516e-06,
400
+ "loss": 0.8722,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.4596774193548387,
405
+ "grad_norm": 124.6083984375,
406
+ "learning_rate": 1.067000201506858e-06,
407
+ "loss": 1.3964,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.46774193548387094,
412
+ "grad_norm": 206.24273681640625,
413
+ "learning_rate": 1.0651049968860643e-06,
414
+ "loss": 7.95,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.47580645161290325,
419
+ "grad_norm": 75.12397766113281,
420
+ "learning_rate": 1.0632097922652704e-06,
421
+ "loss": 0.7911,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.4838709677419355,
426
+ "grad_norm": 234.4656524658203,
427
+ "learning_rate": 1.0613145876444769e-06,
428
+ "loss": 14.9343,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.49193548387096775,
433
+ "grad_norm": 244.2675323486328,
434
+ "learning_rate": 1.0594193830236831e-06,
435
+ "loss": 0.5364,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.5,
440
+ "grad_norm": 134.83209228515625,
441
+ "learning_rate": 1.0575241784028894e-06,
442
+ "loss": 23.3423,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.5080645161290323,
447
+ "grad_norm": 117.73362731933594,
448
+ "learning_rate": 1.0556289737820957e-06,
449
+ "loss": 1.3623,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.5161290322580645,
454
+ "grad_norm": 67.53313446044922,
455
+ "learning_rate": 1.053733769161302e-06,
456
+ "loss": 2.8317,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.5241935483870968,
461
+ "grad_norm": 169.532470703125,
462
+ "learning_rate": 1.0518385645405082e-06,
463
+ "loss": 14.4335,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.532258064516129,
468
+ "grad_norm": 62.2873649597168,
469
+ "learning_rate": 1.0499433599197147e-06,
470
+ "loss": 1.3316,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.5403225806451613,
475
+ "grad_norm": 70.10298156738281,
476
+ "learning_rate": 1.048048155298921e-06,
477
+ "loss": 1.325,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.5483870967741935,
482
+ "grad_norm": 60.70389175415039,
483
+ "learning_rate": 1.0461529506781272e-06,
484
+ "loss": 0.8161,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.5564516129032258,
489
+ "grad_norm": 104.86827087402344,
490
+ "learning_rate": 1.0442577460573335e-06,
491
+ "loss": 1.3411,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.5645161290322581,
496
+ "grad_norm": 78.47552490234375,
497
+ "learning_rate": 1.0423625414365397e-06,
498
+ "loss": 0.7227,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.5725806451612904,
503
+ "grad_norm": 74.83175659179688,
504
+ "learning_rate": 1.040467336815746e-06,
505
+ "loss": 2.0944,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.5806451612903226,
510
+ "grad_norm": 261.87298583984375,
511
+ "learning_rate": 1.0385721321949523e-06,
512
+ "loss": 17.7856,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.5887096774193549,
517
+ "grad_norm": 55.27876663208008,
518
+ "learning_rate": 1.0366769275741585e-06,
519
+ "loss": 0.98,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.5967741935483871,
524
+ "grad_norm": 145.73300170898438,
525
+ "learning_rate": 1.0347817229533648e-06,
526
+ "loss": 1.7042,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.6048387096774194,
531
+ "grad_norm": 125.30365753173828,
532
+ "learning_rate": 1.0328865183325713e-06,
533
+ "loss": 16.882,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.6129032258064516,
538
+ "grad_norm": 26.542724609375,
539
+ "learning_rate": 1.0309913137117776e-06,
540
+ "loss": 0.1877,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.6209677419354839,
545
+ "grad_norm": 37.75946044921875,
546
+ "learning_rate": 1.0290961090909838e-06,
547
+ "loss": 0.2059,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.6290322580645161,
552
+ "grad_norm": 128.78330993652344,
553
+ "learning_rate": 1.02720090447019e-06,
554
+ "loss": 13.4677,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.6370967741935484,
559
+ "grad_norm": 210.26707458496094,
560
+ "learning_rate": 1.0253056998493964e-06,
561
+ "loss": 22.7431,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.6451612903225806,
566
+ "grad_norm": 35.724334716796875,
567
+ "learning_rate": 1.0234104952286026e-06,
568
+ "loss": 1.0135,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.6532258064516129,
573
+ "grad_norm": 66.4817886352539,
574
+ "learning_rate": 1.0215152906078089e-06,
575
+ "loss": 0.4585,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.6612903225806451,
580
+ "grad_norm": 53.807220458984375,
581
+ "learning_rate": 1.0196200859870154e-06,
582
+ "loss": 2.1388,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.6693548387096774,
587
+ "grad_norm": 33.74148178100586,
588
+ "learning_rate": 1.0177248813662214e-06,
589
+ "loss": 0.2365,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.6774193548387096,
594
+ "grad_norm": 47.4416618347168,
595
+ "learning_rate": 1.015829676745428e-06,
596
+ "loss": 1.1393,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.6854838709677419,
601
+ "grad_norm": 280.392578125,
602
+ "learning_rate": 1.0139344721246342e-06,
603
+ "loss": 18.5951,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.6935483870967742,
608
+ "grad_norm": 91.39240264892578,
609
+ "learning_rate": 1.0120392675038404e-06,
610
+ "loss": 14.467,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.7016129032258065,
615
+ "grad_norm": 97.39054107666016,
616
+ "learning_rate": 1.0101440628830467e-06,
617
+ "loss": 2.257,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.7096774193548387,
622
+ "grad_norm": 95.57569885253906,
623
+ "learning_rate": 1.008248858262253e-06,
624
+ "loss": 0.9064,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.717741935483871,
629
+ "grad_norm": 230.4451904296875,
630
+ "learning_rate": 1.0063536536414592e-06,
631
+ "loss": 1.5809,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.7258064516129032,
636
+ "grad_norm": 149.36996459960938,
637
+ "learning_rate": 1.0044584490206655e-06,
638
+ "loss": 1.7855,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.7338709677419355,
643
+ "grad_norm": 319.1463317871094,
644
+ "learning_rate": 1.002563244399872e-06,
645
+ "loss": 19.9728,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.7419354838709677,
650
+ "grad_norm": 132.4302215576172,
651
+ "learning_rate": 1.000668039779078e-06,
652
+ "loss": 7.4546,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.75,
657
+ "grad_norm": 24.757253646850586,
658
+ "learning_rate": 9.987728351582845e-07,
659
+ "loss": 0.1834,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.7580645161290323,
664
+ "grad_norm": 117.05673217773438,
665
+ "learning_rate": 9.968776305374908e-07,
666
+ "loss": 1.7553,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.7661290322580645,
671
+ "grad_norm": 178.41818237304688,
672
+ "learning_rate": 9.94982425916697e-07,
673
+ "loss": 9.5775,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.7741935483870968,
678
+ "grad_norm": 129.0936737060547,
679
+ "learning_rate": 9.930872212959033e-07,
680
+ "loss": 2.2522,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.782258064516129,
685
