antonpolishko commited on
Commit
73d0ec4
·
verified ·
1 Parent(s): 0f6f19c

Model save

Browse files
README.md ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: google/gemma-2-9b
3
+ library_name: transformers
4
+ model_name: gemma-2-9b-tok20k-overfit-ua
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for gemma-2-9b-tok20k-overfit-ua
13
+
14
+ This model is a fine-tuned version of [google/gemma-2-9b](https://huggingface.co/google/gemma-2-9b).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="antonpolishko/gemma-2-9b-tok20k-overfit-ua", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/polyagent/huggingface/runs/26qriwow)
31
+
32
+ This model was trained with SFT.
33
+
34
+ ### Framework versions
35
+
36
+ - TRL: 0.12.1
37
+ - Transformers: 4.46.3
38
+ - Pytorch: 2.6.0.dev20241113+rocm6.2
39
+ - Datasets: 3.1.0
40
+ - Tokenizers: 0.20.3
41
+
42
+ ## Citations
43
+
44
+
45
+
46
+ Cite TRL as:
47
+
48
+ ```bibtex
49
+ @misc{vonwerra2022trl,
50
+ title = {{TRL: Transformer Reinforcement Learning}},
51
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
52
+ year = 2020,
53
+ journal = {GitHub repository},
54
+ publisher = {GitHub},
55
+ howpublished = {\url{https://github.com/huggingface/trl}}
56
+ }
57
+ ```
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 5.361516984661967e+18,
4
+ "train_loss": 5.0057218712328115,
5
+ "train_runtime": 5286.5666,
6
+ "train_samples": 95663,
7
+ "train_samples_per_second": 9.883,
8
+ "train_steps_per_second": 0.155
9
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "cache_implementation": "hybrid",
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.46.3"
8
+ }
runs/Dec26_11-16-33_mia1-gpu-109/events.out.tfevents.1735211862.mia1-gpu-109.3674296.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1f3af49799c4ca2df9032060c66fc6ea44dc79b9a4ab6735d3bbaf85f824d21c
3
- size 41013
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c730283cffe2a7edecc3604cfab235b1b4a8706e3b129a763da5f97d97586b2
3
+ size 41367
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 5.361516984661967e+18,
4
+ "train_loss": 5.0057218712328115,
5
+ "train_runtime": 5286.5666,
6
+ "train_samples": 95663,
7
+ "train_samples_per_second": 9.883,
8
+ "train_steps_per_second": 0.155
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1206 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 300,
6
+ "global_step": 819,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.003663003663003663,
13
+ "grad_norm": 165.0,
14
+ "learning_rate": 2e-06,
15
+ "loss": 9.1083,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.018315018315018316,
20
+ "grad_norm": 64.0,
21
+ "learning_rate": 1.9998820020169668e-06,
22
+ "loss": 8.3175,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03663003663003663,
27
+ "grad_norm": 26.875,
28
+ "learning_rate": 1.999402682936637e-06,
29
+ "loss": 7.3293,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.054945054945054944,
34
+ "grad_norm": 16.75,
35
+ "learning_rate": 1.998554844493029e-06,
36
+ "loss": 6.8602,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.07326007326007326,
41
+ "grad_norm": 11.0,
42
+ "learning_rate": 1.997338799317767e-06,
43
+ "loss": 6.6054,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.09157509157509157,
48
+ "grad_norm": 8.4375,
49
+ "learning_rate": 1.995754995814884e-06,
50
+ "loss": 6.3957,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.10989010989010989,
55
+ "grad_norm": 6.9375,
56
+ "learning_rate": 1.9938040179954784e-06,
57
+ "loss": 6.256,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.1282051282051282,
62
+ "grad_norm": 6.15625,
63
+ "learning_rate": 1.991486585262365e-06,
64
+ "loss": 6.1321,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.14652014652014653,
69
+ "grad_norm": 5.1875,
70
+ "learning_rate": 1.988803552144804e-06,
71
+ "loss": 6.018,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.16483516483516483,
