Jackie999 commited on
Commit
1ce6411
·
verified ·
1 Parent(s): 2d2a40a

Model save

Browse files
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: meta-llama/Meta-Llama-3-8B
9
+ model-index:
10
+ - name: llama3-10p
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # llama3-10p
18
+
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: nan
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0002
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 1
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 4
46
+ - gradient_accumulation_steps: 2
47
+ - total_train_batch_size: 64
48
+ - total_eval_batch_size: 4
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.1
52
+ - num_epochs: 1
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:-----:|:----:|:---------------:|
58
+ | 0.8719 | 1.0 | 974 | nan |
59
+
60
+
61
+ ### Framework versions
62
+
63
+ - PEFT 0.7.1
64
+ - Transformers 4.39.0.dev0
65
+ - Pytorch 2.1.2
66
+ - Datasets 2.14.6
67
+ - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3016b9bc8fc2052b989b91955bd48b1ddb4207d19bed4cb2717d8cfaf1ac9b2f
3
  size 31516744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d791b1b59cd974039a7290e2bb11701b1861bdc35372c3ba58094837e20e6472
3
  size 31516744
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.838743479834445,
4
+ "train_runtime": 18938.875,
5
+ "train_samples": 62359,
6
+ "train_samples_per_second": 3.293,
7
+ "train_steps_per_second": 0.051
8
+ }
runs/May09_05-52-16_COE-CS-sv004/events.out.tfevents.1715233961.COE-CS-sv004.2132936.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af3bf04085315d5e0caf3359be06b9e41ef425d90245ce5c58d83c05386aa8fd
3
- size 42781
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e46f8e3f384e0b98f15311fcd51f4dcd105af25db977a16c34b38b9162f039c
3
+ size 46360
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.838743479834445,
4
+ "train_runtime": 18938.875,
5
+ "train_samples": 62359,
6
+ "train_samples_per_second": 3.293,
7
+ "train_steps_per_second": 0.051
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1403 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9994869163673679,
5
+ "eval_steps": 500,
6
+ "global_step": 974,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "grad_norm": 0.41478540297292715,
14
+ "learning_rate": 2.040816326530612e-06,
15
+ "loss": 1.0591,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "grad_norm": 0.5358244986842327,
21
+ "learning_rate": 1.0204081632653061e-05,
22
+ "loss": 0.9762,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "grad_norm": 0.48168298126198883,
28
+ "learning_rate": 2.0408163265306123e-05,
29
+ "loss": 1.0246,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.02,
34
+ "grad_norm": 0.23532241564777356,
35
+ "learning_rate": 3.061224489795919e-05,
36
+ "loss": 1.0114,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.02,
41
+ "grad_norm": 0.2679701566666925,
42
+ "learning_rate": 4.0816326530612245e-05,
43
+ "loss": 0.9971,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "grad_norm": 0.1583325918997123,
49
+ "learning_rate": 5.102040816326531e-05,
50
+ "loss": 0.9355,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "grad_norm": 0.1840547641781333,
56
+ "learning_rate": 6.122448979591838e-05,
57
+ "loss": 0.943,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.04,
62
+ "grad_norm": 0.239483711202957,
63
+ "learning_rate": 7.142857142857143e-05,
64
+ "loss": 0.8764,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.04,
69
+ "grad_norm": 0.15385235740129172,
70
+ "learning_rate": 8.163265306122449e-05,
71
+ "loss": 0.8824,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.05,
76
+ "grad_norm": 0.13852135103025864,
77
+ "learning_rate": 9.183673469387756e-05,
78
+ "loss": 0.8615,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.05,
83
+ "grad_norm": 0.13036655403253458,
84
+ "learning_rate": 0.00010204081632653062,
85
+ "loss": 0.9202,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.06,
90
+ "grad_norm": 0.138300443551494,
91
+ "learning_rate": 0.00011224489795918367,
92
+ "loss": 0.8849,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.06,
97
+ "grad_norm": 0.12389157364724535,
98
+ "learning_rate": 0.00012244897959183676,
99
+ "loss": 0.862,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.07,
104
+ "grad_norm": 0.1429134980556313,
105
+ "learning_rate": 0.0001326530612244898,
106
+ "loss": 0.8463,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.07,