+ "grad_norm": 120.8997802734375,
686
+ "learning_rate": 9.911920166751096e-07,
687
+ "loss": 1.365,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.7903225806451613,
692
+ "grad_norm": 124.6778335571289,
693
+ "learning_rate": 9.892968120543159e-07,
694
+ "loss": 1.6905,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.7983870967741935,
699
+ "grad_norm": 805.3359375,
700
+ "learning_rate": 9.874016074335221e-07,
701
+ "loss": 2.0304,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.8064516129032258,
706
+ "grad_norm": 258.4070739746094,
707
+ "learning_rate": 9.855064028127286e-07,
708
+ "loss": 18.573,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.8145161290322581,
713
+ "grad_norm": 25.10341453552246,
714
+ "learning_rate": 9.836111981919347e-07,
715
+ "loss": 0.8904,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.8225806451612904,
720
+ "grad_norm": 137.26222229003906,
721
+ "learning_rate": 9.817159935711411e-07,
722
+ "loss": 1.3759,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.8306451612903226,
727
+ "grad_norm": 100.84027862548828,
728
+ "learning_rate": 9.798207889503474e-07,
729
+ "loss": 11.2571,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.8387096774193549,
734
+ "grad_norm": 99.84913635253906,
735
+ "learning_rate": 9.779255843295537e-07,
736
+ "loss": 0.7655,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.8467741935483871,
741
+ "grad_norm": 58.0277099609375,
742
+ "learning_rate": 9.7603037970876e-07,
743
+ "loss": 0.685,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.8548387096774194,
748
+ "grad_norm": 118.5246810913086,
749
+ "learning_rate": 9.741351750879662e-07,
750
+ "loss": 3.1796,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.8629032258064516,
755
+ "grad_norm": 93.28719329833984,
756
+ "learning_rate": 9.722399704671725e-07,
757
+ "loss": 1.5221,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.8709677419354839,
762
+ "grad_norm": 109.0409927368164,
763
+ "learning_rate": 9.703447658463787e-07,
764
+ "loss": 1.2383,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.8790322580645161,
769
+ "grad_norm": 167.7772216796875,
770
+ "learning_rate": 9.684495612255852e-07,
771
+ "loss": 2.5045,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.8870967741935484,
776
+ "grad_norm": 51.926692962646484,
777
+ "learning_rate": 9.665543566047915e-07,
778
+ "loss": 1.0717,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.8951612903225806,
783
+ "grad_norm": 222.6824951171875,
784
+ "learning_rate": 9.646591519839975e-07,
785
+ "loss": 4.1007,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.9032258064516129,
790
+ "grad_norm": 53.584320068359375,
791
+ "learning_rate": 9.62763947363204e-07,
792
+ "loss": 1.257,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.9112903225806451,
797
+ "grad_norm": 92.6733627319336,
798
+ "learning_rate": 9.608687427424103e-07,
799
+ "loss": 0.7563,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.9193548387096774,
804
+ "grad_norm": 23.77389144897461,
805
+ "learning_rate": 9.589735381216166e-07,
806
+ "loss": 0.7985,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.9274193548387096,
811
+ "grad_norm": 374.8204650878906,
812
+ "learning_rate": 9.570783335008228e-07,
813
+ "loss": 5.5011,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.9354838709677419,
818
+ "grad_norm": 83.84817504882812,
819
+ "learning_rate": 9.55183128880029e-07,
820
+ "loss": 2.123,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.9435483870967742,
825
+ "grad_norm": 160.6682891845703,
826
+ "learning_rate": 9.532879242592355e-07,
827
+ "loss": 1.6017,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.9516129032258065,
832
+ "grad_norm": 150.16294860839844,
833
+ "learning_rate": 9.513927196384417e-07,
834
+ "loss": 14.0088,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.9596774193548387,
839
+ "grad_norm": 94.02184295654297,
840
+ "learning_rate": 9.494975150176481e-07,
841
+ "loss": 1.6218,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.967741935483871,
846
+ "grad_norm": 40.898277282714844,
847
+ "learning_rate": 9.476023103968543e-07,
848
+ "loss": 1.4986,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.9758064516129032,
853
+ "grad_norm": 167.00746154785156,
854
+ "learning_rate": 9.457071057760605e-07,
855
+ "loss": 1.4057,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.9838709677419355,
860
+ "grad_norm": 35.20161819458008,
861
+ "learning_rate": 9.438119011552669e-07,
862
+ "loss": 0.5185,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.9919354838709677,
867
+ "grad_norm": 61.312583923339844,
868
+ "learning_rate": 9.419166965344732e-07,
869
+ "loss": 0.6058,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 1.0,
874
+ "grad_norm": 9.15098762512207,
875
+ "learning_rate": 9.400214919136795e-07,
876
+ "loss": 0.008,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 1.0,
881
+ "eval_loss": 3.5903782844543457,
882
+ "eval_mae": 1.1498976945877075,
883
+ "eval_mse": 3.590378522872925,
884
+ "eval_r2": 0.01701676845550537,
885
+ "eval_rmse": 1.8948294178824976,
886
+ "eval_runtime": 1.3414,
887
+ "eval_samples_per_second": 41.002,
888
+ "eval_smape": 46.445611119270325,
889
+ "eval_steps_per_second": 10.437,
890
+ "step": 124
891
+ },
892
+ {
893
+ "epoch": 1.0080645161290323,
894
+ "grad_norm": 93.99532318115234,
895
+ "learning_rate": 9.381262872928857e-07,
896
+ "loss": 1.3843,
897
+ "step": 125
898
+ },
899
+ {
900
+ "epoch": 1.0161290322580645,
901
+ "grad_norm": 131.81895446777344,
902
+ "learning_rate": 9.362310826720921e-07,
903
+ "loss": 0.7999,
904
+ "step": 126
905
+ },
906
+ {
907
+ "epoch": 1.0241935483870968,
908
+ "grad_norm": 137.10763549804688,
909
+ "learning_rate": 9.343358780512983e-07,
910
+ "loss": 6.5196,
911
+ "step": 127
912
+ },
913
+ {
914
+ "epoch": 1.032258064516129,
915
+ "grad_norm": 48.18350601196289,
916
+ "learning_rate": 9.324406734305047e-07,
917
+ "loss": 0.6827,
918
+ "step": 128
919
+ },
920
+ {
921
+ "epoch": 1.0403225806451613,
922
+ "grad_norm": 177.0278778076172,
923
+ "learning_rate": 9.30545468809711e-07,
924
+ "loss": 0.8596,
925
+ "step": 129
926
+ },
927
+ {
928
+ "epoch": 1.0483870967741935,
929
+ "grad_norm": 47.4871711730957,
930
+ "learning_rate": 9.286502641889171e-07,
931
+ "loss": 2.1029,
932
+ "step": 130
933
+ },
934
+ {
935
+ "epoch": 1.0564516129032258,
936
+ "grad_norm": 47.476219177246094,
937
+ "learning_rate": 9.267550595681235e-07,
938
+ "loss": 0.2775,
939
+ "step": 131
940
+ },
941
+ {
942
+ "epoch": 1.064516129032258,
943
+ "grad_norm": 171.87510681152344,
944
+ "learning_rate": 9.248598549473298e-07,
945
+ "loss": 5.2664,
946
+ "step": 132
947
+ },
948
+ {
949
+ "epoch": 1.0725806451612903,
950
+ "grad_norm": 76.38048553466797,
951
+ "learning_rate": 9.229646503265362e-07,
952
+ "loss": 0.9048,
953
+ "step": 133
954
+ },
955
+ {
956
+ "epoch": 1.0806451612903225,
957
+ "grad_norm": 65.83001708984375,
958
+ "learning_rate": 9.210694457057423e-07,
959
+ "loss": 0.6209,
960
+ "step": 134
961
+ },
962
+ {
963
+ "epoch": 1.0887096774193548,
964
+ "grad_norm": 155.04318237304688,
965
+ "learning_rate": 9.191742410849487e-07,
966
+ "loss": 7.5308,
967
+ "step": 135
968