76
+ "grad_norm": 5.59375,
77
+ "learning_rate": 1.9857559079834022e-06,
78
+ "loss": 5.9438,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.18315018315018314,
83
+ "grad_norm": 6.5,
84
+ "learning_rate": 1.982344776565302e-06,
85
+ "loss": 5.8733,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.20146520146520147,
90
+ "grad_norm": 4.84375,
91
+ "learning_rate": 1.978571415709799e-06,
92
+ "loss": 5.8341,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.21978021978021978,
97
+ "grad_norm": 4.125,
98
+ "learning_rate": 1.9744372168045322e-06,
99
+ "loss": 5.793,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.23809523809523808,
104
+ "grad_norm": 3.8125,
105
+ "learning_rate": 1.9699437042924264e-06,
106
+ "loss": 5.7305,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.2564102564102564,
111
+ "grad_norm": 4.0,
112
+ "learning_rate": 1.965092535109567e-06,
113
+ "loss": 5.671,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.27472527472527475,
118
+ "grad_norm": 4.125,
119
+ "learning_rate": 1.959885498074224e-06,
120
+ "loss": 5.6563,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.29304029304029305,
125
+ "grad_norm": 3.765625,
126
+ "learning_rate": 1.954324513227244e-06,
127
+ "loss": 5.5778,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.31135531135531136,
132
+ "grad_norm": 3.75,
133
+ "learning_rate": 1.948411631124053e-06,
134
+ "loss": 5.5838,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.32967032967032966,
139
+ "grad_norm": 6.71875,
140
+ "learning_rate": 1.942149032078538e-06,
141
+ "loss": 5.5305,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.34798534798534797,
146
+ "grad_norm": 4.6875,
147
+ "learning_rate": 1.935539025359077e-06,
148
+ "loss": 5.5108,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.3663003663003663,
153
+ "grad_norm": 3.546875,
154
+ "learning_rate": 1.928584048337022e-06,
155
+ "loss": 5.494,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.38461538461538464,
160
+ "grad_norm": 3.75,
161
+ "learning_rate": 1.9212866655879395e-06,
162
+ "loss": 5.4749,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.40293040293040294,
167
+ "grad_norm": 4.5,
168
+ "learning_rate": 1.9136495679459563e-06,
169
+ "loss": 5.4473,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.42124542124542125,
174
+ "grad_norm": 4.34375,
175
+ "learning_rate": 1.9056755715115372e-06,
176
+ "loss": 5.4071,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.43956043956043955,
181
+ "grad_norm": 3.609375,
182
+ "learning_rate": 1.8973676166130791e-06,
183
+ "loss": 5.3994,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.45787545787545786,
188
+ "grad_norm": 3.859375,
189
+ "learning_rate": 1.8887287667226963e-06,
190
+ "loss": 5.3773,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.47619047619047616,
195
+ "grad_norm": 3.4375,
196
+ "learning_rate": 1.8797622073265943e-06,
197
+ "loss": 5.3734,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.4945054945054945,
202
+ "grad_norm": 3.8125,
203
+ "learning_rate": 1.8704712447504579e-06,
204
+ "loss": 5.323,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.5128205128205128,
209
+ "grad_norm": 3.4375,
210
+ "learning_rate": 1.8608593049402752e-06,
211
+ "loss": 5.2964,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.5311355311355311,
216
+ "grad_norm": 3.484375,
217
+ "learning_rate": 1.850929932199058e-06,
218
+ "loss": 5.2951,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.5494505494505495,
223
+ "grad_norm": 3.375,
224
+ "learning_rate": 1.8406867878799152e-06,
225
+ "loss": 5.2498,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.5677655677655677,
230
+ "grad_norm": 3.859375,
231
+ "learning_rate": 1.8301336490359678e-06,
232
+ "loss": 5.2432,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.5860805860805861,
237
+ "grad_norm": 3.90625,
238
