yihanwang617 commited on
Commit
101c539
1 Parent(s): 499891d

Model save

Browse files
README.md ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: tinyllama-sft-vicuna-full-no-completion-mask
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # tinyllama-sft-vicuna-full-no-completion-mask
17
+
18
+ This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.8864
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 2e-05
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - distributed_type: multi-GPU
44
+ - num_devices: 4
45
+ - gradient_accumulation_steps: 2
46
+ - total_train_batch_size: 128
47
+ - total_eval_batch_size: 32
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: cosine
50
+ - lr_scheduler_warmup_ratio: 0.1
51
+ - num_epochs: 1
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:-----:|:----:|:---------------:|
57
+ | 0.9086 | 1.0 | 732 | 0.8864 |
58
+
59
+
60
+ ### Framework versions
61
+
62
+ - Transformers 4.39.0.dev0
63
+ - Pytorch 2.2.1+cu121
64
+ - Datasets 2.14.6
65
+ - Tokenizers 0.15.0
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.9459957421803084,
4
+ "train_runtime": 6147.7675,
5
+ "train_samples": 93645,
6
+ "train_samples_per_second": 15.232,
7
+ "train_steps_per_second": 0.119
8
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 2048,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.39.0.dev0"
7
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1916191a2eab5f99f70794dd6d8d0c651681386af79e991460e932a9888dbd17
3
  size 2200128056
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e98d2fe686f04193f7630a200b618c0538b453f690c72bbf204f973ada7e02a
3
  size 2200128056
runs/May23_22-38-06_nova.cs.ucla.edu/events.out.tfevents.1716529643.nova.cs.ucla.edu.3260057.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:81dbcf1b18d5430ac8060bc92255891c8dc7f4bfde8d330f32137cb38935877c
3
- size 34562
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b09c9cdd274d4cbb5fadc079eb0d10b47100f871480e1aa367dec9644fc71d66
3
+ size 36453
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.9459957421803084,
4
+ "train_runtime": 6147.7675,
5
+ "train_samples": 93645,
6
+ "train_samples_per_second": 15.232,
7
+ "train_steps_per_second": 0.119
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1067 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 732,
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": 9.658415693590747,
14
+ "learning_rate": 2.702702702702703e-07,
15
+ "loss": 1.4394,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "grad_norm": 9.876292569045443,
21
+ "learning_rate": 1.3513513513513515e-06,
22
+ "loss": 1.4271,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.01,
27
+ "grad_norm": 9.586588537619404,
28
+ "learning_rate": 2.702702702702703e-06,
29
+ "loss": 1.4014,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.02,
34
+ "grad_norm": 3.2928588175082862,
35
+ "learning_rate": 4.0540540540540545e-06,
36
+ "loss": 1.242,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.03,
41
+ "grad_norm": 1.7916899568159483,
42
+ "learning_rate": 5.405405405405406e-06,
43
+ "loss": 1.1312,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "grad_norm": 0.9818778768043777,
49
+ "learning_rate": 6.7567567567567575e-06,
50
+ "loss": 1.0505,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.04,
55
+ "grad_norm": 0.8394686945518413,
56
+ "learning_rate": 8.108108108108109e-06,
57
+ "loss": 1.0424,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.05,
62
+ "grad_norm": 0.7551190468266619,
63
+ "learning_rate": 9.45945945945946e-06,
64
+ "loss": 1.0518,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.05,
69
+ "grad_norm": 0.6711648718562863,
70
+ "learning_rate": 1.0810810810810812e-05,
71
+ "loss": 1.0121,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.06,
76
+ "grad_norm": 0.6200076782701016,
77
+ "learning_rate": 1.2162162162162164e-05,