+ },
969
+ {
970
+ "epoch": 1.096774193548387,
971
+ "grad_norm": 180.7427520751953,
972
+ "learning_rate": 9.17279036464155e-07,
973
+ "loss": 16.1446,
974
+ "step": 136
975
+ },
976
+ {
977
+ "epoch": 1.1048387096774193,
978
+ "grad_norm": 209.10556030273438,
979
+ "learning_rate": 9.153838318433613e-07,
980
+ "loss": 10.6327,
981
+ "step": 137
982
+ },
983
+ {
984
+ "epoch": 1.1129032258064515,
985
+ "grad_norm": 31.048919677734375,
986
+ "learning_rate": 9.134886272225676e-07,
987
+ "loss": 0.4128,
988
+ "step": 138
989
+ },
990
+ {
991
+ "epoch": 1.120967741935484,
992
+ "grad_norm": 419.14520263671875,
993
+ "learning_rate": 9.115934226017738e-07,
994
+ "loss": 45.5376,
995
+ "step": 139
996
+ },
997
+ {
998
+ "epoch": 1.129032258064516,
999
+ "grad_norm": 18.564205169677734,
1000
+ "learning_rate": 9.096982179809801e-07,
1001
+ "loss": 0.1034,
1002
+ "step": 140
1003
+ },
1004
+ {
1005
+ "epoch": 1.1370967741935485,
1006
+ "grad_norm": 38.12897872924805,
1007
+ "learning_rate": 9.078030133601864e-07,
1008
+ "loss": 1.2526,
1009
+ "step": 141
1010
+ },
1011
+ {
1012
+ "epoch": 1.1451612903225807,
1013
+ "grad_norm": 56.46409606933594,
1014
+ "learning_rate": 9.059078087393928e-07,
1015
+ "loss": 1.3236,
1016
+ "step": 142
1017
+ },
1018
+ {
1019
+ "epoch": 1.153225806451613,
1020
+ "grad_norm": 57.76015090942383,
1021
+ "learning_rate": 9.04012604118599e-07,
1022
+ "loss": 0.3464,
1023
+ "step": 143
1024
+ },
1025
+ {
1026
+ "epoch": 1.1612903225806452,
1027
+ "grad_norm": 32.81700134277344,
1028
+ "learning_rate": 9.021173994978053e-07,
1029
+ "loss": 0.2656,
1030
+ "step": 144
1031
+ },
1032
+ {
1033
+ "epoch": 1.1693548387096775,
1034
+ "grad_norm": 40.10895919799805,
1035
+ "learning_rate": 9.002221948770116e-07,
1036
+ "loss": 0.8185,
1037
+ "step": 145
1038
+ },
1039
+ {
1040
+ "epoch": 1.1774193548387097,
1041
+ "grad_norm": 46.941200256347656,
1042
+ "learning_rate": 8.98326990256218e-07,
1043
+ "loss": 2.0041,
1044
+ "step": 146
1045
+ },
1046
+ {
1047
+ "epoch": 1.185483870967742,
1048
+ "grad_norm": 47.112571716308594,
1049
+ "learning_rate": 8.964317856354242e-07,
1050
+ "loss": 0.4361,
1051
+ "step": 147
1052
+ },
1053
+ {
1054
+ "epoch": 1.1935483870967742,
1055
+ "grad_norm": 91.99717712402344,
1056
+ "learning_rate": 8.945365810146304e-07,
1057
+ "loss": 1.2795,
1058
+ "step": 148
1059
+ },
1060
+ {
1061
+ "epoch": 1.2016129032258065,
1062
+ "grad_norm": 321.48345947265625,
1063
+ "learning_rate": 8.926413763938367e-07,
1064
+ "loss": 21.7397,
1065
+ "step": 149
1066
+ },
1067
+ {
1068
+ "epoch": 1.2096774193548387,
1069
+ "grad_norm": 420.6986083984375,
1070
+ "learning_rate": 8.90746171773043e-07,
1071
+ "loss": 37.258,
1072
+ "step": 150
1073
+ },
1074
+ {
1075
+ "epoch": 1.217741935483871,
1076
+ "grad_norm": 195.8186492919922,
1077
+ "learning_rate": 8.888509671522494e-07,
1078
+ "loss": 17.6255,
1079
+ "step": 151
1080
+ },
1081
+ {
1082
+ "epoch": 1.2258064516129032,
1083
+ "grad_norm": 131.71688842773438,
1084
+ "learning_rate": 8.869557625314557e-07,
1085
+ "loss": 1.9709,
1086
+ "step": 152
1087
+ },
1088
+ {
1089
+ "epoch": 1.2338709677419355,
1090
+ "grad_norm": 275.066162109375,
1091
+ "learning_rate": 8.850605579106619e-07,
1092
+ "loss": 10.5323,
1093
+ "step": 153
1094
+ },
1095
+ {
1096
+ "epoch": 1.2419354838709677,
1097
+ "grad_norm": 121.62479400634766,
1098
+ "learning_rate": 8.831653532898682e-07,
1099
+ "loss": 13.0309,
1100
+ "step": 154
1101
+ },
1102
+ {
1103
+ "epoch": 1.25,
1104
+ "grad_norm": 44.131412506103516,
1105
+ "learning_rate": 8.812701486690746e-07,
1106
+ "loss": 0.4236,
1107
+ "step": 155
1108
+ },
1109
+ {
1110
+ "epoch": 1.2580645161290323,
1111
+ "grad_norm": 48.05416488647461,
1112
+ "learning_rate": 8.793749440482808e-07,
1113
+ "loss": 0.787,
1114
+ "step": 156
1115
+ },
1116
+ {
1117
+ "epoch": 1.2661290322580645,
1118
+ "grad_norm": 136.1278839111328,
1119
+ "learning_rate": 8.774797394274871e-07,
1120
+ "loss": 4.653,
1121
+ "step": 157
1122
+ },
1123
+ {
1124
+ "epoch": 1.2741935483870968,
1125
+ "grad_norm": 156.81910705566406,
1126
+ "learning_rate": 8.755845348066934e-07,
1127
+ "loss": 14.6158,
1128
+ "step": 158
1129
+ },
1130
+ {
1131
+ "epoch": 1.282258064516129,
1132
+ "grad_norm": 311.3755798339844,
1133
+ "learning_rate": 8.736893301858996e-07,
1134
+ "loss": 25.6107,
1135
+ "step": 159
1136
+ },
1137
+ {
1138
+ "epoch": 1.2903225806451613,
1139
+ "grad_norm": 147.80181884765625,
1140
+ "learning_rate": 8.71794125565106e-07,
1141
+ "loss": 10.7386,
1142
+ "step": 160
1143
+ },
1144
+ {
1145
+ "epoch": 1.2983870967741935,
1146
+ "grad_norm": 326.3906555175781,
1147
+ "learning_rate": 8.698989209443123e-07,
1148
+ "loss": 41.9267,
1149
+ "step": 161
1150
+ },
1151
+ {
1152
+ "epoch": 1.3064516129032258,
1153
+ "grad_norm": 35.69755172729492,
1154
+ "learning_rate": 8.680037163235186e-07,
1155
+ "loss": 0.9466,
1156
+ "step": 162
1157
+ },
1158
+ {
1159
+ "epoch": 1.314516129032258,
1160
+ "grad_norm": 82.98596954345703,
1161
+ "learning_rate": 8.661085117027248e-07,
1162
+ "loss": 0.6258,
1163
+ "step": 163
1164
+ },
1165
+ {
1166
+ "epoch": 1.3225806451612903,
1167
+ "grad_norm": 105.8860855102539,
1168
+ "learning_rate": 8.642133070819311e-07,
1169
+ "loss": 2.386,
1170
+ "step": 164
1171
+ },
1172
+ {
1173
+ "epoch": 1.3306451612903225,
1174
+ "grad_norm": 196.2127685546875,
1175
+ "learning_rate": 8.623181024611374e-07,
1176
+ "loss": 2.511,
1177
+ "step": 165
1178
+ },
1179
+ {
1180
+ "epoch": 1.3387096774193548,
1181
+ "grad_norm": 145.88681030273438,
1182
+ "learning_rate": 8.604228978403437e-07,
1183
+ "loss": 1.8581,
1184
+ "step": 166
1185
+ },
1186
+ {
1187
+ "epoch": 1.346774193548387,
1188
+ "grad_norm": 151.8137664794922,
1189
+ "learning_rate": 8.5852769321955e-07,
1190
+ "loss": 3.5077,
1191
+ "step": 167
1192
+ },
1193
+ {
1194
+ "epoch": 1.3548387096774195,
1195
+ "grad_norm": 83.18711853027344,
1196
+ "learning_rate": 8.566324885987562e-07,
1197
+ "loss": 0.5735,
1198
+ "step": 168
1199
+ },
1200
+ {
1201
+ "epoch": 1.3629032258064515,
1202
+ "grad_norm": 137.17172241210938,
1203
+ "learning_rate": 8.547372839779626e-07,
1204
+ "loss": 2.0364,
1205
+ "step": 169
1206
+ },
1207
+ {
1208
+ "epoch": 1.370967741935484,
1209
+ "grad_norm": 486.7686767578125,
1210
+ "learning_rate": 8.528420793571689e-07,
1211
+ "loss": 130.1139,
1212
+ "step": 170
1213
+ },
1214
+ {
1215
+ "epoch": 1.379032258064516,
1216
+ "grad_norm": 53.59453582763672,
1217
+ "learning_rate": 8.509468747363753e-07,
1218
+ "loss": 1.0157,
1219
+ "step": 171
1220
+ },
1221
+ {
1222
+ "epoch": 1.3870967741935485,
1223
+ "grad_norm": 361.83270263671875,
1224
+ "learning_rate": 8.490516701155814e-07,
1225
+ "loss": 9.2801,
1226
+ "step": 172
1227
+ },
1228
+ {
1229
+ "epoch": 1.3951612903225805,
1230
+ "grad_norm": 122.44477081298828,
1231
+ "learning_rate": 8.471564654947877e-07,
1232
+ "loss": 1.7689,
1233
+ "step": 173
1234
+ },
1235
+ {
1236
+ "epoch": 1.403225806451613,
1237
+ "grad_norm": 163.0518035888672,
1238
+ "learning_rate": 8.452612608739941e-07,
1239
+ "loss": 8.8246,
1240
+ "step": 174
1241
+ },
1242
+ {
1243
+ "epoch": 1.4112903225806452,
1244
+ "grad_norm": 23.73016357421875,
1245