+ "learning_rate": 1.819274407027599e-06,
239
+ "loss": 5.2436,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.6043956043956044,
244
+ "grad_norm": 4.6875,
245
+ "learning_rate": 1.8081130660875555e-06,
246
+ "loss": 5.2218,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.6227106227106227,
251
+ "grad_norm": 3.5625,
252
+ "learning_rate": 1.79665374184443e-06,
253
+ "loss": 5.2194,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.6410256410256411,
258
+ "grad_norm": 3.546875,
259
+ "learning_rate": 1.7849006598050625e-06,
260
+ "loss": 5.1988,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.6593406593406593,
265
+ "grad_norm": 3.9375,
266
+ "learning_rate": 1.7728581537964318e-06,
267
+ "loss": 5.1814,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.6776556776556777,
272
+ "grad_norm": 4.40625,
273
+ "learning_rate": 1.7605306643676006e-06,
274
+ "loss": 5.1318,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.6959706959706959,
279
+ "grad_norm": 3.4375,
280
+ "learning_rate": 1.747922737152308e-06,
281
+ "loss": 5.1609,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.7142857142857143,
286
+ "grad_norm": 4.125,
287
+ "learning_rate": 1.7350390211928166e-06,
288
+ "loss": 5.1395,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.7326007326007326,
293
+ "grad_norm": 3.375,
294
+ "learning_rate": 1.721884267225624e-06,
295
+ "loss": 5.0794,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.7509157509157509,
300
+ "grad_norm": 3.421875,
301
+ "learning_rate": 1.7084633259296795e-06,
302
+ "loss": 5.0974,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.7692307692307693,
307
+ "grad_norm": 3.46875,
308
+ "learning_rate": 1.6947811461377467e-06,
309
+ "loss": 5.1007,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.7875457875457875,
314
+ "grad_norm": 3.984375,
315
+ "learning_rate": 1.6808427730115712e-06,
316
+ "loss": 5.0777,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.8058608058608059,
321
+ "grad_norm": 3.828125,
322
+ "learning_rate": 1.6666533461815323e-06,
323
+ "loss": 5.0517,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.8241758241758241,
328
+ "grad_norm": 3.546875,
329
+ "learning_rate": 1.6522180978514552e-06,
330
+ "loss": 5.0462,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.8424908424908425,
335
+ "grad_norm": 3.78125,
336
+ "learning_rate": 1.6375423508692912e-06,
337
+ "loss": 5.0538,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.8608058608058609,
342
+ "grad_norm": 3.515625,
343
+ "learning_rate": 1.622631516764372e-06,
344
+ "loss": 5.0335,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.8791208791208791,
349
+ "grad_norm": 3.359375,
350
+ "learning_rate": 1.607491093751966e-06,
351
+ "loss": 5.0177,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.8974358974358975,
356
+ "grad_norm": 4.8125,
357
+ "learning_rate": 1.592126664705868e-06,
358
+ "loss": 4.998,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.9157509157509157,
363
+ "grad_norm": 3.671875,
364
+ "learning_rate": 1.5765438950997703e-06,
365
+ "loss": 4.992,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.9340659340659341,
370
+ "grad_norm": 3.828125,
371
+ "learning_rate": 1.5607485309181812e-06,
372
+ "loss": 4.9831,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.9523809523809523,
377
+ "grad_norm": 3.90625,
378
+ "learning_rate": 1.544746396537651e-06,
379
+ "loss": 4.9835,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.9706959706959707,
384
+ "grad_norm": 3.484375,
385
+ "learning_rate": 1.5285433925790945e-06,
386
+ "loss": 4.962,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.989010989010989,
391
+ "grad_norm": 3.421875,
392
+ "learning_rate": 1.5121454937319975e-06,
393
+ "loss": 4.9708,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 1.0073260073260073,
398
+ "grad_norm": 3.9375,
399
+ "learning_rate": 1.4955587465513128e-06,