78
+ "loss": 1.057,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.07,
83
+ "grad_norm": 0.658782644893673,
84
+ "learning_rate": 1.3513513513513515e-05,
85
+ "loss": 1.013,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.08,
90
+ "grad_norm": 0.5356817576419274,
91
+ "learning_rate": 1.4864864864864865e-05,
92
+ "loss": 1.0,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.08,
97
+ "grad_norm": 0.5048984401091999,
98
+ "learning_rate": 1.6216216216216218e-05,
99
+ "loss": 1.0219,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.09,
104
+ "grad_norm": 0.517547275458162,
105
+ "learning_rate": 1.756756756756757e-05,
106
+ "loss": 1.0327,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "grad_norm": 0.4925361764186976,
112
+ "learning_rate": 1.891891891891892e-05,
113
+ "loss": 1.0107,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.1,
118
+ "grad_norm": 0.4945886669805455,
119
+ "learning_rate": 1.999988602302209e-05,
120
+ "loss": 1.0252,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.11,
125
+ "grad_norm": 0.49947937008697635,
126
+ "learning_rate": 1.9995897101594454e-05,
127
+ "loss": 0.9825,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.12,
132
+ "grad_norm": 0.48799350342765896,
133
+ "learning_rate": 1.99862119291555e-05,
134
+ "loss": 0.9848,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.12,
139
+ "grad_norm": 0.49925057798584527,
140
+ "learning_rate": 1.997083602488702e-05,
141
+ "loss": 0.9982,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.13,
146
+ "grad_norm": 0.533633849127977,
147
+ "learning_rate": 1.994977815088504e-05,
148
+ "loss": 1.0089,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "grad_norm": 0.5205221053880554,
154
+ "learning_rate": 1.9923050307166655e-05,
155
+ "loss": 1.0036,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.14,
160
+ "grad_norm": 0.4728771654069777,
161
+ "learning_rate": 1.989066772483171e-05,
162
+ "loss": 1.0057,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "grad_norm": 0.5330378379531007,
168
+ "learning_rate": 1.9852648857383224e-05,
169
+ "loss": 1.0004,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.16,
174
+ "grad_norm": 0.5474915550296676,
175
+ "learning_rate": 1.9809015370211505e-05,
176
+ "loss": 0.9582,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.16,
181
+ "grad_norm": 0.5065591909160228,
182
+ "learning_rate": 1.9759792128247922e-05,
183
+ "loss": 0.9875,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.17,
188
+ "grad_norm": 0.5041379472534961,
189
+ "learning_rate": 1.9705007181795416e-05,
190
+ "loss": 0.9839,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "grad_norm": 0.5152712578505293,
196
+ "learning_rate": 1.964469175054377e-05,
197
+ "loss": 0.9817,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.18,
202
+ "grad_norm": 0.5181637583730708,
203
+ "learning_rate": 1.9578880205778793e-05,
204
+ "loss": 0.9601,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "grad_norm": 0.5082366273570456,
210
+ "learning_rate": 1.950761005079556e-05,
211
+ "loss": 0.9594,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.2,
216
+ "grad_norm": 0.5163440396909443,
217
+ "learning_rate": 1.9430921899526786e-05,
218
+ "loss": 0.9601,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.2,
223
+ "grad_norm": 0.5286098804973952,
224
+ "learning_rate": 1.934885945339865e-05,
225
+ "loss": 0.9603,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.21,
230
+ "grad_norm": 0.5128368272170564,
231
+ "learning_rate": 1.9261469476427122e-05,
232
+ "loss": 0.9959,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "grad_norm": 0.5153476529706232,
238
+ "learning_rate": 1.916880176856909e-05,
239
+ "loss": 0.9494,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.23,
244
+ "grad_norm": 0.4950657638957343,
245
+ "learning_rate": 1.907090913734341e-05,