+ "learning_rate": 8.433660562532003e-07,
1246
+ "loss": 0.3707,
1247
+ "step": 175
1248
+ },
1249
+ {
1250
+ "epoch": 1.4193548387096775,
1251
+ "grad_norm": 392.39739990234375,
1252
+ "learning_rate": 8.414708516324067e-07,
1253
+ "loss": 19.7684,
1254
+ "step": 176
1255
+ },
1256
+ {
1257
+ "epoch": 1.4274193548387097,
1258
+ "grad_norm": 101.17041015625,
1259
+ "learning_rate": 8.395756470116129e-07,
1260
+ "loss": 1.2123,
1261
+ "step": 177
1262
+ },
1263
+ {
1264
+ "epoch": 1.435483870967742,
1265
+ "grad_norm": 129.82452392578125,
1266
+ "learning_rate": 8.376804423908192e-07,
1267
+ "loss": 1.3362,
1268
+ "step": 178
1269
+ },
1270
+ {
1271
+ "epoch": 1.4435483870967742,
1272
+ "grad_norm": 93.25836944580078,
1273
+ "learning_rate": 8.357852377700255e-07,
1274
+ "loss": 2.67,
1275
+ "step": 179
1276
+ },
1277
+ {
1278
+ "epoch": 1.4516129032258065,
1279
+ "grad_norm": 144.9728240966797,
1280
+ "learning_rate": 8.338900331492319e-07,
1281
+ "loss": 1.4356,
1282
+ "step": 180
1283
+ },
1284
+ {
1285
+ "epoch": 1.4596774193548387,
1286
+ "grad_norm": 106.55712127685547,
1287
+ "learning_rate": 8.31994828528438e-07,
1288
+ "loss": 1.0411,
1289
+ "step": 181
1290
+ },
1291
+ {
1292
+ "epoch": 1.467741935483871,
1293
+ "grad_norm": 50.07807922363281,
1294
+ "learning_rate": 8.300996239076443e-07,
1295
+ "loss": 1.0384,
1296
+ "step": 182
1297
+ },
1298
+ {
1299
+ "epoch": 1.4758064516129032,
1300
+ "grad_norm": 47.749080657958984,
1301
+ "learning_rate": 8.282044192868507e-07,
1302
+ "loss": 0.8744,
1303
+ "step": 183
1304
+ },
1305
+ {
1306
+ "epoch": 1.4838709677419355,
1307
+ "grad_norm": 505.6190185546875,
1308
+ "learning_rate": 8.263092146660569e-07,
1309
+ "loss": 3.8061,
1310
+ "step": 184
1311
+ },
1312
+ {
1313
+ "epoch": 1.4919354838709677,
1314
+ "grad_norm": 126.373779296875,
1315
+ "learning_rate": 8.244140100452633e-07,
1316
+ "loss": 1.3906,
1317
+ "step": 185
1318
+ },
1319
+ {
1320
+ "epoch": 1.5,
1321
+ "grad_norm": 95.14514923095703,
1322
+ "learning_rate": 8.225188054244695e-07,
1323
+ "loss": 0.7666,
1324
+ "step": 186
1325
+ },
1326
+ {
1327
+ "epoch": 1.5080645161290323,
1328
+ "grad_norm": 171.452392578125,
1329
+ "learning_rate": 8.206236008036759e-07,
1330
+ "loss": 4.4224,
1331
+ "step": 187
1332
+ },
1333
+ {
1334
+ "epoch": 1.5161290322580645,
1335
+ "grad_norm": 58.15391159057617,
1336
+ "learning_rate": 8.187283961828821e-07,
1337
+ "loss": 2.4317,
1338
+ "step": 188
1339
+ },
1340
+ {
1341
+ "epoch": 1.5241935483870968,
1342
+ "grad_norm": 73.09158325195312,
1343
+ "learning_rate": 8.168331915620885e-07,
1344
+ "loss": 1.3731,
1345
+ "step": 189
1346
+ },
1347
+ {
1348
+ "epoch": 1.532258064516129,
1349
+ "grad_norm": 92.6122817993164,
1350
+ "learning_rate": 8.149379869412948e-07,
1351
+ "loss": 1.0538,
1352
+ "step": 190
1353
+ },
1354
+ {
1355
+ "epoch": 1.5403225806451613,
1356
+ "grad_norm": 61.34569549560547,
1357
+ "learning_rate": 8.130427823205009e-07,
1358
+ "loss": 0.8926,
1359
+ "step": 191
1360
+ },
1361
+ {
1362
+ "epoch": 1.5483870967741935,
1363
+ "grad_norm": 46.87858200073242,
1364
+ "learning_rate": 8.111475776997073e-07,
1365
+ "loss": 0.774,
1366
+ "step": 192
1367
+ },
1368
+ {
1369
+ "epoch": 1.5564516129032258,
1370
+ "grad_norm": 32.91598129272461,
1371
+ "learning_rate": 8.092523730789136e-07,
1372
+ "loss": 0.6749,
1373
+ "step": 193
1374
+ },
1375
+ {
1376
+ "epoch": 1.564516129032258,
1377
+ "grad_norm": 69.08930206298828,
1378
+ "learning_rate": 8.073571684581199e-07,
1379
+ "loss": 0.7667,
1380
+ "step": 194
1381
+ },
1382
+ {
1383
+ "epoch": 1.5725806451612905,
1384
+ "grad_norm": 61.4034538269043,
1385
+ "learning_rate": 8.054619638373261e-07,
1386
+ "loss": 0.6325,
1387
+ "step": 195
1388
+ },
1389
+ {
1390
+ "epoch": 1.5806451612903225,
1391
+ "grad_norm": 252.8091583251953,
1392
+ "learning_rate": 8.035667592165325e-07,
1393
+ "loss": 9.2466,
1394
+ "step": 196
1395
+ },
1396
+ {
1397
+ "epoch": 1.588709677419355,
1398
+ "grad_norm": 60.166046142578125,
1399
+ "learning_rate": 8.016715545957387e-07,
1400
+ "loss": 1.7571,
1401
+ "step": 197
1402
+ },
1403
+ {
1404
+ "epoch": 1.596774193548387,
1405
+ "grad_norm": 42.4896125793457,
1406
+ "learning_rate": 7.997763499749451e-07,
1407
+ "loss": 0.5738,
1408
+ "step": 198
1409
+ },
1410
+ {
1411
+ "epoch": 1.6048387096774195,
1412
+ "grad_norm": 62.25552749633789,
1413
+ "learning_rate": 7.978811453541514e-07,
1414
+ "loss": 0.356,
1415
+ "step": 199
1416
+ },
1417
+ {
1418
+ "epoch": 1.6129032258064515,
1419
+ "grad_norm": 131.12069702148438,
1420
+ "learning_rate": 7.959859407333575e-07,
1421
+ "loss": 1.0522,
1422
+ "step": 200
1423
+ },
1424
+ {
1425
+ "epoch": 1.620967741935484,
1426
+ "grad_norm": 13.091470718383789,
1427
+ "learning_rate": 7.940907361125639e-07,
1428
+ "loss": 0.2088,
1429
+ "step": 201
1430
+ },
1431
+ {
1432
+ "epoch": 1.629032258064516,
1433
+ "grad_norm": 61.03318786621094,
1434
+ "learning_rate": 7.921955314917702e-07,
1435
+ "loss": 1.6789,
1436
+ "step": 202
1437
+ },
1438
+ {
1439
+ "epoch": 1.6370967741935485,
1440
+ "grad_norm": 14.173700332641602,
1441
+ "learning_rate": 7.903003268709766e-07,
1442
+ "loss": 0.1112,
1443
+ "step": 203
1444
+ },
1445
+ {
1446
+ "epoch": 1.6451612903225805,
1447
+ "grad_norm": 70.016845703125,
1448
+ "learning_rate": 7.884051222501828e-07,
1449
+ "loss": 0.8495,
1450
+ "step": 204
1451
+ },
1452
+ {
1453
+ "epoch": 1.653225806451613,
1454
+ "grad_norm": 217.45611572265625,
1455
+ "learning_rate": 7.865099176293891e-07,
1456
+ "loss": 15.7051,
1457
+ "step": 205
1458
+ },
1459
+ {
1460
+ "epoch": 1.661290322580645,
1461
+ "grad_norm": 99.29098510742188,
1462
+ "learning_rate": 7.846147130085954e-07,
1463
+ "loss": 1.0348,
1464
+ "step": 206
1465
+ },
1466
+ {
1467
+ "epoch": 1.6693548387096775,
1468
+ "grad_norm": 33.91508102416992,
1469
+ "learning_rate": 7.827195083878016e-07,
1470
+ "loss": 0.7734,
1471
+ "step": 207
1472
+ },
1473
+ {
1474
+ "epoch": 1.6774193548387095,
1475
+ "grad_norm": 93.15892028808594,
1476
+ "learning_rate": 7.80824303767008e-07,
1477
+ "loss": 0.3326,
1478
+ "step": 208
1479
+ },
1480
+ {
1481
+ "epoch": 1.685483870967742,
1482
+ "grad_norm": 14.714130401611328,
1483
+ "learning_rate": 7.789290991462143e-07,
1484
+ "loss": 0.3946,
1485
+ "step": 209
1486
+ },
1487
+ {
1488
+ "epoch": 1.6935483870967742,
1489
+ "grad_norm": 22.98834800720215,
1490
+ "learning_rate": 7.770338945254205e-07,
1491
+ "loss": 0.2196,
1492
+ "step": 210
1493
+ },
1494
+ {
1495
+ "epoch": 1.7016129032258065,
1496
+ "grad_norm": 165.18626403808594,
1497
+ "learning_rate": 7.751386899046268e-07,
1498
+ "loss": 3.0653,
1499
+ "step": 211
1500
+ },
1501
+ {
1502
+ "epoch": 1.7096774193548387,
1503
+ "grad_norm": 55.49916458129883,
1504
+ "learning_rate": 7.732434852838332e-07,
1505
+ "loss": 0.8254,
1506
+ "step": 212
1507
+ },
1508
+ {
1509
+ "epoch": 1.717741935483871,
1510
+ "grad_norm": 25.563867568969727,
1511
+ "learning_rate": 7.713482806630394e-07,
1512
+ "loss": 0.1924,
1513
+ "step": 213
1514
+ },
1515
+ {
1516
+ "epoch": 1.7258064516129032,
1517
+ "grad_norm": 277.5033874511719,
1518
+ "learning_rate": 7.694530760422456e-07,
1519
+ "loss": 26.1251,
1520
+ "step": 214
1521
+ },
1522