400
+ "loss": 4.9494,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 1.0256410256410255,
405
+ "grad_norm": 3.703125,
406
+ "learning_rate": 1.4787892672278555e-06,
407
+ "loss": 4.9172,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 1.043956043956044,
412
+ "grad_norm": 3.515625,
413
+ "learning_rate": 1.461843239333021e-06,
414
+ "loss": 4.9262,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 1.0622710622710623,
419
+ "grad_norm": 3.75,
420
+ "learning_rate": 1.444726911538657e-06,
421
+ "loss": 4.9115,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 1.0805860805860805,
426
+ "grad_norm": 3.671875,
427
+ "learning_rate": 1.4274465953129325e-06,
428
+ "loss": 4.8926,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 1.098901098901099,
433
+ "grad_norm": 3.625,
434
+ "learning_rate": 1.4100086625930462e-06,
435
+ "loss": 4.8845,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 1.098901098901099,
440
+ "eval_loss": 4.895308017730713,
441
+ "eval_runtime": 14.6132,
442
+ "eval_samples_per_second": 36.2,
443
+ "eval_steps_per_second": 1.163,
444
+ "step": 300
445
+ },
446
+ {
447
+ "epoch": 1.1172161172161172,
448
+ "grad_norm": 3.5,
449
+ "learning_rate": 1.3924195434356441e-06,
450
+ "loss": 4.8754,
451
+ "step": 305
452
+ },
453
+ {
454
+ "epoch": 1.1355311355311355,
455
+ "grad_norm": 3.484375,
456
+ "learning_rate": 1.3746857236458005e-06,
457
+ "loss": 4.8566,
458
+ "step": 310
459
+ },
460
+ {
461
+ "epoch": 1.1538461538461537,
462
+ "grad_norm": 3.875,
463
+ "learning_rate": 1.3568137423854457e-06,
464
+ "loss": 4.873,
465
+ "step": 315
466
+ },
467
+ {
468
+ "epoch": 1.1721611721611722,
469
+ "grad_norm": 3.515625,
470
+ "learning_rate": 1.3388101897621183e-06,
471
+ "loss": 4.8476,
472
+ "step": 320
473
+ },
474
+ {
475
+ "epoch": 1.1904761904761905,
476
+ "grad_norm": 4.0,
477
+ "learning_rate": 1.32068170439893e-06,
478
+ "loss": 4.8846,
479
+ "step": 325
480
+ },
481
+ {
482
+ "epoch": 1.2087912087912087,
483
+ "grad_norm": 3.453125,
484
+ "learning_rate": 1.3024349709866448e-06,
485
+ "loss": 4.864,
486
+ "step": 330
487
+ },
488
+ {
489
+ "epoch": 1.2271062271062272,
490
+ "grad_norm": 5.125,
491
+ "learning_rate": 1.2840767178187654e-06,
492
+ "loss": 4.8457,
493
+ "step": 335
494
+ },
495
+ {
496
+ "epoch": 1.2454212454212454,
497
+ "grad_norm": 3.828125,
498
+ "learning_rate": 1.265613714310548e-06,
499
+ "loss": 4.8469,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 1.2637362637362637,
504
+ "grad_norm": 3.59375,
505
+ "learning_rate": 1.2470527685028482e-06,
506
+ "loss": 4.8406,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 1.282051282051282,
511
+ "grad_norm": 3.609375,
512
+ "learning_rate": 1.228400724551728e-06,
513
+ "loss": 4.8545,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 1.3003663003663004,
518
+ "grad_norm": 3.578125,
519
+ "learning_rate": 1.2096644602047445e-06,
520
+ "loss": 4.8236,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 1.3186813186813187,
525
+ "grad_norm": 3.515625,
526
+ "learning_rate": 1.1908508842648505e-06,
527
+ "loss": 4.8126,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 1.3369963369963371,
532
+ "grad_norm": 3.453125,
533
+ "learning_rate": 1.171966934042847e-06,
534
+ "loss": 4.8038,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 1.3553113553113554,
539
+ "grad_norm": 3.484375,
540
+ "learning_rate": 1.1530195727993199e-06,
541
+ "loss": 4.8278,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 1.3736263736263736,
546
+ "grad_norm": 3.53125,
547
+ "learning_rate": 1.1340157871770115e-06,
548
+ "loss": 4.8217,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 1.3919413919413919,
553
+ "grad_norm": 3.6875,
554
+ "learning_rate": 1.1149625846245681e-06,
555
+ "loss": 4.7912,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 1.4102564102564101,
560
+ "grad_norm": 3.515625,
561