246
+ "loss": 0.9522,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "grad_norm": 0.48067724327955486,
252
+ "learning_rate": 1.896784736773805e-05,
253
+ "loss": 0.9881,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.24,
258
+ "grad_norm": 0.5010564864661735,
259
+ "learning_rate": 1.885967519042054e-05,
260
+ "loss": 0.9664,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.25,
265
+ "grad_norm": 0.5157306256625271,
266
+ "learning_rate": 1.8746454248269777e-05,
267
+ "loss": 0.992,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.25,
272
+ "grad_norm": 0.5051241069978706,
273
+ "learning_rate": 1.862824906124826e-05,
274
+ "loss": 0.9569,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "grad_norm": 0.5095950283674965,
280
+ "learning_rate": 1.850512698963485e-05,
281
+ "loss": 0.9741,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.27,
286
+ "grad_norm": 0.5227451618958809,
287
+ "learning_rate": 1.8377158195638877e-05,
288
+ "loss": 0.9331,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "grad_norm": 0.5181101956320976,
294
+ "learning_rate": 1.8244415603417603e-05,
295
+ "loss": 0.9603,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.28,
300
+ "grad_norm": 0.510093680653842,
301
+ "learning_rate": 1.8106974857519737e-05,
302
+ "loss": 0.9562,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.29,
307
+ "grad_norm": 0.5408000977823474,
308
+ "learning_rate": 1.7964914279778716e-05,
309
+ "loss": 0.9525,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.29,
314
+ "grad_norm": 0.5221005548724054,
315
+ "learning_rate": 1.78183148246803e-05,
316
+ "loss": 0.9461,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "grad_norm": 0.5228654219110396,
322
+ "learning_rate": 1.7667260033229953e-05,
323
+ "loss": 0.93,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.31,
328
+ "grad_norm": 0.5156616142594198,
329
+ "learning_rate": 1.751183598534625e-05,
330
+ "loss": 0.9523,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "grad_norm": 0.4992025615094016,
336
+ "learning_rate": 1.7352131250807466e-05,
337
+ "loss": 0.9447,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.32,
342
+ "grad_norm": 0.47695347755887807,
343
+ "learning_rate": 1.7188236838779297e-05,
344
+ "loss": 0.933,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.33,
349
+ "grad_norm": 0.5219986591713203,
350
+ "learning_rate": 1.702024614595248e-05,
351
+ "loss": 0.9359,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.33,
356
+ "grad_norm": 0.5247969957734948,
357
+ "learning_rate": 1.6848254903319866e-05,
358
+ "loss": 0.9378,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.34,
363
+ "grad_norm": 0.5260544851199538,
364
+ "learning_rate": 1.6672361121623238e-05,
365
+ "loss": 0.919,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.35,
370
+ "grad_norm": 0.5249428172457546,
371
+ "learning_rate": 1.6492665035501048e-05,
372
+ "loss": 0.9369,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.36,
377
+ "grad_norm": 0.5162356214174872,
378
+ "learning_rate": 1.6309269046368777e-05,
379
+ "loss": 0.9515,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.36,
384
+ "grad_norm": 0.5010926857825899,
385
+ "learning_rate": 1.612227766406461e-05,
386
+ "loss": 0.9609,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.37,
391
+ "grad_norm": 0.5382068418300701,
392
+ "learning_rate": 1.5931797447293553e-05,
393
+ "loss": 0.9447,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.38,
398
+ "grad_norm": 0.5295330507632734,
399
+ "learning_rate": 1.5737936942904025e-05,
400
+ "loss": 0.9561,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.38,
405
+ "grad_norm": 0.5785403455994167,
406
+ "learning_rate": 1.554080662403144e-05,
407
+ "loss": 0.928,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.39,
412
+ "grad_norm": 0.49666012212229643,
413