111
+ "grad_norm": 0.14541426965897125,
112
+ "learning_rate": 0.00014285714285714287,
113
+ "loss": 0.8289,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.08,
118
+ "grad_norm": 0.15560066242946408,
119
+ "learning_rate": 0.0001530612244897959,
120
+ "loss": 0.8285,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.08,
125
+ "grad_norm": 0.1382713303784162,
126
+ "learning_rate": 0.00016326530612244898,
127
+ "loss": 0.8267,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.09,
132
+ "grad_norm": 0.1367658702482613,
133
+ "learning_rate": 0.00017346938775510205,
134
+ "loss": 0.8875,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.09,
139
+ "grad_norm": 0.15684837460586795,
140
+ "learning_rate": 0.00018367346938775512,
141
+ "loss": 0.8758,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1,
146
+ "grad_norm": 0.13232147433779046,
147
+ "learning_rate": 0.00019387755102040816,
148
+ "loss": 0.8457,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.1,
153
+ "grad_norm": 0.12839583748095243,
154
+ "learning_rate": 0.0001999974277115551,
155
+ "loss": 0.8492,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.11,
160
+ "grad_norm": 0.1321027185844607,
161
+ "learning_rate": 0.00019996849098629418,
162
+ "loss": 0.787,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.11,
167
+ "grad_norm": 0.13562231873522493,
168
+ "learning_rate": 0.00019990741151022301,
169
+ "loss": 0.8034,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.12,
174
+ "grad_norm": 0.1872551577847746,
175
+ "learning_rate": 0.0001998142089221534,
176
+ "loss": 0.8897,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.12,
181
+ "grad_norm": 0.11987471778945524,
182
+ "learning_rate": 0.0001996889131894033,
183
+ "loss": 0.8548,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.13,
188
+ "grad_norm": 0.13758980449343652,
189
+ "learning_rate": 0.00019953156459816179,
190
+ "loss": 0.8126,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.13,
195
+ "grad_norm": 0.12651848093312767,
196
+ "learning_rate": 0.0001993422137405354,
197
+ "loss": 0.8606,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.14,
202
+ "grad_norm": 0.14411919368392379,
203
+ "learning_rate": 0.00019912092149828174,
204
+ "loss": 0.8858,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.14,
209
+ "grad_norm": 0.150512988037676,
210
+ "learning_rate": 0.00019886775902323405,
211
+ "loss": 0.8593,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.15,
216
+ "grad_norm": 0.15943254043396862,
217
+ "learning_rate": 0.00019858280771442385,
218
+ "loss": 0.8577,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.15,
223
+ "grad_norm": 0.1934721467912684,
224
+ "learning_rate": 0.00019826615919190887,
225
+ "loss": 0.8628,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.16,
230
+ "grad_norm": 0.1232771853573132,
231
+ "learning_rate": 0.00019791791526731445,
232
+ "loss": 0.84,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.16,
237
+ "grad_norm": 0.14669498826369695,
238
+ "learning_rate": 0.00019753818791109828,
239
+ "loss": 0.8185,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.17,
244
+ "grad_norm": 0.14085085759954724,
245
+ "learning_rate": 0.0001971270992165486,
246
+ "loss": 0.7795,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.17,
251
+ "grad_norm": 0.1335331389930546,
252
+ "learning_rate": 0.00019668478136052774,
253
+ "loss": 0.8159,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.18,
258
+ "grad_norm": 0.14301329004691762,
259
+ "learning_rate": 0.0001962113765609735,
260
+ "loss": 0.8008,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.18,
265
+ "grad_norm": 0.1349013012428684,
266
+ "learning_rate": 0.0001957070370311717,
267
+ "loss": 0.8523,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.19,
272
+ "grad_norm": 0.15345671624450827,
273
+ "learning_rate": 0.00019517192493081565,
274
+ "loss": 0.827,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.19,
279
+ "grad_norm": 0.13105575744786446,
280
+ "learning_rate": 0.00019460621231386676,
281
+ "loss": 0.8844,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.2,
286
+ "grad_norm": 0.14047662719322554,
287
+ "learning_rate": 0.00019401008107323455,
288
+ "loss": 0.8593,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.21,
293
+ "grad_norm": 0.16716795882975244,
294
+ "learning_rate": 0.0001933837228822925,
295