+ "learning_rate": 1.095866990812615e-06,
562
+ "loss": 4.8004,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 1.4285714285714286,
567
+ "grad_norm": 3.5625,
568
+ "learning_rate": 1.0767360470431157e-06,
569
+ "loss": 4.7896,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 1.4468864468864469,
574
+ "grad_norm": 3.625,
575
+ "learning_rate": 1.0575768076529625e-06,
576
+ "loss": 4.7788,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 1.4652014652014653,
581
+ "grad_norm": 3.640625,
582
+ "learning_rate": 1.0383963374127645e-06,
583
+ "loss": 4.8106,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 1.4835164835164836,
588
+ "grad_norm": 3.71875,
589
+ "learning_rate": 1.0192017089217861e-06,
590
+ "loss": 4.7749,
591
+ "step": 405
592
+ },
593
+ {
594
+ "epoch": 1.5018315018315018,
595
+ "grad_norm": 3.484375,
596
+ "learning_rate": 1e-06,
597
+ "loss": 4.7864,
598
+ "step": 410
599
+ },
600
+ {
601
+ "epoch": 1.52014652014652,
602
+ "grad_norm": 3.59375,
603
+ "learning_rate": 9.80798291078214e-07,
604
+ "loss": 4.7869,
605
+ "step": 415
606
+ },
607
+ {
608
+ "epoch": 1.5384615384615383,
609
+ "grad_norm": 3.625,
610
+ "learning_rate": 9.616036625872356e-07,
611
+ "loss": 4.8314,
612
+ "step": 420
613
+ },
614
+ {
615
+ "epoch": 1.5567765567765568,
616
+ "grad_norm": 3.734375,
617
+ "learning_rate": 9.424231923470376e-07,
618
+ "loss": 4.7831,
619
+ "step": 425
620
+ },
621
+ {
622
+ "epoch": 1.575091575091575,
623
+ "grad_norm": 3.578125,
624
+ "learning_rate": 9.232639529568842e-07,
625
+ "loss": 4.7636,
626
+ "step": 430
627
+ },
628
+ {
629
+ "epoch": 1.5934065934065935,
630
+ "grad_norm": 3.71875,
631
+ "learning_rate": 9.041330091873851e-07,
632
+ "loss": 4.7465,
633
+ "step": 435
634
+ },
635
+ {
636
+ "epoch": 1.6117216117216118,
637
+ "grad_norm": 3.546875,
638
+ "learning_rate": 8.850374153754321e-07,
639
+ "loss": 4.7616,
640
+ "step": 440
641
+ },
642
+ {
643
+ "epoch": 1.63003663003663,
644
+ "grad_norm": 3.46875,
645
+ "learning_rate": 8.659842128229886e-07,
646
+ "loss": 4.7542,
647
+ "step": 445
648
+ },
649
+ {
650
+ "epoch": 1.6483516483516483,
651
+ "grad_norm": 3.5,
652
+ "learning_rate": 8.4698042720068e-07,
653
+ "loss": 4.745,
654
+ "step": 450
655
+ },
656
+ {
657
+ "epoch": 1.6666666666666665,
658
+ "grad_norm": 3.921875,
659
+ "learning_rate": 8.280330659571531e-07,
660
+ "loss": 4.738,
661
+ "step": 455
662
+ },
663
+ {
664
+ "epoch": 1.684981684981685,
665
+ "grad_norm": 3.75,
666
+ "learning_rate": 8.091491157351493e-07,
667
+ "loss": 4.7335,
668
+ "step": 460
669
+ },
670
+ {
671
+ "epoch": 1.7032967032967035,
672
+ "grad_norm": 3.546875,
673
+ "learning_rate": 7.903355397952556e-07,
674
+ "loss": 4.7792,
675
+ "step": 465
676
+ },
677
+ {
678
+ "epoch": 1.7216117216117217,
679
+ "grad_norm": 3.546875,
680
+ "learning_rate": 7.715992754482718e-07,
681
+ "loss": 4.7288,
682
+ "step": 470
683
+ },
684
+ {
685
+ "epoch": 1.73992673992674,
686
+ "grad_norm": 3.515625,
687
+ "learning_rate": 7.529472314971522e-07,
688
+ "loss": 4.7489,
689
+ "step": 475
690
+ },
691
+ {
692
+ "epoch": 1.7582417582417582,
693
+ "grad_norm": 3.484375,
694
+ "learning_rate": 7.34386285689452e-07,
695
+ "loss": 4.7163,
696
+ "step": 480
697
+ },
698
+ {
699
+ "epoch": 1.7765567765567765,
700
+ "grad_norm": 3.828125,
701
+ "learning_rate": 7.159232821812347e-07,
702
+ "loss": 4.7201,
703
+ "step": 485
704
+ },
705
+ {
706
+ "epoch": 1.7948717948717947,
707
+ "grad_norm": 3.5625,
708
+ "learning_rate": 6.975650290133554e-07,
709
+ "loss": 4.7334,
710
+ "step": 490
711
+ },
712
+ {
713
+ "epoch": 1.8131868131868132,
714
+ "grad_norm": 3.609375,
715
+ "learning_rate": 6.793182956010699e-07,
716
+ "loss": 4.7339,
717
+ "step": 495
718
+ },
719
+ {
720
+ "epoch": 1.8315018315018317,
721
+ "grad_norm": 3.484375,
722
+ "learning_rate": 6.611898102378818e-07,
723
+ "loss": 4.7384,
724
+ "step": 500
725
+ },
726
+ {
727