+ "learning_rate": 1.5340518827144145e-05,
414
+ "loss": 0.9481,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.4,
419
+ "grad_norm": 0.5289929313831263,
420
+ "learning_rate": 1.5137187688027437e-05,
421
+ "loss": 0.9284,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.4,
426
+ "grad_norm": 0.4842301638025414,
427
+ "learning_rate": 1.4930929076742317e-05,
428
+ "loss": 0.9154,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.41,
433
+ "grad_norm": 0.5192170165075863,
434
+ "learning_rate": 1.4721860531595868e-05,
435
+ "loss": 0.922,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.42,
440
+ "grad_norm": 0.503974988503286,
441
+ "learning_rate": 1.451010119216102e-05,
442
+ "loss": 0.9149,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.42,
447
+ "grad_norm": 0.507463468927673,
448
+ "learning_rate": 1.4295771731383799e-05,
449
+ "loss": 0.9264,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.43,
454
+ "grad_norm": 0.4963800283400335,
455
+ "learning_rate": 1.4078994286816768e-05,
456
+ "loss": 0.9354,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.44,
461
+ "grad_norm": 0.5178048187839067,
462
+ "learning_rate": 1.3859892391017867e-05,
463
+ "loss": 0.9248,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.44,
468
+ "grad_norm": 0.4961104164383763,
469
+ "learning_rate": 1.3638590901154276e-05,
470
+ "loss": 0.947,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.45,
475
+ "grad_norm": 0.5205766773557013,
476
+ "learning_rate": 1.341521592785145e-05,
477
+ "loss": 0.9416,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.46,
482
+ "grad_norm": 0.5453988235731033,
483
+ "learning_rate": 1.3189894763327851e-05,
484
+ "loss": 0.9288,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.46,
489
+ "grad_norm": 0.5021959145630082,
490
+ "learning_rate": 1.2962755808856341e-05,
491
+ "loss": 0.9545,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.47,
496
+ "grad_norm": 0.5269775945473103,
497
+ "learning_rate": 1.2733928501593587e-05,
498
+ "loss": 0.9568,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.48,
503
+ "grad_norm": 0.5460646239634884,
504
+ "learning_rate": 1.2503543240819127e-05,
505
+ "loss": 0.9281,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.48,
510
+ "grad_norm": 0.5290020549006504,
511
+ "learning_rate": 1.227173131362619e-05,
512
+ "loss": 0.9178,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.49,
517
+ "grad_norm": 0.517202069917391,
518
+ "learning_rate": 1.2038624820106572e-05,
519
+ "loss": 0.8777,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.5,
524
+ "grad_norm": 0.5016455236354784,
525
+ "learning_rate": 1.1804356598072223e-05,
526
+ "loss": 0.9372,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.51,
531
+ "grad_norm": 0.49187913515804305,
532
+ "learning_rate": 1.1569060147356441e-05,
533
+ "loss": 0.8909,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.51,
538
+ "grad_norm": 0.5137148402464546,
539
+ "learning_rate": 1.133286955373779e-05,
540
+ "loss": 0.9219,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.52,
545
+ "grad_norm": 0.48367513170223,
546
+ "learning_rate": 1.1095919412530136e-05,
547
+ "loss": 0.9152,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.53,
552
+ "grad_norm": 0.5431723767746645,
553
+ "learning_rate": 1.0858344751882304e-05,
554
+ "loss": 0.9299,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.53,
559
+ "grad_norm": 0.499436195510236,
560
+ "learning_rate": 1.0620280955831088e-05,
561
+ "loss": 0.9133,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.54,
566
+ "grad_norm": 0.4798245609936736,
567
+ "learning_rate": 1.038186368715145e-05,
568
+ "loss": 0.9202,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.55,
573
+ "grad_norm": 0.5224088939393612,
574
+ "learning_rate": 1.0143228810047877e-05,
575
+ "loss": 0.9039,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.55,