+ "loss": 0.8105,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.21,
300
+ "grad_norm": 0.14011283820384335,
301
+ "learning_rate": 0.0001927273391332499,
302
+ "loss": 0.8575,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.22,
307
+ "grad_norm": 0.13652402697672666,
308
+ "learning_rate": 0.00019204114087239806,
309
+ "loss": 0.8615,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.22,
314
+ "grad_norm": 0.1366384894487137,
315
+ "learning_rate": 0.00019132534873225323,
316
+ "loss": 0.8373,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.23,
321
+ "grad_norm": 0.1418034030520275,
322
+ "learning_rate": 0.00019058019286061665,
323
+ "loss": 0.8335,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.23,
328
+ "grad_norm": 0.12973981150967331,
329
+ "learning_rate": 0.00018980591284657535,
330
+ "loss": 0.8393,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.24,
335
+ "grad_norm": 0.14098955432684662,
336
+ "learning_rate": 0.00018900275764346768,
337
+ "loss": 0.8324,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.24,
342
+ "grad_norm": 0.1483857739019612,
343
+ "learning_rate": 0.0001881709854888372,
344
+ "loss": 0.8224,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.25,
349
+ "grad_norm": 0.1207564764356093,
350
+ "learning_rate": 0.00018731086382140226,
351
+ "loss": 0.8115,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.25,
356
+ "grad_norm": 0.147034450761397,
357
+ "learning_rate": 0.00018642266919506644,
358
+ "loss": 0.8473,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.26,
363
+ "grad_norm": 0.16090713319886477,
364
+ "learning_rate": 0.00018550668718999872,
365
+ "loss": 0.7954,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.26,
370
+ "grad_norm": 0.12088332198568695,
371
+ "learning_rate": 0.0001845632123208111,
372
+ "loss": 0.8553,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.27,
377
+ "grad_norm": 0.13281843244805125,
378
+ "learning_rate": 0.0001835925479418637,
379
+ "loss": 0.8602,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.27,
384
+ "grad_norm": 0.11395640254482223,
385
+ "learning_rate": 0.0001825950061497276,
386
+ "loss": 0.799,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.28,
391
+ "grad_norm": 0.18338212518218974,
392
+ "learning_rate": 0.00018157090768283678,
393
+ "loss": 0.818,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.28,
398
+ "grad_norm": 0.14005425053437323,
399
+ "learning_rate": 0.00018052058181836151,
400
+ "loss": 0.8221,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.29,
405
+ "grad_norm": 0.15372527780279158,
406
+ "learning_rate": 0.00017944436626633623,
407
+ "loss": 0.8417,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.29,
412
+ "grad_norm": 0.14434072633220638,
413
+ "learning_rate": 0.00017834260706107595,
414
+ "loss": 0.8108,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.3,
419
+ "grad_norm": 0.12627644118944992,
420
+ "learning_rate": 0.00017721565844991643,
421
+ "loss": 0.8068,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.3,
426
+ "grad_norm": 0.14772289738958688,
427
+ "learning_rate": 0.00017606388277931328,
428
+ "loss": 0.8744,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.31,
433
+ "grad_norm": 0.13984571980575156,
434
+ "learning_rate": 0.0001748876503783373,
435
+ "loss": 0.835,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.31,
440
+ "grad_norm": 0.13153373295044746,
441
+ "learning_rate": 0.00017368733943960276,
442
+ "loss": 0.8383,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.32,
447
+ "grad_norm": 0.1291787835295784,
448
+ "learning_rate": 0.00017246333589766787,
449
+ "loss": 0.8532,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.32,
454
+ "grad_norm": 0.1313754669175255,
455
+ "learning_rate": 0.00017121603330494544,
456
+ "loss": 0.8638,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.33,
461
+ "grad_norm": 0.14687623249066953,
462
+ "learning_rate": 0.0001699458327051647,
463
+ "loss": 0.8385,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.33,
468
+ "grad_norm": 0.14850598828123243,
469
+ "learning_rate": 0.00016865314250442398,
470
+ "loss": 0.8169,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.34,
475
+ "grad_norm": 0.14865425552332714,
476
+ "learning_rate": 0.00016733837833987633,
477
+ "loss": 0.8129,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.34,
482
+ "grad_norm": 0.14327447625468243,
483