+ "epoch": 1.84981684981685,
728
+ "grad_norm": 3.546875,
729
+ "learning_rate": 6.431862576145544e-07,
730
+ "loss": 4.7402,
731
+ "step": 505
732
+ },
733
+ {
734
+ "epoch": 1.8681318681318682,
735
+ "grad_norm": 3.703125,
736
+ "learning_rate": 6.253142763541995e-07,
737
+ "loss": 4.723,
738
+ "step": 510
739
+ },
740
+ {
741
+ "epoch": 1.8864468864468864,
742
+ "grad_norm": 3.5,
743
+ "learning_rate": 6.075804565643561e-07,
744
+ "loss": 4.7457,
745
+ "step": 515
746
+ },
747
+ {
748
+ "epoch": 1.9047619047619047,
749
+ "grad_norm": 3.5625,
750
+ "learning_rate": 5.899913374069538e-07,
751
+ "loss": 4.712,
752
+ "step": 520
753
+ },
754
+ {
755
+ "epoch": 1.9230769230769231,
756
+ "grad_norm": 3.703125,
757
+ "learning_rate": 5.725534046870677e-07,
758
+ "loss": 4.7273,
759
+ "step": 525
760
+ },
761
+ {
762
+ "epoch": 1.9413919413919414,
763
+ "grad_norm": 3.5625,
764
+ "learning_rate": 5.552730884613428e-07,
765
+ "loss": 4.6981,
766
+ "step": 530
767
+ },
768
+ {
769
+ "epoch": 1.9597069597069599,
770
+ "grad_norm": 3.609375,
771
+ "learning_rate": 5.381567606669793e-07,
772
+ "loss": 4.7482,
773
+ "step": 535
774
+ },
775
+ {
776
+ "epoch": 1.978021978021978,
777
+ "grad_norm": 3.765625,
778
+ "learning_rate": 5.212107327721445e-07,
779
+ "loss": 4.718,
780
+ "step": 540
781
+ },
782
+ {
783
+ "epoch": 1.9963369963369964,
784
+ "grad_norm": 3.6875,
785
+ "learning_rate": 5.044412534486873e-07,
786
+ "loss": 4.7148,
787
+ "step": 545
788
+ },
789
+ {
790
+ "epoch": 2.0146520146520146,
791
+ "grad_norm": 3.71875,
792
+ "learning_rate": 4.878545062680026e-07,
793
+ "loss": 4.6767,
794
+ "step": 550
795
+ },
796
+ {
797
+ "epoch": 2.032967032967033,
798
+ "grad_norm": 3.453125,
799
+ "learning_rate": 4.7145660742090575e-07,
800
+ "loss": 4.706,
801
+ "step": 555
802
+ },
803
+ {
804
+ "epoch": 2.051282051282051,
805
+ "grad_norm": 4.0,
806
+ "learning_rate": 4.5525360346234907e-07,
807
+ "loss": 4.7142,
808
+ "step": 560
809
+ },
810
+ {
811
+ "epoch": 2.06959706959707,
812
+ "grad_norm": 3.546875,
813
+ "learning_rate": 4.392514690818193e-07,
814
+ "loss": 4.7053,
815
+ "step": 565
816
+ },
817
+ {
818
+ "epoch": 2.087912087912088,
819
+ "grad_norm": 3.765625,
820
+ "learning_rate": 4.2345610490022996e-07,
821
+ "loss": 4.6978,
822
+ "step": 570
823
+ },
824
+ {
825
+ "epoch": 2.1062271062271063,
826
+ "grad_norm": 3.53125,
827
+ "learning_rate": 4.078733352941321e-07,
828
+ "loss": 4.7147,
829
+ "step": 575
830
+ },
831
+ {
832
+ "epoch": 2.1245421245421245,
833
+ "grad_norm": 3.546875,
834
+ "learning_rate": 3.925089062480339e-07,
835
+ "loss": 4.6861,
836
+ "step": 580
837
+ },
838
+ {
839
+ "epoch": 2.142857142857143,
840
+ "grad_norm": 3.703125,
841
+ "learning_rate": 3.77368483235628e-07,
842
+ "loss": 4.725,
843
+ "step": 585
844
+ },
845
+ {
846
+ "epoch": 2.161172161172161,
847
+ "grad_norm": 3.453125,
848
+ "learning_rate": 3.6245764913070875e-07,
849
+ "loss": 4.7223,
850
+ "step": 590
851
+ },
852
+ {
853
+ "epoch": 2.1794871794871793,
854
+ "grad_norm": 3.578125,
855
+ "learning_rate": 3.477819021485447e-07,
856
+ "loss": 4.7254,
857
+ "step": 595
858
+ },
859
+ {
860
+ "epoch": 2.197802197802198,
861
+ "grad_norm": 3.578125,
862
+ "learning_rate": 3.333466538184674e-07,
863
+ "loss": 4.7074,
864
+ "step": 600
865
+ },
866
+ {
867
+ "epoch": 2.197802197802198,
868
+ "eval_loss": 4.718916893005371,
869
+ "eval_runtime": 14.5971,
870
+ "eval_samples_per_second": 36.24,
871
+ "eval_steps_per_second": 1.165,
872
+ "step": 600
873
+ },
874
+ {
875
+ "epoch": 2.2161172161172162,
876
+ "grad_norm": 3.546875,
877
+ "learning_rate": 3.1915722698842874e-07,
878
+ "loss": 4.705,
879
+ "step": 605
880
+ },
881
+ {
882
+ "epoch": 2.2344322344322345,
883
+ "grad_norm": 3.734375,
884
+ "learning_rate": 3.0521885386225344e-07,
885
+ "loss": 4.7118,
886
+ "step": 610
887
+ },
888
+ {
889
+ "epoch": 2.2527472527472527,