580
+ "grad_norm": 0.5281235095829454,
581
+ "learning_rate": 9.904512312730948e-06,
582
+ "loss": 0.9149,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.56,
587
+ "grad_norm": 0.527644785869163,
588
+ "learning_rate": 9.665850229923258e-06,
589
+ "loss": 0.9152,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.57,
594
+ "grad_norm": 0.5060754191665021,
595
+ "learning_rate": 9.4273785653388e-06,
596
+ "loss": 0.9283,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.57,
601
+ "grad_norm": 0.47138126227952987,
602
+ "learning_rate": 9.189233214180057e-06,
603
+ "loss": 0.8978,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.58,
608
+ "grad_norm": 0.4952269194383896,
609
+ "learning_rate": 8.951549885696889e-06,
610
+ "loss": 0.9172,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.59,
615
+ "grad_norm": 0.524189618003053,
616
+ "learning_rate": 8.714464025851428e-06,
617
+ "loss": 0.923,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.59,
622
+ "grad_norm": 0.48690727308343434,
623
+ "learning_rate": 8.478110740132971e-06,
624
+ "loss": 0.926,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.6,
629
+ "grad_norm": 0.5143235296126093,
630
+ "learning_rate": 8.242624716566928e-06,
631
+ "loss": 0.9351,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.61,
636
+ "grad_norm": 0.5131865229287261,
637
+ "learning_rate": 8.008140148961642e-06,
638
+ "loss": 0.9169,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.61,
643
+ "grad_norm": 0.49549144988639615,
644
+ "learning_rate": 7.774790660436857e-06,
645
+ "loss": 0.9089,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.62,
650
+ "grad_norm": 0.48031633340251334,
651
+ "learning_rate": 7.542709227277396e-06,
652
+ "loss": 0.923,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.63,
657
+ "grad_norm": 0.5244467963752784,
658
+ "learning_rate": 7.312028103155426e-06,
659
+ "loss": 0.8991,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.64,
664
+ "grad_norm": 0.5448301982574317,
665
+ "learning_rate": 7.0828787437645455e-06,
666
+ "loss": 0.8748,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.64,
671
+ "grad_norm": 0.5320333401965247,
672
+ "learning_rate": 6.8553917319085676e-06,
673
+ "loss": 0.9169,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.65,
678
+ "grad_norm": 0.4778407461352207,
679
+ "learning_rate": 6.629696703087755e-06,
680
+ "loss": 0.9127,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.66,
685
+ "grad_norm": 0.5225492873477832,
686
+ "learning_rate": 6.405922271624874e-06,
687
+ "loss": 0.9112,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.66,
692
+ "grad_norm": 0.5123626688586312,
693
+ "learning_rate": 6.184195957373176e-06,
694
+ "loss": 0.9097,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.67,
699
+ "grad_norm": 0.5025191272317765,
700
+ "learning_rate": 5.964644113048079e-06,
701
+ "loss": 0.9147,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.68,
706
+ "grad_norm": 0.48965471328713384,
707
+ "learning_rate": 5.74739185222394e-06,
708
+ "loss": 0.9222,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.68,
713
+ "grad_norm": 0.5288978316966069,
714
+ "learning_rate": 5.532562978036964e-06,
715
+ "loss": 0.8984,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.69,
720
+ "grad_norm": 0.5102011962342545,
721
+ "learning_rate": 5.320279912634907e-06,
722
+ "loss": 0.8971,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.7,
727
+ "grad_norm": 0.5046037978403557,
728
+ "learning_rate": 5.110663627413695e-06,
729
+ "loss": 0.8965,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.7,
734
+ "grad_norm": 0.5527850768232141,
735
+ "learning_rate": 4.903833574080825e-06,
736
+ "loss": 0.9333,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.71,
741
+ "grad_norm": 0.5015924720632916,
742
+ "learning_rate": 4.6999076165847214e-06,