+ {
1523
+ "epoch": 1.7338709677419355,
1524
+ "grad_norm": 218.53463745117188,
1525
+ "learning_rate": 7.67557871421452e-07,
1526
+ "loss": 16.69,
1527
+ "step": 215
1528
+ },
1529
+ {
1530
+ "epoch": 1.7419354838709677,
1531
+ "grad_norm": 48.32767868041992,
1532
+ "learning_rate": 7.656626668006582e-07,
1533
+ "loss": 1.267,
1534
+ "step": 216
1535
+ },
1536
+ {
1537
+ "epoch": 1.75,
1538
+ "grad_norm": 297.5074462890625,
1539
+ "learning_rate": 7.637674621798646e-07,
1540
+ "loss": 8.8952,
1541
+ "step": 217
1542
+ },
1543
+ {
1544
+ "epoch": 1.7580645161290323,
1545
+ "grad_norm": 19.5325927734375,
1546
+ "learning_rate": 7.618722575590709e-07,
1547
+ "loss": 0.1047,
1548
+ "step": 218
1549
+ },
1550
+ {
1551
+ "epoch": 1.7661290322580645,
1552
+ "grad_norm": 382.64434814453125,
1553
+ "learning_rate": 7.599770529382771e-07,
1554
+ "loss": 18.6839,
1555
+ "step": 219
1556
+ },
1557
+ {
1558
+ "epoch": 1.7741935483870968,
1559
+ "grad_norm": 18.517972946166992,
1560
+ "learning_rate": 7.580818483174834e-07,
1561
+ "loss": 0.1679,
1562
+ "step": 220
1563
+ },
1564
+ {
1565
+ "epoch": 1.782258064516129,
1566
+ "grad_norm": 36.31782913208008,
1567
+ "learning_rate": 7.561866436966898e-07,
1568
+ "loss": 0.1592,
1569
+ "step": 221
1570
+ },
1571
+ {
1572
+ "epoch": 1.7903225806451613,
1573
+ "grad_norm": 44.47116470336914,
1574
+ "learning_rate": 7.54291439075896e-07,
1575
+ "loss": 0.6091,
1576
+ "step": 222
1577
+ },
1578
+ {
1579
+ "epoch": 1.7983870967741935,
1580
+ "grad_norm": 16.11575698852539,
1581
+ "learning_rate": 7.523962344551024e-07,
1582
+ "loss": 0.1174,
1583
+ "step": 223
1584
+ },
1585
+ {
1586
+ "epoch": 1.8064516129032258,
1587
+ "grad_norm": 288.2611999511719,
1588
+ "learning_rate": 7.505010298343086e-07,
1589
+ "loss": 18.947,
1590
+ "step": 224
1591
+ },
1592
+ {
1593
+ "epoch": 1.814516129032258,
1594
+ "grad_norm": 267.47540283203125,
1595
+ "learning_rate": 7.486058252135149e-07,
1596
+ "loss": 10.6328,
1597
+ "step": 225
1598
+ },
1599
+ {
1600
+ "epoch": 1.8225806451612905,
1601
+ "grad_norm": 33.86650848388672,
1602
+ "learning_rate": 7.467106205927212e-07,
1603
+ "loss": 2.1226,
1604
+ "step": 226
1605
+ },
1606
+ {
1607
+ "epoch": 1.8306451612903225,
1608
+ "grad_norm": 121.6884765625,
1609
+ "learning_rate": 7.448154159719275e-07,
1610
+ "loss": 0.371,
1611
+ "step": 227
1612
+ },
1613
+ {
1614
+ "epoch": 1.838709677419355,
1615
+ "grad_norm": 42.18556594848633,
1616
+ "learning_rate": 7.429202113511338e-07,
1617
+ "loss": 0.7408,
1618
+ "step": 228
1619
+ },
1620
+ {
1621
+ "epoch": 1.846774193548387,
1622
+ "grad_norm": 64.0304946899414,
1623
+ "learning_rate": 7.4102500673034e-07,
1624
+ "loss": 0.7927,
1625
+ "step": 229
1626
+ },
1627
+ {
1628
+ "epoch": 1.8548387096774195,
1629
+ "grad_norm": 211.26959228515625,
1630
+ "learning_rate": 7.391298021095464e-07,
1631
+ "loss": 8.6268,
1632
+ "step": 230
1633
+ },
1634
+ {
1635
+ "epoch": 1.8629032258064515,
1636
+ "grad_norm": 97.74939727783203,
1637
+ "learning_rate": 7.372345974887527e-07,
1638
+ "loss": 1.1381,
1639
+ "step": 231
1640
+ },
1641
+ {
1642
+ "epoch": 1.870967741935484,
1643
+ "grad_norm": 325.8834228515625,
1644
+ "learning_rate": 7.35339392867959e-07,
1645
+ "loss": 1.0786,
1646
+ "step": 232
1647
+ },
1648
+ {
1649
+ "epoch": 1.879032258064516,
1650
+ "grad_norm": 49.10792541503906,
1651
+ "learning_rate": 7.334441882471652e-07,
1652
+ "loss": 0.5383,
1653
+ "step": 233
1654
+ },
1655
+ {
1656
+ "epoch": 1.8870967741935485,
1657
+ "grad_norm": 27.410547256469727,
1658
+ "learning_rate": 7.315489836263715e-07,
1659
+ "loss": 0.4044,
1660
+ "step": 234
1661
+ },
1662
+ {
1663
+ "epoch": 1.8951612903225805,
1664
+ "grad_norm": 31.08856773376465,
1665
+ "learning_rate": 7.296537790055778e-07,
1666
+ "loss": 0.2857,
1667
+ "step": 235
1668
+ },
1669
+ {
1670
+ "epoch": 1.903225806451613,
1671
+ "grad_norm": 236.7850341796875,
1672
+ "learning_rate": 7.277585743847841e-07,
1673
+ "loss": 10.882,
1674
+ "step": 236
1675
+ },
1676
+ {
1677
+ "epoch": 1.911290322580645,
1678
+ "grad_norm": 85.23385620117188,
1679
+ "learning_rate": 7.258633697639905e-07,
1680
+ "loss": 6.385,
1681
+ "step": 237
1682
+ },
1683
+ {
1684
+ "epoch": 1.9193548387096775,
1685
+ "grad_norm": 42.652278900146484,
1686
+ "learning_rate": 7.239681651431966e-07,
1687
+ "loss": 0.7745,
1688
+ "step": 238
1689
+ },
1690
+ {
1691
+ "epoch": 1.9274193548387095,
1692
+ "grad_norm": 37.44199752807617,
1693
+ "learning_rate": 7.22072960522403e-07,
1694
+ "loss": 0.6186,
1695
+ "step": 239
1696
+ },
1697
+ {
1698
+ "epoch": 1.935483870967742,
1699
+ "grad_norm": 122.3661117553711,
1700
+ "learning_rate": 7.201777559016093e-07,
1701
+ "loss": 3.3579,
1702
+ "step": 240
1703
+ },
1704
+ {
1705
+ "epoch": 1.9435483870967742,
1706
+ "grad_norm": 29.615245819091797,
1707
+ "learning_rate": 7.182825512808157e-07,
1708
+ "loss": 1.0682,
1709
+ "step": 241
1710
+ },
1711
+ {
1712
+ "epoch": 1.9516129032258065,
1713
+ "grad_norm": 38.644981384277344,
1714
+ "learning_rate": 7.163873466600218e-07,
1715
+ "loss": 1.6679,
1716
+ "step": 242
1717
+ },
1718
+ {
1719
+ "epoch": 1.9596774193548387,
1720
+ "grad_norm": 151.14231872558594,
1721
+ "learning_rate": 7.144921420392281e-07,
1722
+ "loss": 3.0028,
1723
+ "step": 243
1724
+ },
1725
+ {
1726
+ "epoch": 1.967741935483871,
1727
+ "grad_norm": 90.12117767333984,
1728
+ "learning_rate": 7.125969374184345e-07,
1729
+ "loss": 3.1295,
1730
+ "step": 244
1731
+ },
1732
+ {
1733
+ "epoch": 1.9758064516129032,
1734
+ "grad_norm": 65.80992889404297,
1735
+ "learning_rate": 7.107017327976407e-07,
1736
+ "loss": 0.8262,
1737
+ "step": 245
1738
+ },
1739
+ {
1740
+ "epoch": 1.9838709677419355,
1741
+ "grad_norm": 224.11376953125,
1742
+ "learning_rate": 7.088065281768471e-07,
1743
+ "loss": 2.9559,
1744
+ "step": 246
1745
+ },
1746
+ {
1747
+ "epoch": 1.9919354838709677,
1748
+ "grad_norm": 165.00535583496094,
1749
+ "learning_rate": 7.069113235560533e-07,
1750
+ "loss": 1.3005,
1751
+ "step": 247
1752
+ },
1753
+ {
1754
+ "epoch": 2.0,
1755
+ "grad_norm": 178.8065185546875,
1756
+ "learning_rate": 7.050161189352596e-07,
1757
+ "loss": 2.704,
1758
+ "step": 248
1759
+ },
1760
+ {
1761
+ "epoch": 2.0,
1762
+ "eval_loss": 3.476463556289673,
1763
+ "eval_mae": 1.1249998807907104,
1764
+ "eval_mse": 3.476463794708252,
1765
+ "eval_r2": 0.04820472002029419,
1766
+ "eval_rmse": 1.8645277672129885,
1767
+ "eval_runtime": 1.4258,
1768
+ "eval_samples_per_second": 38.574,
1769
+ "eval_smape": 47.31981158256531,
1770
+ "eval_steps_per_second": 9.819,
1771
+ "step": 248
1772
+ }
1773
+ ],
1774
+ "logging_steps": 1,
1775
+ "max_steps": 620,
1776
+ "num_input_tokens_seen": 0,
1777
+ "num_train_epochs": 5,
1778
+ "save_steps": 500,
1779
+ "stateful_callbacks": {
1780
+ "TrainerControl": {
1781
+ "args": {
1782
+ "should_epoch_stop": false,
1783
+ "should_evaluate": false,
1784
+ "should_log": false,
1785
+ "should_save": true,
1786
+ "should_training_stop": false
1787
+ },
1788
+ "attributes": {}
1789
+ }
1790
+ },
1791
+ "total_flos": 1343939627507712.0,
1792
+ "train_batch_size": 4,
1793
+ "trial_name": null,
1794
+ "trial_params": {
1795
+ "learning_rate": 1.1750268648920993e-06
1796
+ }
1797