890
+ "grad_norm": 3.703125,
891
+ "learning_rate": 2.9153667407032066e-07,
892
+ "loss": 4.7285,
893
+ "step": 615
894
+ },
895
+ {
896
+ "epoch": 2.271062271062271,
897
+ "grad_norm": 3.59375,
898
+ "learning_rate": 2.7811573277437603e-07,
899
+ "loss": 4.681,
900
+ "step": 620
901
+ },
902
+ {
903
+ "epoch": 2.2893772893772892,
904
+ "grad_norm": 3.546875,
905
+ "learning_rate": 2.649609788071836e-07,
906
+ "loss": 4.707,
907
+ "step": 625
908
+ },
909
+ {
910
+ "epoch": 2.3076923076923075,
911
+ "grad_norm": 3.46875,
912
+ "learning_rate": 2.520772628476919e-07,
913
+ "loss": 4.7278,
914
+ "step": 630
915
+ },
916
+ {
917
+ "epoch": 2.326007326007326,
918
+ "grad_norm": 3.59375,
919
+ "learning_rate": 2.394693356323997e-07,
920
+ "loss": 4.7202,
921
+ "step": 635
922
+ },
923
+ {
924
+ "epoch": 2.3443223443223444,
925
+ "grad_norm": 3.609375,
926
+ "learning_rate": 2.2714184620356826e-07,
927
+ "loss": 4.706,
928
+ "step": 640
929
+ },
930
+ {
931
+ "epoch": 2.3626373626373627,
932
+ "grad_norm": 3.5,
933
+ "learning_rate": 2.150993401949376e-07,
934
+ "loss": 4.732,
935
+ "step": 645
936
+ },
937
+ {
938
+ "epoch": 2.380952380952381,
939
+ "grad_norm": 4.46875,
940
+ "learning_rate": 2.0334625815557026e-07,
941
+ "loss": 4.6981,
942
+ "step": 650
943
+ },
944
+ {
945
+ "epoch": 2.399267399267399,
946
+ "grad_norm": 3.5625,
947
+ "learning_rate": 1.9188693391244438e-07,
948
+ "loss": 4.7334,
949
+ "step": 655
950
+ },
951
+ {
952
+ "epoch": 2.4175824175824174,
953
+ "grad_norm": 3.609375,
954
+ "learning_rate": 1.8072559297240097e-07,
955
+ "loss": 4.7284,
956
+ "step": 660
957
+ },
958
+ {
959
+ "epoch": 2.435897435897436,
960
+ "grad_norm": 3.515625,
961
+ "learning_rate": 1.6986635096403212e-07,
962
+ "loss": 4.7189,
963
+ "step": 665
964
+ },
965
+ {
966
+ "epoch": 2.4542124542124544,
967
+ "grad_norm": 3.484375,
968
+ "learning_rate": 1.5931321212008465e-07,
969
+ "loss": 4.7007,
970
+ "step": 670
971
+ },
972
+ {
973
+ "epoch": 2.4725274725274726,
974
+ "grad_norm": 3.53125,
975
+ "learning_rate": 1.490700678009421e-07,
976
+ "loss": 4.6923,
977
+ "step": 675
978
+ },
979
+ {
980
+ "epoch": 2.490842490842491,
981
+ "grad_norm": 3.546875,
982
+ "learning_rate": 1.3914069505972482e-07,
983
+ "loss": 4.7265,
984
+ "step": 680
985
+ },
986
+ {
987
+ "epoch": 2.509157509157509,
988
+ "grad_norm": 3.6875,
989
+ "learning_rate": 1.2952875524954232e-07,
990
+ "loss": 4.7318,
991
+ "step": 685
992
+ },
993
+ {
994
+ "epoch": 2.5274725274725274,
995
+ "grad_norm": 3.546875,
996
+ "learning_rate": 1.2023779267340563e-07,
997
+ "loss": 4.7144,
998
+ "step": 690
999
+ },
1000
+ {
1001
+ "epoch": 2.5457875457875456,
1002
+ "grad_norm": 3.609375,
1003
+ "learning_rate": 1.112712332773038e-07,
1004
+ "loss": 4.6979,
1005
+ "step": 695
1006
+ },
1007
+ {
1008
+ "epoch": 2.564102564102564,
1009
+ "grad_norm": 3.5625,
1010
+ "learning_rate": 1.026323833869206e-07,
1011
+ "loss": 4.6933,
1012
+ "step": 700
1013
+ },
1014
+ {
1015
+ "epoch": 2.5824175824175826,
1016
+ "grad_norm": 3.59375,
1017
+ "learning_rate": 9.432442848846289e-08,
1018
+ "loss": 4.731,
1019
+ "step": 705
1020
+ },
1021
+ {
1022
+ "epoch": 2.600732600732601,
1023
+ "grad_norm": 3.53125,
1024
+ "learning_rate": 8.63504320540438e-08,
1025
+ "loss": 4.7246,
1026
+ "step": 710
1027
+ },
1028
+ {
1029
+ "epoch": 2.619047619047619,
1030
+ "grad_norm": 3.53125,
1031
+ "learning_rate": 7.871333441206052e-08,
1032
+ "loss": 4.7311,
1033
+ "step": 715
1034
+ },
1035
+ {
1036
+ "epoch": 2.6373626373626373,
1037
+ "grad_norm": 3.53125,
1038
+ "learning_rate": 7.141595166297832e-08,
1039
+ "loss": 4.7213,
1040
+ "step": 720
1041
+ },
1042
+ {
1043
+ "epoch": 2.6556776556776556,
1044
+ "grad_norm": 3.484375,
1045
+ "learning_rate": 6.446097464092248e-08,
1046
+ "loss": 4.6983,
1047
+ "step": 725
1048
+ },
1049
+ {
1050
+ "epoch": 2.6739926739926743,
1051
+ "grad_norm": 3.546875,