+ "learning_rate": 0.00016600196294609045,
484
+ "loss": 0.8112,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.35,
489
+ "grad_norm": 0.1422668164659587,
490
+ "learning_rate": 0.00016464432601912912,
491
+ "loss": 0.8306,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.35,
496
+ "grad_norm": 0.12679383458857735,
497
+ "learning_rate": 0.0001632659040783897,
498
+ "loss": 0.8476,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.36,
503
+ "grad_norm": 0.13298721576358818,
504
+ "learning_rate": 0.00016186714032625035,
505
+ "loss": 0.8527,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.36,
510
+ "grad_norm": 0.15009708989503034,
511
+ "learning_rate": 0.00016044848450556787,
512
+ "loss": 0.819,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.37,
517
+ "grad_norm": 0.1535277202050242,
518
+ "learning_rate": 0.00015901039275507245,
519
+ "loss": 0.8305,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.37,
524
+ "grad_norm": 0.13563353212881551,
525
+ "learning_rate": 0.00015755332746270572,
526
+ "loss": 0.8104,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.38,
531
+ "grad_norm": 0.14208140877531353,
532
+ "learning_rate": 0.00015607775711694977,
533
+ "loss": 0.8381,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.38,
538
+ "grad_norm": 0.14084759655058215,
539
+ "learning_rate": 0.00015458415615619484,
540
+ "loss": 0.8186,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.39,
545
+ "grad_norm": 0.1202290236098903,
546
+ "learning_rate": 0.00015307300481619333,
547
+ "loss": 0.8578,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.4,
552
+ "grad_norm": 0.1427466626848493,
553
+ "learning_rate": 0.00015154478897565045,
554
+ "loss": 0.8816,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.4,
559
+ "grad_norm": 0.11439109448818731,
560
+ "learning_rate": 0.00015000000000000001,
561
+ "loss": 0.8142,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.41,
566
+ "grad_norm": 0.1402674024104851,
567
+ "learning_rate": 0.00014843913458341645,
568
+ "loss": 0.8011,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.41,
573
+ "grad_norm": 0.15497764154867305,
574
+ "learning_rate": 0.00014686269458911332,
575
+ "loss": 0.8179,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.42,
580
+ "grad_norm": 0.14725740205615323,
581
+ "learning_rate": 0.00014527118688797963,
582
+ "loss": 0.9044,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.42,
587
+ "grad_norm": 0.1441807780130091,
588
+ "learning_rate": 0.0001436651231956064,
589
+ "loss": 0.8295,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.43,
594
+ "grad_norm": 0.14314161398501463,
595
+ "learning_rate": 0.00014204501990775533,
596
+ "loss": 0.8471,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.43,
601
+ "grad_norm": 0.12813797598238083,
602
+ "learning_rate": 0.00014041139793432274,
603
+ "loss": 0.8262,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.44,
608
+ "grad_norm": 0.14355138774702908,
609
+ "learning_rate": 0.00013876478253185183,
610
+ "loss": 0.8544,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.44,
615
+ "grad_norm": 0.14065230597830594,
616
+ "learning_rate": 0.00013710570313464778,
617
+ "loss": 0.8273,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.45,
622
+ "grad_norm": 0.1502376541192836,
623
+ "learning_rate": 0.0001354346931845492,
624
+ "loss": 0.8629,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.45,
629
+ "grad_norm": 0.11612404557295304,
630
+ "learning_rate": 0.00013375228995941133,
631
+ "loss": 0.8474,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.46,
636
+ "grad_norm": 0.12394764980545889,
637
+ "learning_rate": 0.0001320590344003557,
638
+ "loss": 0.8398,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.46,
643
+ "grad_norm": 0.14219342684795552,
644
+ "learning_rate": 0.00013035547093784186,
645
+ "loss": 0.8386,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.47,
650
+ "grad_norm": 0.13096131144311657,
651
+ "learning_rate": 0.00012864214731661742,
652
+ "loss": 0.8175,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.47,
657
+ "grad_norm": 0.143251617809686,
658
+ "learning_rate": 0.00012691961441960238,
659
+ "loss": 0.8085,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.48,
664
+ "grad_norm": 0.12498048690896839,
665
+ "learning_rate": 0.00012518842609076413,
666
+ "loss": 0.8259,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.48,