+ }
run-3/checkpoint-248/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:337213442d1d720bd2aad2abb97c003e5615771062861d03db91a7a8f9420b71
3
+ size 5432
run-4/checkpoint-620/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "answerdotai/ModernBERT-base",
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": 768,
23
+ "id2label": {
24
+ "0": "LABEL_0"
25
+ },
26
+ "initializer_cutoff_factor": 2.0,
27
+ "initializer_range": 0.02,
28
+ "intermediate_size": 1152,
29
+ "label2id": {
30
+ "LABEL_0": 0
31
+ },
32
+ "layer_norm_eps": 1e-05,
33
+ "local_attention": 128,
34
+ "local_rope_theta": 10000.0,
35
+ "max_position_embeddings": 8192,
36
+ "mlp_bias": false,
37
+ "mlp_dropout": 0.0,
38
+ "model_type": "modernbert",
39
+ "norm_bias": false,
40
+ "norm_eps": 1e-05,
41
+ "num_attention_heads": 12,
42
+ "num_hidden_layers": 22,
43
+ "pad_token_id": 50283,
44
+ "position_embedding_type": "absolute",
45
+ "problem_type": "regression",
46
+ "reference_compile": true,
47
+ "sep_token_id": 50282,
48
+ "sparse_pred_ignore_index": -100,
49
+ "sparse_prediction": false,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.48.0.dev0",
52
+ "vocab_size": 50368
53
+ }
run-4/checkpoint-620/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c118bd41ab3f9771fa546fe780f292cd77bcc4690108473ac02c0a5823716a6
3
+ size 598436708
run-4/checkpoint-620/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:664d9e5c6ea2e501054a861155046ec480def3555d356b6164853bd29beeb3dd
3
+ size 1196963194
run-4/checkpoint-620/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5c27b160b159e371c7d5122ee5052944383a2bf6a7498d51d058a4a6a3dd7d
3
+ size 14244
run-4/checkpoint-620/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:049fae4dfe486113308d6608738a93bfaaaa58f9971eab6f1cbf4e5222184c38
3
+ size 1064
run-4/checkpoint-620/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": true,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
run-4/checkpoint-620/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-620/tokenizer_config.json ADDED
@@ -0,0 +1,945 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "|||IP_ADDRESS|||",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": false
10
+ },
11
+ "1": {
12
+ "content": "<|padding|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "50254": {
20
+ "content": " ",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": false
26
+ },
27
+ "50255": {
28
+ "content": " ",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": false
34
+ },
35
+ "50256": {
36
+ "content": " ",
37
+ "lstrip": false,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": false
42
+ },
43
+ "50257": {
44
+ "content": " ",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ },
51
+ "50258": {
52
+ "content": " ",
53
+ "lstrip": false,
54
+ "normalized": true,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": false
58
+ },
59
+ "50259": {
60
+ "content": " ",
61
+ "lstrip": false,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": false
66
+ },
67
+ "50260": {
68
+ "content": " ",
69
+ "lstrip": false,
70
+ "normalized": true,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": false
74
+ },
75
+ "50261": {
76
+ "content": " ",
77
+ "lstrip": false,
78
+ "normalized": true,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": false
82
+ },
83
+ "50262": {
84
+ "content": " ",
85
+ "lstrip": false,
86
+ "normalized": true,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": false
90
+ },
91
+ "50263": {
92
+ "content": " ",
93
+ "lstrip": false,
94
+ "normalized": true,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": false
98
+ },
99
+ "50264": {
100
+ "content": " ",
101
+ "lstrip": false,
102
+ "normalized": true,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": false
106
+ },
107
+ "50265": {
108
+ "content": " ",
109
+ "lstrip": false,
110
+ "normalized": true,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": false
114
+ },
115
+ "50266": {
116
+ "content": " ",
117
+ "lstrip": false,
118
+ "normalized": true,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": false
122
+ },
123
+ "50267": {
124
+ "content": " ",
125
+ "lstrip": false,
126
+ "normalized": true,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": false
130
+ },
131
+ "50268": {
132
+ "content": " ",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": false
138
+ },
139
+ "50269": {
140
+ "content": " ",
141
+ "lstrip": false,
142
+ "normalized": true,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": false
146
+ },
147
+ "50270": {
148
+ "content": " ",
149
+ "lstrip": false,
150
+ "normalized": true,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": false
154
+ },
155
+ "50271": {
156
+ "content": " ",
157
+ "lstrip": false,
158
+ "normalized": true,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": false
162
+ },
163
+ "50272": {
164
+ "content": " ",
165
+ "lstrip": false,
166
+ "normalized": true,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": false
170
+ },
171
+ "50273": {
172
+ "content": " ",
173
+ "lstrip": false,
174
+ "normalized": true,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": false
178
+ },
179
+ "50274": {
180
+ "content": " ",
181
+ "lstrip": false,
182
+ "normalized": true,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": false
186
+ },
187
+ "50275": {
188
+ "content": " ",
189
+ "lstrip": false,
190
+ "normalized": true,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": false
194
+ },
195
+ "50276": {
196
+ "content": " ",
197
+ "lstrip": false,
198
+ "normalized": true,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": false
202
+ },
203
+ "50277": {
204
+ "content": "|||EMAIL_ADDRESS|||",
205
+ "lstrip": false,
206
+ "normalized": true,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": false
210
+ },
211
+ "50278": {
212
+ "content": "|||PHONE_NUMBER|||",
213
+ "lstrip": false,
214
+ "normalized": true,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": false
218
+ },
219
+ "50279": {
220
+ "content": "<|endoftext|>",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "50280": {
228
+ "content": "[UNK]",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "50281": {
236
+ "content": "[CLS]",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "50282": {
244
+ "content": "[SEP]",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "50283": {
252
+ "content": "[PAD]",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "50284": {
260
+ "content": "[MASK]",
261
+ "lstrip": true,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "50285": {
268
+ "content": "[unused0]",
269
+ "lstrip": false,
270
+ "normalized": true,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": false
274
+ },
275
+ "50286": {
276
+ "content": "[unused1]",
277
+ "lstrip": false,
278
+ "normalized": true,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": false
282
+ },
283
+ "50287": {
284
+ "content": "[unused2]",
285
+ "lstrip": false,
286
+ "normalized": true,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": false
290
+ },
291
+ "50288": {
292
+ "content": "[unused3]",
293
+ "lstrip": false,
294
+ "normalized": true,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": false
298
+ },