743
+ "loss": 0.9128,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.72,
748
+ "grad_norm": 0.5259030791021958,
749
+ "learning_rate": 4.499001963948929e-06,
750
+ "loss": 0.8977,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.72,
755
+ "grad_norm": 0.4922610443183964,
756
+ "learning_rate": 4.301231104049359e-06,
757
+ "loss": 0.8812,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.73,
762
+ "grad_norm": 0.513772562818194,
763
+ "learning_rate": 4.106707738372357e-06,
764
+ "loss": 0.9142,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.74,
769
+ "grad_norm": 0.48791049706062967,
770
+ "learning_rate": 3.915542717790759e-06,
771
+ "loss": 0.9158,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.74,
776
+ "grad_norm": 0.5169968541006628,
777
+ "learning_rate": 3.727844979394526e-06,
778
+ "loss": 0.9277,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.75,
783
+ "grad_norm": 0.5077943410753647,
784
+ "learning_rate": 3.543721484411976e-06,
785
+ "loss": 0.8959,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.76,
790
+ "grad_norm": 0.4817889325039375,
791
+ "learning_rate": 3.3632771572569878e-06,
792
+ "loss": 0.9101,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.77,
797
+ "grad_norm": 0.5385996666793137,
798
+ "learning_rate": 3.1866148257368666e-06,
799
+ "loss": 0.8938,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.77,
804
+ "grad_norm": 0.5225728816923544,
805
+ "learning_rate": 3.0138351624550165e-06,
806
+ "loss": 0.8894,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.78,
811
+ "grad_norm": 0.5124915386197431,
812
+ "learning_rate": 2.845036627441755e-06,
813
+ "loss": 0.9064,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.79,
818
+ "grad_norm": 0.5061857842019366,
819
+ "learning_rate": 2.6803154120460007e-06,
820
+ "loss": 0.8816,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.79,
825
+ "grad_norm": 0.5114177573176913,
826
+ "learning_rate": 2.5197653841197546e-06,
827
+ "loss": 0.9156,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.8,
832
+ "grad_norm": 0.49787558407973875,
833
+ "learning_rate": 2.3634780345266805e-06,
834
+ "loss": 0.9244,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.81,
839
+ "grad_norm": 0.5208494862823159,
840
+ "learning_rate": 2.211542425005223e-06,
841
+ "loss": 0.8811,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.81,
846
+ "grad_norm": 0.5079793748028331,
847
+ "learning_rate": 2.064045137415982e-06,
848
+ "loss": 0.9066,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.82,
853
+ "grad_norm": 0.49224088294773566,
854
+ "learning_rate": 1.9210702244022616e-06,
855
+ "loss": 0.9038,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.83,
860
+ "grad_norm": 0.5159941101908895,
861
+ "learning_rate": 1.7826991614919264e-06,
862
+ "loss": 0.9188,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.83,
867
+ "grad_norm": 0.5064241877692456,
868
+ "learning_rate": 1.6490108006678495e-06,
869
+ "loss": 0.8913,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.84,
874
+ "grad_norm": 0.504152765761692,
875
+ "learning_rate": 1.5200813254334013e-06,
876
+ "loss": 0.9058,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.85,
881
+ "grad_norm": 0.4798941893934502,
882
+ "learning_rate": 1.3959842073986085e-06,
883
+ "loss": 0.9029,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.85,
888
+ "grad_norm": 0.5155303614528817,
889
+ "learning_rate": 1.2767901644116943e-06,
890
+ "loss": 0.9095,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.86,
895
+ "grad_norm": 0.4959233986544881,
896
+ "learning_rate": 1.1625671202598875e-06,
897
+ "loss": 0.8903,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.87,
902
+ "grad_norm": 0.5554850290932043,
903
+ "learning_rate": 1.0533801659624531e-06,
904
+ "loss": 0.9355,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.87,