671
+ "grad_norm": 0.14752810775417383,
672
+ "learning_rate": 0.00012344913895704097,
673
+ "loss": 0.8157,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.49,
678
+ "grad_norm": 0.13076168345663483,
679
+ "learning_rate": 0.00012170231224937032,
680
+ "loss": 0.8102,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.49,
685
+ "grad_norm": 0.14100703225317754,
686
+ "learning_rate": 0.00011994850762287989,
687
+ "loss": 0.8308,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.5,
692
+ "grad_norm": 0.1406207969502619,
693
+ "learning_rate": 0.0001181882889762994,
694
+ "loss": 0.8538,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.5,
699
+ "grad_norm": 0.12811621773423584,
700
+ "learning_rate": 0.00011642222227065089,
701
+ "loss": 0.8096,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.51,
706
+ "grad_norm": 0.16240280156036038,
707
+ "learning_rate": 0.00011465087534727587,
708
+ "loss": 0.893,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.51,
713
+ "grad_norm": 0.15744615260545805,
714
+ "learning_rate": 0.0001128748177452581,
715
+ "loss": 0.8271,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.52,
720
+ "grad_norm": 0.13453736557228535,
721
+ "learning_rate": 0.00011109462051830017,
722
+ "loss": 0.8932,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.52,
727
+ "grad_norm": 0.13488475951319145,
728
+ "learning_rate": 0.00010931085605111354,
729
+ "loss": 0.805,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.53,
734
+ "grad_norm": 0.13615523867504728,
735
+ "learning_rate": 0.00010752409787538,
736
+ "loss": 0.8358,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.53,
741
+ "grad_norm": 0.11608735125900543,
742
+ "learning_rate": 0.00010573492048534515,
743
+ "loss": 0.7792,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.54,
748
+ "grad_norm": 0.12529530307366807,
749
+ "learning_rate": 0.00010394389915310149,
750
+ "loss": 0.7858,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.54,
755
+ "grad_norm": 0.1477196524951594,
756
+ "learning_rate": 0.00010215160974362223,
757
+ "loss": 0.8949,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.55,
762
+ "grad_norm": 0.13478280062013237,
763
+ "learning_rate": 0.00010035862852960387,
764
+ "loss": 0.8675,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.55,
769
+ "grad_norm": 0.18751542446376887,
770
+ "learning_rate": 9.856553200617805e-05,
771
+ "loss": 0.8002,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.56,
776
+ "grad_norm": 0.13384075191607198,
777
+ "learning_rate": 9.677289670555169e-05,
778
+ "loss": 0.8342,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.56,
783
+ "grad_norm": 0.12830956280090938,
784
+ "learning_rate": 9.49812990116353e-05,
785
+ "loss": 0.8542,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.57,
790
+ "grad_norm": 0.13139317688343669,
791
+ "learning_rate": 9.319131497471894e-05,
792
+ "loss": 0.857,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.57,
797
+ "grad_norm": 0.11825939494831025,
798
+ "learning_rate": 9.140352012625537e-05,
799
+ "loss": 0.8286,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.58,
804
+ "grad_norm": 0.1319552841228723,
805
+ "learning_rate": 8.961848929381026e-05,
806
+ "loss": 0.8513,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.58,
811
+ "grad_norm": 0.13846831656086162,
812
+ "learning_rate": 8.783679641623845e-05,
813
+ "loss": 0.8575,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.59,
818
+ "grad_norm": 0.1565132050678128,
819
+ "learning_rate": 8.605901435914607e-05,
820
+ "loss": 0.8095,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.6,
825
+ "grad_norm": 0.1349070576898933,
826
+ "learning_rate": 8.428571473069775e-05,
827
+ "loss": 0.8364,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.6,
832
+ "grad_norm": 0.1478930655811248,
833
+ "learning_rate": 8.25174676978282e-05,
834
+ "loss": 0.7913,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.61,
839
+ "grad_norm": 0.14739768176324797,
840
+ "learning_rate": 8.075484180291701e-05,
841
+ "loss": 0.8095,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.61,
846
+ "grad_norm": 0.13061570270501272,
847
+ "learning_rate": 7.899840378098588e-05,
848
+ "loss": 0.8456,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.62,
853
+ "grad_norm": 0.1400435628778491,
854
+ "learning_rate": 7.724871837747707e-05,
855
+ "loss": 0.854,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.62,