299
+ "50289": {
300
+ "content": "[unused4]",
301
+ "lstrip": false,
302
+ "normalized": true,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": false
306
+ },
307
+ "50290": {
308
+ "content": "[unused5]",
309
+ "lstrip": false,
310
+ "normalized": true,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": false
314
+ },
315
+ "50291": {
316
+ "content": "[unused6]",
317
+ "lstrip": false,
318
+ "normalized": true,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": false
322
+ },
323
+ "50292": {
324
+ "content": "[unused7]",
325
+ "lstrip": false,
326
+ "normalized": true,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": false
330
+ },
331
+ "50293": {
332
+ "content": "[unused8]",
333
+ "lstrip": false,
334
+ "normalized": true,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": false
338
+ },
339
+ "50294": {
340
+ "content": "[unused9]",
341
+ "lstrip": false,
342
+ "normalized": true,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": false
346
+ },
347
+ "50295": {
348
+ "content": "[unused10]",
349
+ "lstrip": false,
350
+ "normalized": true,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": false
354
+ },
355
+ "50296": {
356
+ "content": "[unused11]",
357
+ "lstrip": false,
358
+ "normalized": true,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": false
362
+ },
363
+ "50297": {
364
+ "content": "[unused12]",
365
+ "lstrip": false,
366
+ "normalized": true,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": false
370
+ },
371
+ "50298": {
372
+ "content": "[unused13]",
373
+ "lstrip": false,
374
+ "normalized": true,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": false
378
+ },
379
+ "50299": {
380
+ "content": "[unused14]",
381
+ "lstrip": false,
382
+ "normalized": true,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": false
386
+ },
387
+ "50300": {
388
+ "content": "[unused15]",
389
+ "lstrip": false,
390
+ "normalized": true,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": false
394
+ },
395
+ "50301": {
396
+ "content": "[unused16]",
397
+ "lstrip": false,
398
+ "normalized": true,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": false
402
+ },
403
+ "50302": {
404
+ "content": "[unused17]",
405
+ "lstrip": false,
406
+ "normalized": true,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": false
410
+ },
411
+ "50303": {
412
+ "content": "[unused18]",
413
+ "lstrip": false,
414
+ "normalized": true,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": false
418
+ },
419
+ "50304": {
420
+ "content": "[unused19]",
421
+ "lstrip": false,
422
+ "normalized": true,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": false
426
+ },
427
+ "50305": {
428
+ "content": "[unused20]",
429
+ "lstrip": false,
430
+ "normalized": true,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": false
434
+ },
435
+ "50306": {
436
+ "content": "[unused21]",
437
+ "lstrip": false,
438
+ "normalized": true,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": false
442
+ },
443
+ "50307": {
444
+ "content": "[unused22]",
445
+ "lstrip": false,
446
+ "normalized": true,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": false
450
+ },
451
+ "50308": {
452
+ "content": "[unused23]",
453
+ "lstrip": false,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": false
458
+ },
459
+ "50309": {
460
+ "content": "[unused24]",
461
+ "lstrip": false,
462
+ "normalized": true,
463
+ "rstrip": false,
464
+ "single_word": false,
465
+ "special": false
466
+ },
467
+ "50310": {
468
+ "content": "[unused25]",
469
+ "lstrip": false,
470
+ "normalized": true,
471
+ "rstrip": false,
472
+ "single_word": false,
473
+ "special": false
474
+ },
475
+ "50311": {
476
+ "content": "[unused26]",
477
+ "lstrip": false,
478
+ "normalized": true,
479
+ "rstrip": false,
480
+ "single_word": false,
481
+ "special": false
482
+ },
483
+ "50312": {
484
+ "content": "[unused27]",
485
+ "lstrip": false,
486
+ "normalized": true,
487
+ "rstrip": false,
488
+ "single_word": false,
489
+ "special": false
490
+ },
491
+ "50313": {
492
+ "content": "[unused28]",
493
+ "lstrip": false,
494
+ "normalized": true,
495
+ "rstrip": false,
496
+ "single_word": false,
497
+ "special": false
498
+ },
499
+ "50314": {
500
+ "content": "[unused29]",
501
+ "lstrip": false,
502
+ "normalized": true,
503
+ "rstrip": false,
504
+ "single_word": false,
505
+ "special": false
506
+ },
507
+ "50315": {
508
+ "content": "[unused30]",
509
+ "lstrip": false,
510
+ "normalized": true,
511
+ "rstrip": false,
512
+ "single_word": false,
513
+ "special": false
514
+ },
515
+ "50316": {
516
+ "content": "[unused31]",
517
+ "lstrip": false,
518
+ "normalized": true,
519
+ "rstrip": false,
520
+ "single_word": false,
521
+ "special": false
522
+ },
523
+ "50317": {
524
+ "content": "[unused32]",
525
+ "lstrip": false,
526
+ "normalized": true,
527
+ "rstrip": false,
528
+ "single_word": false,
529
+ "special": false
530
+ },
531
+ "50318": {
532
+ "content": "[unused33]",
533
+ "lstrip": false,
534
+ "normalized": true,
535
+ "rstrip": false,
536
+ "single_word": false,
537
+ "special": false
538
+ },
539
+ "50319": {
540
+ "content": "[unused34]",
541
+ "lstrip": false,
542
+ "normalized": true,
543
+ "rstrip": false,
544
+ "single_word": false,
545
+ "special": false
546
+ },
547
+ "50320": {
548
+ "content": "[unused35]",
549
+ "lstrip": false,
550
+ "normalized": true,
551
+ "rstrip": false,
552
+ "single_word": false,
553
+ "special": false
554
+ },
555
+ "50321": {
556
+ "content": "[unused36]",
557
+ "lstrip": false,
558
+ "normalized": true,
559
+ "rstrip": false,
560
+ "single_word": false,
561
+ "special": false
562
+ },
563
+ "50322": {
564
+ "content": "[unused37]",
565
+ "lstrip": false,
566
+ "normalized": true,
567
+ "rstrip": false,
568
+ "single_word": false,
569
+ "special": false
570
+ },
571
+ "50323": {
572
+ "content": "[unused38]",
573
+ "lstrip": false,
574
+ "normalized": true,
575
+ "rstrip": false,
576
+ "single_word": false,
577
+ "special": false
578
+ },
579
+ "50324": {
580
+ "content": "[unused39]",
581
+ "lstrip": false,
582
+ "normalized": true,
583
+ "rstrip": false,
584
+ "single_word": false,
585
+ "special": false
586
+ },
587
+ "50325": {
588
+ "content": "[unused40]",
589
+ "lstrip": false,
590
+ "normalized": true,
591
+ "rstrip": false,
592
+ "single_word": false,
593
+ "special": false
594
+ },
595
+ "50326": {
596
+ "content": "[unused41]",
597
+ "lstrip": false,
598
+ "normalized": true,
599
+ "rstrip": false,
600
+ "single_word": false,
601
+ "special": false
602
+ },
603
+ "50327": {
604
+ "content": "[unused42]",
605
+ "lstrip": false,
606
+ "normalized": true,
607
+ "rstrip": false,
608
+ "single_word": false,
609
+ "special": false
610
+ },
611
+ "50328": {
612
+ "content": "[unused43]",
613
+ "lstrip": false,
614
+ "normalized": true,
615
+ "rstrip": false,
616
+ "single_word": false,
617
+ "special": false
618
+ },
619
+ "50329": {
620
+ "content": "[unused44]",
621
+ "lstrip": false,
622
+ "normalized": true,
623
+ "rstrip": false,
624
+ "single_word": false,
625
+ "special": false
626
+ },
627
+ "50330": {
628
+ "content": "[unused45]",
629
+ "lstrip": false,
630
+ "normalized": true,
631
+ "rstrip": false,