1052
+ "learning_rate": 5.78509679214616e-08,
1053
+ "loss": 4.6839,
1054
+ "step": 730
1055
+ },
1056
+ {
1057
+ "epoch": 2.6923076923076925,
1058
+ "grad_norm": 3.5625,
1059
+ "learning_rate": 5.1588368875946864e-08,
1060
+ "loss": 4.7098,
1061
+ "step": 735
1062
+ },
1063
+ {
1064
+ "epoch": 2.7106227106227108,
1065
+ "grad_norm": 3.65625,
1066
+ "learning_rate": 4.567548677275601e-08,
1067
+ "loss": 4.6954,
1068
+ "step": 740
1069
+ },
1070
+ {
1071
+ "epoch": 2.728937728937729,
1072
+ "grad_norm": 3.578125,
1073
+ "learning_rate": 4.0114501925775925e-08,
1074
+ "loss": 4.7284,
1075
+ "step": 745
1076
+ },
1077
+ {
1078
+ "epoch": 2.7472527472527473,
1079
+ "grad_norm": 3.765625,
1080
+ "learning_rate": 3.490746489043317e-08,
1081
+ "loss": 4.7099,
1082
+ "step": 750
1083
+ },
1084
+ {
1085
+ "epoch": 2.7655677655677655,
1086
+ "grad_norm": 3.5625,
1087
+ "learning_rate": 3.005629570757373e-08,
1088
+ "loss": 4.7144,
1089
+ "step": 755
1090
+ },
1091
+ {
1092
+ "epoch": 2.7838827838827838,
1093
+ "grad_norm": 3.625,
1094
+ "learning_rate": 2.5562783195467675e-08,
1095
+ "loss": 4.6949,
1096
+ "step": 760
1097
+ },
1098
+ {
1099
+ "epoch": 2.802197802197802,
1100
+ "grad_norm": 3.515625,
1101
+ "learning_rate": 2.1428584290201114e-08,
1102
+ "loss": 4.7045,
1103
+ "step": 765
1104
+ },
1105
+ {
1106
+ "epoch": 2.8205128205128203,
1107
+ "grad_norm": 3.703125,
1108
+ "learning_rate": 1.7655223434698053e-08,
1109
+ "loss": 4.688,
1110
+ "step": 770
1111
+ },
1112
+ {
1113
+ "epoch": 2.838827838827839,
1114
+ "grad_norm": 3.578125,
1115
+ "learning_rate": 1.4244092016597931e-08,
1116
+ "loss": 4.7177,
1117
+ "step": 775
1118
+ },
1119
+ {
1120
+ "epoch": 2.857142857142857,
1121
+ "grad_norm": 3.59375,
1122
+ "learning_rate": 1.11964478551958e-08,
1123
+ "loss": 4.7073,
1124
+ "step": 780
1125
+ },
1126
+ {
1127
+ "epoch": 2.8754578754578755,
1128
+ "grad_norm": 3.453125,
1129
+ "learning_rate": 8.513414737635005e-09,
1130
+ "loss": 4.7176,
1131
+ "step": 785
1132
+ },
1133
+ {
1134
+ "epoch": 2.8937728937728937,
1135
+ "grad_norm": 3.515625,
1136
+ "learning_rate": 6.1959820045215385e-09,
1137
+ "loss": 4.6779,
1138
+ "step": 790
1139
+ },
1140
+ {
1141
+ "epoch": 2.912087912087912,
1142
+ "grad_norm": 3.484375,
1143
+ "learning_rate": 4.245004185115752e-09,
1144
+ "loss": 4.7052,
1145
+ "step": 795
1146
+ },
1147
+ {
1148
+ "epoch": 2.9304029304029307,
1149
+ "grad_norm": 3.484375,
1150
+ "learning_rate": 2.661200682232745e-09,
1151
+ "loss": 4.7179,
1152
+ "step": 800
1153
+ },
1154
+ {
1155
+ "epoch": 2.948717948717949,
1156
+ "grad_norm": 3.640625,
1157
+ "learning_rate": 1.4451555069708853e-09,
1158
+ "loss": 4.6552,
1159
+ "step": 805
1160
+ },
1161
+ {
1162
+ "epoch": 2.967032967032967,
1163
+ "grad_norm": 4.21875,
1164
+ "learning_rate": 5.973170633631897e-10,
1165
+ "loss": 4.6979,
1166
+ "step": 810
1167
+ },
1168
+ {
1169
+ "epoch": 2.9853479853479854,
1170
+ "grad_norm": 3.578125,
1171
+ "learning_rate": 1.1799798303335772e-10,
1172
+ "loss": 4.7113,
1173
+ "step": 815
1174
+ },
1175
+ {
1176
+ "epoch": 3.0,
1177
+ "step": 819,
1178
+ "total_flos": 5.361516984661967e+18,
1179
+ "train_loss": 5.0057218712328115,
1180
+ "train_runtime": 5286.5666,
1181
+ "train_samples_per_second": 9.883,
1182
+ "train_steps_per_second": 0.155
1183
+ }
1184
+ ],
1185
+ "logging_steps": 5,
1186
+ "max_steps": 819,
1187
+ "num_input_tokens_seen": 0,
1188
+ "num_train_epochs": 3,
1189
+ "save_steps": 300,
1190
+ "stateful_callbacks": {
1191
+ "TrainerControl": {
1192
+ "args": {
1193
+ "should_epoch_stop": false,
1194
+ "should_evaluate": false,
1195
+ "should_log": false,
1196
+ "should_save": true,
1197
+ "should_training_stop": true
1198
+ },
1199
+ "attributes": {}
1200
+ }
1201
+ },
1202
+ "total_flos": 5.361516984661967e+18,
1203
+ "train_batch_size": 8,
1204
+ "trial_name": null,
1205
+ "trial_params": null
1206
+ }