860
+ "grad_norm": 0.14619473905537722,
861
+ "learning_rate": 7.550634816667142e-05,
862
+ "loss": 0.7878,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.63,
867
+ "grad_norm": 0.1669686546083547,
868
+ "learning_rate": 7.377185337080442e-05,
869
+ "loss": 0.8293,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.63,
874
+ "grad_norm": 0.15161889018158878,
875
+ "learning_rate": 7.204579167993881e-05,
876
+ "loss": 0.8327,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.64,
881
+ "grad_norm": 0.11802481456975004,
882
+ "learning_rate": 7.032871807265096e-05,
883
+ "loss": 0.8549,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.64,
888
+ "grad_norm": 0.1428032348359374,
889
+ "learning_rate": 6.862118463758943e-05,
890
+ "loss": 0.8113,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.65,
895
+ "grad_norm": 0.12867903817761486,
896
+ "learning_rate": 6.69237403959624e-05,
897
+ "loss": 0.8219,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.65,
902
+ "grad_norm": 0.14771825500860036,
903
+ "learning_rate": 6.52369311250116e-05,
904
+ "loss": 0.8042,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.66,
909
+ "grad_norm": 0.13258568806230905,
910
+ "learning_rate": 6.356129918252927e-05,
911
+ "loss": 0.8156,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.66,
916
+ "grad_norm": 0.14732793338159728,
917
+ "learning_rate": 6.189738333247432e-05,
918
+ "loss": 0.8121,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.67,
923
+ "grad_norm": 0.15966142035007472,
924
+ "learning_rate": 6.024571857174443e-05,
925
+ "loss": 0.8595,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.67,
930
+ "grad_norm": 0.13684066043985474,
931
+ "learning_rate": 5.860683595815893e-05,
932
+ "loss": 0.8109,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.68,
937
+ "grad_norm": 0.12490641045647621,
938
+ "learning_rate": 5.698126243970845e-05,
939
+ "loss": 0.8125,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.68,
944
+ "grad_norm": 0.13460863241729326,
945
+ "learning_rate": 5.536952068512608e-05,
946
+ "loss": 0.7947,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.69,
951
+ "grad_norm": 0.13010809900192385,
952
+ "learning_rate": 5.3772128915834184e-05,
953
+ "loss": 0.9054,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.69,
958
+ "grad_norm": 0.12951084999951157,
959
+ "learning_rate": 5.218960073932122e-05,
960
+ "loss": 0.8647,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.7,
965
+ "grad_norm": 0.12953418083912346,
966
+ "learning_rate": 5.062244498400228e-05,
967
+ "loss": 0.8474,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.7,
972
+ "grad_norm": 0.13349564170084294,
973
+ "learning_rate": 4.907116553561607e-05,
974
+ "loss": 0.8127,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.71,
979
+ "grad_norm": 0.11698335940312711,
980
+ "learning_rate": 4.753626117521103e-05,
981
+ "loss": 0.7862,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.71,
986
+ "grad_norm": 0.1417412770863787,
987
+ "learning_rate": 4.601822541877291e-05,
988
+ "loss": 0.7952,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.72,
993
+ "grad_norm": 0.1490752355077661,
994
+ "learning_rate": 4.451754635854517e-05,
995
+ "loss": 0.8458,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.72,
1000
+ "grad_norm": 0.1331101949433864,
1001
+ "learning_rate": 4.303470650609325e-05,
1002
+ "loss": 0.8044,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.73,
1007
+ "grad_norm": 0.13611270992724214,
1008
+ "learning_rate": 4.1570182637163155e-05,
1009
+ "loss": 0.8855,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.73,
1014
+ "grad_norm": 0.13998690568497368,
1015
+ "learning_rate": 4.0124445638384366e-05,
1016
+ "loss": 0.7553,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.74,
1021
+ "grad_norm": 0.14235394415159117,
1022
+ "learning_rate": 3.869796035586625e-05,
1023
+ "loss": 0.8048,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.74,
1028
+ "grad_norm": 0.13460415814884583,
1029
+ "learning_rate": 3.7291185445736444e-05,
1030
+ "loss": 0.808,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 0.75,
1035
+ "grad_norm": 0.13186297670947147,
1036
+ "learning_rate": 3.590457322666997e-05,
1037
+ "loss": 0.8512,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 0.75,
1042
+ "grad_norm": 0.13821443417948492,
1043
+ "learning_rate": 3.453856953445557e-05,
1044
+ "loss": 0.8445,
1045
+ "step": 735
1046
+ },
1047
+ {
1048
+ "epoch": 0.76,
1049
+ "grad_norm": 0.11419625295174221,
1050
+ "learning_rate": 3.319361357864663e-05,
1051
+ "loss": 0.8169,
1052
+ "step": 740
1053
+ },
1054
+ {
1055
+ "epoch": 0.76,
1056
+ "grad_norm": 0.12406541104741345,
1057
+ "learning_rate": 3.187013780134291e-05,
1058
+ "loss": 0.8428,
1059
+ "step": 745
1060
+ },
1061
+ {
1062
+ "epoch": 0.77,
1063
+ "grad_norm": 0.1286809514714421,
1064
+ "learning_rate": 3.05685677381475e-05,
1065
+ "loss": 0.85,
1066
+ "step": 750
1067
+ },
1068
+ {
1069
+ "epoch": 0.77,
1070
+ "grad_norm": 0.1445484766618607,
1071
+ "learning_rate": 2.9289321881345254e-05,
1072
+ "loss": 0.8106,
1073
+ "step": 755
1074
+ },
1075
+ {
1076
+ "epoch": 0.78,
1077
+ "grad_norm": 0.14262528146944597,
1078
+ "learning_rate": 2.8032811545345294e-05,
1079
+ "loss": 0.8609,
1080
+ "step": 760
1081
+ },
1082
+ {
1083
+ "epoch": 0.79,
1084
+ "grad_norm": 0.1769607359536476,
1085
+ "learning_rate": 2.679944073443158e-05,
1086
+ "loss": 0.8045,
1087
+ "step": 765
1088
+ },
1089
+ {
1090
+ "epoch": 0.79,
1091
+ "grad_norm": 0.14280821635903052,
1092
+ "learning_rate": 2.5589606012863963e-05,
1093
+ "loss": 0.866,
1094
+ "step": 770
1095
+ },
1096
+ {
1097
+ "epoch": 0.8,
1098
+ "grad_norm": 0.1429520090865704,
1099
+ "learning_rate": 2.4403696377371142e-05,
1100
+ "loss": 0.8667,
1101
+ "step": 775
1102
+ },
1103
+ {
1104
+ "epoch": 0.8,
1105
+ "grad_norm": 0.13554177395410102,
1106
+ "learning_rate": 2.324209313207736e-05,
1107
+ "loss": 0.8577,
1108
+ "step": 780
1109
+ },
1110
+ {
1111
+ "epoch": 0.81,
1112
+ "grad_norm": 0.1270863540245396,
1113
+ "learning_rate": 2.210516976590179e-05,
1114
+ "loss": 0.839,
1115
+ "step": 785
1116
+ },
1117
+ {
1118
+ "epoch": 0.81,
1119
+ "grad_norm": 0.15157612066040596,
1120
+ "learning_rate": 2.099329183247126e-05,
1121
+ "loss": 0.8193,
1122
+ "step": 790
1123
+ },
1124
+ {
1125
+ "epoch": 0.82,
1126
+ "grad_norm": 0.13563544257174368,
1127
+ "learning_rate": 1.9906816832584253e-05,
1128
+ "loss": 0.7841,
1129
+ "step": 795
1130
+ },
1131
+ {
1132
+ "epoch": 0.82,
1133
+ "grad_norm": 0.1438162068777128,
1134
+ "learning_rate": 1.8846094099263912e-05,
1135
+ "loss": 0.8317,
1136
+ "step": 800
1137
+ },
1138
+ {
1139
+ "epoch": 0.83,
1140
+ "grad_norm": 0.14098563226328448,
1141
+ "learning_rate": 1.781146468543765e-05,
1142
+ "loss": 0.7682,
1143
+ "step": 805
1144
+ },
1145
+ {
1146
+ "epoch": 0.83,
1147
+ "grad_norm": 0.12904056725011795,
1148
+ "learning_rate": 1.6803261254278636e-05,
1149
+ "loss": 0.8184,
1150
+ "step": 810
1151
+ },
1152
+ {
1153
+ "epoch": 0.84,
1154
+ "grad_norm": 0.1577908132849086,
1155
+ "learning_rate": 1.582180797224507e-05,
1156
+ "loss": 0.8208,
1157
+ "step": 815
1158
+ },
1159
+ {
1160
+ "epoch": 0.84,
1161
+ "grad_norm": 0.1448578052047199,
1162
+ "learning_rate": 1.4867420404851307e-05,
1163
+ "loss": 0.828,
1164
+ "step": 820
1165
+ },
1166
+ {
1167
+ "epoch": 0.85,
1168
+ "grad_norm": 0.11696975769424958,
1169
+ "learning_rate": 1.3940405415204416e-05,
1170
+ "loss": 0.8405,
1171
+ "step": 825
1172
+ },
1173
+ {
1174
+ "epoch": 0.85,
1175
+ "grad_norm": 0.12780753501046968,
1176
+ "learning_rate": 1.30410610653389e-05,
1177
+ "loss": 0.8647,
1178
+ "step": 830
1179
+ },
1180
+ {
1181
+ "epoch": 0.86,
1182
+ "grad_norm": 0.1401641424372156,
1183
+ "learning_rate": 1.2169676520381168e-05,
1184
+ "loss": 0.8346,
1185
+ "step": 835
1186
+ },
1187
+ {
1188
+ "epoch": 0.86,
1189
+ "grad_norm": 0.1507027680390559,
1190
+ "learning_rate": 1.1326531955574526e-05,
1191
+ "loss": 0.8532,
1192
+ "step": 840
1193
+ },
1194
+ {
1195
+ "epoch": 0.87,
1196
+ "grad_norm": 0.13417895219140907,
1197
+ "learning_rate": 1.0511898466194903e-05,
1198
+ "loss": 0.8092,
1199
+ "step": 845
1200
+ },
1201
+ {
1202
+ "epoch": 0.87,
1203
+ "grad_norm": 0.127954613918298,
1204
+ "learning_rate": 9.726037980385738e-06,
1205
+ "loss": 0.7773,
1206
+ "step": 850
1207
+ },
1208
+ {
1209
+ "epoch": 0.88,
1210
+ "grad_norm": 0.1366880841170061,
1211
+ "learning_rate": 8.969203174940654e-06,
1212
+ "loss": 0.8018,
1213
+ "step": 855
1214
+ },
1215
+ {
1216
+ "epoch": 0.88,
1217
+ "grad_norm": 0.11768872682875378,
1218
+ "learning_rate": 8.24163739406062e-06,
1219
+ "loss": 0.8483,
1220
+ "step": 860
1221
+ },
1222
+ {
1223
+ "epoch": 0.89,
1224
+ "grad_norm": 0.1419611383110849,
1225
+ "learning_rate": 7.543574571111655e-06,
1226
+ "loss": 0.844,
1227
+ "step": 865
1228
+ },
1229
+ {
1230
+ "epoch": 0.89,
1231
+ "grad_norm": 0.14220531250615795,
1232
+ "learning_rate": 6.875239153408542e-06,
1233
+ "loss": 0.7975,
1234
+ "step": 870
1235
+ },
1236
+ {
1237
+ "epoch": 0.9,
1238
+ "grad_norm": 0.1566277092103407,
1239
+ "learning_rate": 6.236846030048604e-06,
1240
+ "loss": 0.8063,
1241
+ "step": 875
1242
+ },
1243
+ {
1244
+ "epoch": 0.9,
1245
+ "grad_norm": 0.1268170098289612,
1246
+ "learning_rate": 5.6286004628186675e-06,
1247
+ "loss": 0.8484,
1248
+ "step": 880
1249
+ },
1250
+ {
1251
+ "epoch": 0.91,
1252
+ "grad_norm": 0.13964640969590117,
1253
+ "learning_rate": 5.0506980201973974e-06,
1254
+ "loss": 0.8284,
1255
+ "step": 885
1256
+ },
1257
+ {
1258
+ "epoch": 0.91,
1259
+ "grad_norm": 0.14682575127314904,
1260
+ "learning_rate": 4.503324514474483e-06,
1261
+ "loss": 0.8706,
1262
+ "step": 890
1263
+ },
1264
+ {
1265
+ "epoch": 0.92,
1266
+ "grad_norm": 0.13996242963540317,
1267
+ "learning_rate": 3.986655942006579e-06,
1268
+ "loss": 0.8293,
1269
+ "step": 895
1270
+ },
1271
+ {
1272
+ "epoch": 0.92,
1273
+ "grad_norm": 0.16776823076764466,
1274
+ "learning_rate": 3.5008584266294386e-06,
1275
+ "loss": 0.8063,
1276
+ "step": 900
1277
+ },
1278
+ {
1279
+ "epoch": 0.93,
1280
+ "grad_norm": 0.1510152152640099,
1281
+ "learning_rate": 3.0460881662442763e-06,
1282
+ "loss": 0.8718,
1283
+ "step": 905
1284
+ },
1285
+ {
1286
+ "epoch": 0.93,
1287
+ "grad_norm": 0.1366248738676145,
1288
+ "learning_rate": 2.622491382595693e-06,
1289
+ "loss": 0.8484,
1290
+ "step": 910
1291
+ },
1292
+ {
1293
+ "epoch": 0.94,
1294
+ "grad_norm": 0.1404750930152043,
1295
+ "learning_rate": 2.2302042742571193e-06,
1296
+ "loss": 0.8313,
1297
+ "step": 915
1298
+ },
1299
+ {
1300
+ "epoch": 0.94,
1301
+ "grad_norm": 0.1270118228269091,
1302
+ "learning_rate": 1.869352972839067e-06,
1303
+ "loss": 0.8117,
1304
+ "step": 920
1305
+ },
1306
+ {
1307
+ "epoch": 0.95,
1308
+ "grad_norm": 0.16190747206009376,
1309
+ "learning_rate": 1.5400535024342022e-06,
1310
+ "loss": 0.8031,
1311
+ "step": 925
1312
+ },
1313
+ {
1314
+ "epoch": 0.95,
1315
+ "grad_norm": 0.1529977790553585,
1316
+ "learning_rate": 1.2424117423122328e-06,
1317
+ "loss": 0.8316,
1318
+ "step": 930
1319
+ },
1320
+ {
1321
+ "epoch": 0.96,
1322
+ "grad_norm": 0.12937443064788462,
1323
+ "learning_rate": 9.765233928766493e-07,
1324
+ "loss": 0.7992,
1325
+ "step": 935
1326
+ },
1327
+ {
1328
+ "epoch": 0.96,
1329
+ "grad_norm": 0.1449373612680855,
1330
+ "learning_rate": 7.42473944894384e-07,
1331
+ "loss": 0.8107,
1332
+ "step": 940
1333
+ },
1334
+ {
1335
+ "epoch": 0.97,
1336
+ "grad_norm": 0.12996651495057476,
1337
+ "learning_rate": 5.403386520079323e-07,
1338
+ "loss": 0.7778,
1339
+ "step": 945
1340
+ },
1341
+ {
1342
+ "epoch": 0.97,
1343
+ "grad_norm": 0.12904475182773067,
1344
+ "learning_rate": 3.701825065392184e-07,
1345
+ "loss": 0.8515,
1346
+ "step": 950
1347
+ },
1348
+ {
1349
+ "epoch": 0.98,
1350
+ "grad_norm": 0.12597952420128744,
1351
+ "learning_rate": 2.320602185927001e-07,
1352
+ "loss": 0.7737,
1353
+ "step": 955
1354
+ },
1355
+ {
1356
+ "epoch": 0.99,
1357
+ "grad_norm": 0.1376254333598378,
1358
+ "learning_rate": 1.2601619846444035e-07,
1359
+ "loss": 0.8671,
1360
+ "step": 960
1361
+ },
1362
+ {
1363
+ "epoch": 0.99,
1364
+ "grad_norm": 0.13114870010510168,
1365
+ "learning_rate": 5.208454236296234e-08,
1366
+ "loss": 0.8289,
1367
+ "step": 965
1368
+ },
1369
+ {
1370
+ "epoch": 1.0,
1371
+ "grad_norm": 0.14712841990455128,
1372
+ "learning_rate": 1.0289021446308056e-08,
1373
+ "loss": 0.8719,
1374
+ "step": 970
1375
+ },
1376
+ {
1377
+ "epoch": 1.0,
1378
+ "eval_loss": NaN,
1379
+ "eval_runtime": 1840.0201,
1380
+ "eval_samples_per_second": 3.767,
1381
+ "eval_steps_per_second": 0.942,
1382
+ "step": 974
1383
+ },
1384
+ {
1385
+ "epoch": 1.0,
1386
+ "step": 974,
1387
+ "total_flos": 1.052863723732992e+16,
1388
+ "train_loss": 0.838743479834445,
1389
+ "train_runtime": 18938.875,
1390
+ "train_samples_per_second": 3.293,
1391
+ "train_steps_per_second": 0.051
1392
+ }
1393
+ ],
1394
+ "logging_steps": 5,
1395
+ "max_steps": 974,
1396
+ "num_input_tokens_seen": 0,
1397
+ "num_train_epochs": 1,
1398
+ "save_steps": 100,
1399
+ "total_flos": 1.052863723732992e+16,
1400
+ "train_batch_size": 8,
1401
+ "trial_name": null,
1402
+ "trial_params": null
1403
+ }