909
+ "grad_norm": 0.5083400957738834,
910
+ "learning_rate": 9.492915226779809e-07,
911
+ "loss": 0.9263,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.88,
916
+ "grad_norm": 0.45712422940481173,
917
+ "learning_rate": 8.503605062471187e-07,
918
+ "loss": 0.9049,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.89,
923
+ "grad_norm": 0.5031609635595137,
924
+ "learning_rate": 7.566434933909006e-07,
925
+ "loss": 0.9123,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.89,
930
+ "grad_norm": 0.4981707548553113,
931
+ "learning_rate": 6.681938895839746e-07,
932
+ "loss": 0.8784,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.9,
937
+ "grad_norm": 0.4720191005790808,
938
+ "learning_rate": 5.850620986210198e-07,
939
+ "loss": 0.8619,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.91,
944
+ "grad_norm": 0.5135344387921242,
945
+ "learning_rate": 5.072954938936925e-07,
946
+ "loss": 0.9377,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.92,
951
+ "grad_norm": 0.5221400934605558,
952
+ "learning_rate": 4.3493839139447716e-07,
953
+ "loss": 0.9072,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.92,
958
+ "grad_norm": 0.46747238598666885,
959
+ "learning_rate": 3.6803202446282217e-07,
960
+ "loss": 0.8966,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.93,
965
+ "grad_norm": 0.5058287813526128,
966
+ "learning_rate": 3.0661452028795335e-07,
967
+ "loss": 0.922,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.94,
972
+ "grad_norm": 0.4891511781301968,
973
+ "learning_rate": 2.507208781817638e-07,
974
+ "loss": 0.9009,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.94,
979
+ "grad_norm": 0.4800266914233775,
980
+ "learning_rate": 2.0038294963413251e-07,
981
+ "loss": 0.8974,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.95,
986
+ "grad_norm": 0.4962683953232167,
987
+ "learning_rate": 1.556294201620734e-07,
988
+ "loss": 0.9288,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.96,
993
+ "grad_norm": 0.529691888573167,
994
+ "learning_rate": 1.1648579296304252e-07,
995
+ "loss": 0.8941,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.96,
1000
+ "grad_norm": 0.4622837679446212,
1001
+ "learning_rate": 8.297437438170797e-08,
1002
+ "loss": 0.9002,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.97,
1007
+ "grad_norm": 0.5095413176345687,
1008
+ "learning_rate": 5.51142611984834e-08,
1009
+ "loss": 0.9122,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.98,
1014
+ "grad_norm": 0.5169397392729471,
1015
+ "learning_rate": 3.2921329747056527e-08,
1016
+ "loss": 0.8869,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.98,
1021
+ "grad_norm": 0.4761774021864735,
1022
+ "learning_rate": 1.6408226867118404e-08,
1023
+ "loss": 0.8698,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.99,
1028
+ "grad_norm": 0.49969690572300623,
1029
+ "learning_rate": 5.584362697453882e-09,
1030
+ "loss": 0.9178,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 1.0,
1035
+ "grad_norm": 0.5099959778794521,
1036
+ "learning_rate": 4.5590531348227443e-10,
1037
+ "loss": 0.9086,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 1.0,
1042
+ "eval_loss": 0.8863584399223328,
1043
+ "eval_runtime": 9.4646,
1044
+ "eval_samples_per_second": 52.828,
1045
+ "eval_steps_per_second": 1.691,
1046
+ "step": 732
1047
+ },
1048
+ {
1049
+ "epoch": 1.0,
1050
+ "step": 732,
1051
+ "total_flos": 106034689474560.0,
1052
+ "train_loss": 0.9459957421803084,
1053
+ "train_runtime": 6147.7675,
1054
+ "train_samples_per_second": 15.232,
1055
+ "train_steps_per_second": 0.119
1056
+ }
1057
+ ],
1058
+ "logging_steps": 5,
1059
+ "max_steps": 732,
1060
+ "num_input_tokens_seen": 0,
1061
+ "num_train_epochs": 1,
1062
+ "save_steps": 100,
1063
+ "total_flos": 106034689474560.0,
1064
+ "train_batch_size": 16,
1065
+ "trial_name": null,
1066
+ "trial_params": null
1067
+ }