632
+ "single_word": false,
633
+ "special": false
634
+ },
635
+ "50331": {
636
+ "content": "[unused46]",
637
+ "lstrip": false,
638
+ "normalized": true,
639
+ "rstrip": false,
640
+ "single_word": false,
641
+ "special": false
642
+ },
643
+ "50332": {
644
+ "content": "[unused47]",
645
+ "lstrip": false,
646
+ "normalized": true,
647
+ "rstrip": false,
648
+ "single_word": false,
649
+ "special": false
650
+ },
651
+ "50333": {
652
+ "content": "[unused48]",
653
+ "lstrip": false,
654
+ "normalized": true,
655
+ "rstrip": false,
656
+ "single_word": false,
657
+ "special": false
658
+ },
659
+ "50334": {
660
+ "content": "[unused49]",
661
+ "lstrip": false,
662
+ "normalized": true,
663
+ "rstrip": false,
664
+ "single_word": false,
665
+ "special": false
666
+ },
667
+ "50335": {
668
+ "content": "[unused50]",
669
+ "lstrip": false,
670
+ "normalized": true,
671
+ "rstrip": false,
672
+ "single_word": false,
673
+ "special": false
674
+ },
675
+ "50336": {
676
+ "content": "[unused51]",
677
+ "lstrip": false,
678
+ "normalized": true,
679
+ "rstrip": false,
680
+ "single_word": false,
681
+ "special": false
682
+ },
683
+ "50337": {
684
+ "content": "[unused52]",
685
+ "lstrip": false,
686
+ "normalized": true,
687
+ "rstrip": false,
688
+ "single_word": false,
689
+ "special": false
690
+ },
691
+ "50338": {
692
+ "content": "[unused53]",
693
+ "lstrip": false,
694
+ "normalized": true,
695
+ "rstrip": false,
696
+ "single_word": false,
697
+ "special": false
698
+ },
699
+ "50339": {
700
+ "content": "[unused54]",
701
+ "lstrip": false,
702
+ "normalized": true,
703
+ "rstrip": false,
704
+ "single_word": false,
705
+ "special": false
706
+ },
707
+ "50340": {
708
+ "content": "[unused55]",
709
+ "lstrip": false,
710
+ "normalized": true,
711
+ "rstrip": false,
712
+ "single_word": false,
713
+ "special": false
714
+ },
715
+ "50341": {
716
+ "content": "[unused56]",
717
+ "lstrip": false,
718
+ "normalized": true,
719
+ "rstrip": false,
720
+ "single_word": false,
721
+ "special": false
722
+ },
723
+ "50342": {
724
+ "content": "[unused57]",
725
+ "lstrip": false,
726
+ "normalized": true,
727
+ "rstrip": false,
728
+ "single_word": false,
729
+ "special": false
730
+ },
731
+ "50343": {
732
+ "content": "[unused58]",
733
+ "lstrip": false,
734
+ "normalized": true,
735
+ "rstrip": false,
736
+ "single_word": false,
737
+ "special": false
738
+ },
739
+ "50344": {
740
+ "content": "[unused59]",
741
+ "lstrip": false,
742
+ "normalized": true,
743
+ "rstrip": false,
744
+ "single_word": false,
745
+ "special": false
746
+ },
747
+ "50345": {
748
+ "content": "[unused60]",
749
+ "lstrip": false,
750
+ "normalized": true,
751
+ "rstrip": false,
752
+ "single_word": false,
753
+ "special": false
754
+ },
755
+ "50346": {
756
+ "content": "[unused61]",
757
+ "lstrip": false,
758
+ "normalized": true,
759
+ "rstrip": false,
760
+ "single_word": false,
761
+ "special": false
762
+ },
763
+ "50347": {
764
+ "content": "[unused62]",
765
+ "lstrip": false,
766
+ "normalized": true,
767
+ "rstrip": false,
768
+ "single_word": false,
769
+ "special": false
770
+ },
771
+ "50348": {
772
+ "content": "[unused63]",
773
+ "lstrip": false,
774
+ "normalized": true,
775
+ "rstrip": false,
776
+ "single_word": false,
777
+ "special": false
778
+ },
779
+ "50349": {
780
+ "content": "[unused64]",
781
+ "lstrip": false,
782
+ "normalized": true,
783
+ "rstrip": false,
784
+ "single_word": false,
785
+ "special": false
786
+ },
787
+ "50350": {
788
+ "content": "[unused65]",
789
+ "lstrip": false,
790
+ "normalized": true,
791
+ "rstrip": false,
792
+ "single_word": false,
793
+ "special": false
794
+ },
795
+ "50351": {
796
+ "content": "[unused66]",
797
+ "lstrip": false,
798
+ "normalized": true,
799
+ "rstrip": false,
800
+ "single_word": false,
801
+ "special": false
802
+ },
803
+ "50352": {
804
+ "content": "[unused67]",
805
+ "lstrip": false,
806
+ "normalized": true,
807
+ "rstrip": false,
808
+ "single_word": false,
809
+ "special": false
810
+ },
811
+ "50353": {
812
+ "content": "[unused68]",
813
+ "lstrip": false,
814
+ "normalized": true,
815
+ "rstrip": false,
816
+ "single_word": false,
817
+ "special": false
818
+ },
819
+ "50354": {
820
+ "content": "[unused69]",
821
+ "lstrip": false,
822
+ "normalized": true,
823
+ "rstrip": false,
824
+ "single_word": false,
825
+ "special": false
826
+ },
827
+ "50355": {
828
+ "content": "[unused70]",
829
+ "lstrip": false,
830
+ "normalized": true,
831
+ "rstrip": false,
832
+ "single_word": false,
833
+ "special": false
834
+ },
835
+ "50356": {
836
+ "content": "[unused71]",
837
+ "lstrip": false,
838
+ "normalized": true,
839
+ "rstrip": false,
840
+ "single_word": false,
841
+ "special": false
842
+ },
843
+ "50357": {
844
+ "content": "[unused72]",
845
+ "lstrip": false,
846
+ "normalized": true,
847
+ "rstrip": false,
848
+ "single_word": false,
849
+ "special": false
850
+ },
851
+ "50358": {
852
+ "content": "[unused73]",
853
+ "lstrip": false,
854
+ "normalized": true,
855
+ "rstrip": false,
856
+ "single_word": false,
857
+ "special": false
858
+ },
859
+ "50359": {
860
+ "content": "[unused74]",
861
+ "lstrip": false,
862
+ "normalized": true,
863
+ "rstrip": false,
864
+ "single_word": false,
865
+ "special": false
866
+ },
867
+ "50360": {
868
+ "content": "[unused75]",
869
+ "lstrip": false,
870
+ "normalized": true,
871
+ "rstrip": false,
872
+ "single_word": false,
873
+ "special": false
874
+ },
875
+ "50361": {
876
+ "content": "[unused76]",
877
+ "lstrip": false,
878
+ "normalized": true,
879
+ "rstrip": false,
880
+ "single_word": false,
881
+ "special": false
882
+ },
883
+ "50362": {
884
+ "content": "[unused77]",
885
+ "lstrip": false,
886
+ "normalized": true,
887
+ "rstrip": false,
888
+ "single_word": false,
889
+ "special": false
890
+ },
891
+ "50363": {
892
+ "content": "[unused78]",
893
+ "lstrip": false,
894
+ "normalized": true,
895
+ "rstrip": false,
896
+ "single_word": false,
897
+ "special": false
898
+ },
899
+ "50364": {
900
+ "content": "[unused79]",
901
+ "lstrip": false,
902
+ "normalized": true,
903
+ "rstrip": false,
904
+ "single_word": false,
905
+ "special": false
906
+ },
907
+ "50365": {
908
+ "content": "[unused80]",
909
+ "lstrip": false,
910
+ "normalized": true,
911
+ "rstrip": false,
912
+ "single_word": false,
913
+ "special": false
914
+ },
915
+ "50366": {
916
+ "content": "[unused81]",
917
+ "lstrip": false,
918
+ "normalized": true,
919
+ "rstrip": false,
920
+ "single_word": false,
921
+ "special": false
922
+ },
923
+ "50367": {
924
+ "content": "[unused82]",
925
+ "lstrip": false,
926
+ "normalized": true,
927
+ "rstrip": false,
928
+ "single_word": false,
929
+ "special": false
930
+ }
931
+ },
932
+ "clean_up_tokenization_spaces": true,
933
+ "cls_token": "[CLS]",
934
+ "extra_special_tokens": {},
935
+ "mask_token": "[MASK]",
936
+ "model_input_names": [
937
+ "input_ids",
938
+ "attention_mask"
939
+ ],
940
+ "model_max_length": 2048,
941
+ "pad_token": "[PAD]",
942
+ "sep_token": "[SEP]",
943
+ "tokenizer_class": "PreTrainedTokenizerFast",
944
+ "unk_token": "[UNK]"
945
+ }
run-4/checkpoint-620/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff