yakazimir commited on
Commit
bb0edb6
·
verified ·
1 Parent(s): a62ef22

Model save

Browse files
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: HuggingFaceTB/SmolLM2-1.7B
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: smollm_1_7B_ultrachat_v2
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
+ # smollm_1_7B_ultrachat_v2
18
+
19
+ This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 1.2902
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: 5e-06
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 8
46
+ - gradient_accumulation_steps: 4
47
+ - total_train_batch_size: 256
48
+ - total_eval_batch_size: 64
49
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
+ - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.05
52
+ - num_epochs: 1.0
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:------:|:----:|:---------------:|
58
+ | 1.3939 | 0.2463 | 200 | 1.3921 |
59
+ | 1.2806 | 0.4926 | 400 | 1.3156 |
60
+ | 1.2827 | 0.7389 | 600 | 1.2938 |
61
+ | 1.2807 | 0.9852 | 800 | 1.2902 |
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 4.48.3
67
+ - Pytorch 2.2.2+cu121
68
+ - Datasets 2.18.0
69
+ - Tokenizers 0.21.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 277556883980288.0,
4
+ "train_loss": 1.458759990231744,
5
+ "train_runtime": 5428.2383,
6
+ "train_samples": 207865,
7
+ "train_samples_per_second": 38.293,
8
+ "train_steps_per_second": 0.15
9
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.48.3"
7
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 277556883980288.0,
4
+ "train_loss": 1.458759990231744,
5
+ "train_runtime": 5428.2383,
6
+ "train_samples": 207865,
7
+ "train_samples_per_second": 38.293,
8
+ "train_steps_per_second": 0.15
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1215 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 200,
6
+ "global_step": 812,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0012315270935960591,
13
+ "grad_norm": 53.61643111350721,
14
+ "learning_rate": 1.2195121951219514e-07,
15
+ "loss": 4.2831,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.006157635467980296,
20
+ "grad_norm": 48.00417504001698,
21
+ "learning_rate": 6.097560975609757e-07,
22
+ "loss": 4.1517,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.012315270935960592,
27
+ "grad_norm": 48.656126717388894,
28
+ "learning_rate": 1.2195121951219514e-06,
29
+ "loss": 4.1506,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.01847290640394089,
34
+ "grad_norm": 52.06244750805184,
35
+ "learning_rate": 1.8292682926829268e-06,
36
+ "loss": 4.0921,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.024630541871921183,
41
+ "grad_norm": 41.842979233989844,
42
+ "learning_rate": 2.4390243902439027e-06,
43
+ "loss": 3.832,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.03078817733990148,
48
+ "grad_norm": 26.054139160186445,
49
+ "learning_rate": 3.0487804878048782e-06,
50
+ "loss": 3.4798,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.03694581280788178,
55
+ "grad_norm": 22.768708360256497,
56
+ "learning_rate": 3.6585365853658537e-06,
57
+ "loss": 3.1223,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.04310344827586207,
62
+ "grad_norm": 11.807556286649826,
63
+ "learning_rate": 4.268292682926829e-06,
64
+ "loss": 2.774,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.04926108374384237,
69
+ "grad_norm": 10.768737194323332,
70
+ "learning_rate": 4.8780487804878055e-06,
71
+ "loss": 2.4807,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.05541871921182266,
76
+ "grad_norm": 7.978916355789078,
77
+ "learning_rate": 4.999667943970417e-06,
78
+ "loss": 2.3202,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.06157635467980296,
83
+ "grad_norm": 6.323820375851387,
84
+ "learning_rate": 4.9983191175272635e-06,
85
+ "loss": 2.1276,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.06773399014778325,
90
+ "grad_norm": 5.180335911696107,
91
+ "learning_rate": 4.995933326588439e-06,
92
+ "loss": 1.9964,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.07389162561576355,
97
+ "grad_norm": 4.28298322285663,
98
+ "learning_rate": 4.992511561411944e-06,
99
+ "loss": 1.8959,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.08004926108374384,
104
+ "grad_norm": 3.8950053865689425,
105
+ "learning_rate": 4.988055242252294e-06,
106
+ "loss": 1.7978,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.08620689655172414,
111
+ "grad_norm": 3.5678308957701725,
112
+ "learning_rate": 4.982566218771011e-06,
113
+ "loss": 1.7083,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.09236453201970443,
118
+ "grad_norm": 3.280549835228229,
119
+ "learning_rate": 4.976046769268905e-06,
120
+ "loss": 1.6391,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.09852216748768473,
125
+ "grad_norm": 3.0095284508920943,
126
+ "learning_rate": 4.968499599740427e-06,
127
+ "loss": 1.6052,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.10467980295566502,
132
+ "grad_norm": 2.728976091948925,
133
+ "learning_rate": 4.959927842750501e-06,
134
+ "loss": 1.5852,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.11083743842364532,
139
+ "grad_norm": 2.773190869191402,
140
+ "learning_rate": 4.9503350561343165e-06,
141
+ "loss": 1.5389,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.11699507389162561,
146
+ "grad_norm": 2.495284943078218,
147
+ "learning_rate": 4.939725221520586e-06,
148
+ "loss": 1.529,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.12315270935960591,
153
+ "grad_norm": 2.2585070403096448,
154
+ "learning_rate": 4.928102742678918e-06,
155
+ "loss": 1.5012,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.12931034482758622,
160
+ "grad_norm": 2.3030504202779203,
161
+ "learning_rate": 4.915472443691962e-06,
162
+ "loss": 1.5228,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.1354679802955665,
167
+ "grad_norm": 2.139871025358791,
168
+ "learning_rate": 4.901839566953105e-06,
169
+ "loss": 1.4999,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.1416256157635468,
174
+ "grad_norm": 2.041219388782717,
175
+ "learning_rate": 4.887209770990537e-06,
176
+ "loss": 1.4933,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.1477832512315271,
181
+ "grad_norm": 1.943568456417746,
182
+ "learning_rate": 4.871589128118593e-06,
183
+ "loss": 1.4794,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.1539408866995074,
188
+ "grad_norm": 1.9973156144136615,
189
+ "learning_rate": 4.854984121917349e-06,
190
+ "loss": 1.4658,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.16009852216748768,
195
+ "grad_norm": 2.0438506441497934,
196
+ "learning_rate": 4.83740164454152e-06,
197
+ "loss": 1.4371,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.16625615763546797,
202
+ "grad_norm": 1.856918766416631,
203
+ "learning_rate": 4.8188489938597635e-06,
204
+ "loss": 1.4343,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.1724137931034483,
209
+ "grad_norm": 1.815968456698851,
210
+ "learning_rate": 4.799333870425594e-06,
211
+ "loss": 1.436,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.17857142857142858,
216
+ "grad_norm": 1.833214631845253,
217
+ "learning_rate": 4.778864374281147e-06,
218
+ "loss": 1.4358,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.18472906403940886,
223
+ "grad_norm": 1.7361805528112662,
224
+ "learning_rate": 4.757449001595149e-06,
225
+ "loss": 1.4141,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.19088669950738915,
230
+ "grad_norm": 1.663939988626721,
231
+ "learning_rate": 4.735096641136443e-06,
232
+ "loss": 1.4235,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.19704433497536947,
237
+ "grad_norm": 1.6174000907070356,
238
+ "learning_rate": 4.711816570584583e-06,
239
+ "loss": 1.4094,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.20320197044334976,
244
+ "grad_norm": 1.5393337213532907,
245
+ "learning_rate": 4.687618452678986e-06,
246
+ "loss": 1.423,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.20935960591133004,
251
+ "grad_norm": 1.6022576546137135,
252
+ "learning_rate": 4.662512331208277e-06,
253
+ "loss": 1.4229,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.21551724137931033,
258
+ "grad_norm": 1.4720779248871585,
259
+ "learning_rate": 4.636508626841458e-06,
260
+ "loss": 1.3835,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.22167487684729065,
265
+ "grad_norm": 1.4933242533565454,
266
+ "learning_rate": 4.609618132802661e-06,
267
+ "loss": 1.42,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.22783251231527094,
272
+ "grad_norm": 1.513300883824977,
273
+ "learning_rate": 4.581852010391253e-06,
274
+ "loss": 1.3995,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.23399014778325122,
279
+ "grad_norm": 1.4476120919655675,
280
+ "learning_rate": 4.55322178434918e-06,
281
+ "loss": 1.381,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.24014778325123154,
286
+ "grad_norm": 1.3952114387215153,
287
+ "learning_rate": 4.523739338077443e-06,
288
+ "loss": 1.3886,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.24630541871921183,
293
+ "grad_norm": 1.3697029918579813,
294
+ "learning_rate": 4.493416908703725e-06,
295
+ "loss": 1.3939,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.24630541871921183,
300
+ "eval_loss": 1.3920977115631104,
301
+ "eval_runtime": 145.3824,
302
+ "eval_samples_per_second": 158.96,
303
+ "eval_steps_per_second": 2.49,
304
+ "step": 200
305
+ },
306
+ {
307
+ "epoch": 0.2524630541871921,
308
+ "grad_norm": 1.2995335774783041,
309
+ "learning_rate": 4.462267082003183e-06,
310
+ "loss": 1.3624,
311
+ "step": 205
312
+ },
313
+ {
314
+ "epoch": 0.25862068965517243,
315
+ "grad_norm": 1.3420631400831042,
316
+ "learning_rate": 4.430302787174535e-06,
317
+ "loss": 1.3932,
318
+ "step": 210
319
+ },
320
+ {
321
+ "epoch": 0.2647783251231527,
322
+ "grad_norm": 1.249545343253794,
323
+ "learning_rate": 4.397537291473607e-06,
324
+ "loss": 1.36,
325
+ "step": 215
326
+ },
327
+ {
328
+ "epoch": 0.270935960591133,
329
+ "grad_norm": 1.2639981922437173,
330
+ "learning_rate": 4.36398419470656e-06,
331
+ "loss": 1.3546,
332
+ "step": 220
333
+ },
334
+ {
335
+ "epoch": 0.2770935960591133,
336
+ "grad_norm": 1.3038810090775859,
337
+ "learning_rate": 4.329657423585088e-06,
338
+ "loss": 1.3748,
339
+ "step": 225
340
+ },
341
+ {
342
+ "epoch": 0.2832512315270936,
343
+ "grad_norm": 1.2680995555961931,
344
+ "learning_rate": 4.2945712259459245e-06,
345
+ "loss": 1.3566,
346
+ "step": 230
347
+ },
348
+ {
349
+ "epoch": 0.2894088669950739,
350
+ "grad_norm": 1.1649048616562434,
351
+ "learning_rate": 4.258740164837079e-06,
352
+ "loss": 1.3858,
353
+ "step": 235
354
+ },
355
+ {
356
+ "epoch": 0.2955665024630542,
357
+ "grad_norm": 1.1320815166399936,
358
+ "learning_rate": 4.222179112473211e-06,
359
+ "loss": 1.3447,
360
+ "step": 240
361
+ },
362
+ {
363
+ "epoch": 0.3017241379310345,
364
+ "grad_norm": 1.1611812170906886,
365
+ "learning_rate": 4.184903244062709e-06,
366
+ "loss": 1.3779,
367
+ "step": 245
368
+ },
369
+ {
370
+ "epoch": 0.3078817733990148,
371
+ "grad_norm": 1.1257135240111318,
372
+ "learning_rate": 4.146928031508988e-06,
373
+ "loss": 1.3279,
374
+ "step": 250
375
+ },
376
+ {
377
+ "epoch": 0.31403940886699505,
378
+ "grad_norm": 1.1200758606679568,
379
+ "learning_rate": 4.108269236988648e-06,
380
+ "loss": 1.3651,
381
+ "step": 255
382
+ },
383
+ {
384
+ "epoch": 0.32019704433497537,
385
+ "grad_norm": 0.9671290050907145,
386
+ "learning_rate": 4.068942906409148e-06,
387
+ "loss": 1.3496,
388
+ "step": 260
389
+ },
390
+ {
391
+ "epoch": 0.3263546798029557,
392
+ "grad_norm": 1.0185962333371894,
393
+ "learning_rate": 4.028965362748714e-06,
394
+ "loss": 1.3436,
395
+ "step": 265
396
+ },
397
+ {
398
+ "epoch": 0.33251231527093594,
399
+ "grad_norm": 0.978818844718456,
400
+ "learning_rate": 3.988353199281242e-06,
401
+ "loss": 1.3178,
402
+ "step": 270
403
+ },
404
+ {
405
+ "epoch": 0.33866995073891626,
406
+ "grad_norm": 0.973392817512532,
407
+ "learning_rate": 3.9471232726890185e-06,
408
+ "loss": 1.3292,
409
+ "step": 275
410
+ },
411
+ {
412
+ "epoch": 0.3448275862068966,
413
+ "grad_norm": 1.0993106332736733,
414
+ "learning_rate": 3.905292696066095e-06,
415
+ "loss": 1.3295,
416
+ "step": 280
417
+ },
418
+ {
419
+ "epoch": 0.35098522167487683,
420
+ "grad_norm": 1.0275184655442997,
421
+ "learning_rate": 3.862878831815253e-06,
422
+ "loss": 1.3583,
423
+ "step": 285
424
+ },
425
+ {
426
+ "epoch": 0.35714285714285715,
427
+ "grad_norm": 0.9854194583937018,
428
+ "learning_rate": 3.8198992844414736e-06,
429
+ "loss": 1.2935,
430
+ "step": 290
431
+ },
432
+ {
433
+ "epoch": 0.3633004926108374,
434
+ "grad_norm": 0.9611577984967152,
435
+ "learning_rate": 3.7763718932449322e-06,
436
+ "loss": 1.3332,
437
+ "step": 295
438
+ },
439
+ {
440
+ "epoch": 0.3694581280788177,
441
+ "grad_norm": 1.0605919781420619,
442
+ "learning_rate": 3.7323147249165255e-06,
443
+ "loss": 1.3076,
444
+ "step": 300
445
+ },
446
+ {
447
+ "epoch": 0.37561576354679804,
448
+ "grad_norm": 0.9263868172490802,
449
+ "learning_rate": 3.6877460660390257e-06,
450
+ "loss": 1.309,
451
+ "step": 305
452
+ },
453
+ {
454
+ "epoch": 0.3817733990147783,
455
+ "grad_norm": 0.9572326588734382,
456
+ "learning_rate": 3.6426844154969654e-06,
457
+ "loss": 1.3042,
458
+ "step": 310
459
+ },
460
+ {
461
+ "epoch": 0.3879310344827586,
462
+ "grad_norm": 0.9676601659166285,
463
+ "learning_rate": 3.597148476798393e-06,
464
+ "loss": 1.3091,
465
+ "step": 315
466
+ },
467
+ {
468
+ "epoch": 0.39408866995073893,
469
+ "grad_norm": 0.9053200553500108,
470
+ "learning_rate": 3.5511571503117125e-06,
471
+ "loss": 1.3234,
472
+ "step": 320
473
+ },
474
+ {
475
+ "epoch": 0.4002463054187192,
476
+ "grad_norm": 0.8972007834782274,
477
+ "learning_rate": 3.504729525420798e-06,
478
+ "loss": 1.2932,
479
+ "step": 325
480
+ },
481
+ {
482
+ "epoch": 0.4064039408866995,
483
+ "grad_norm": 0.8726544346292117,
484
+ "learning_rate": 3.4578848726016646e-06,
485
+ "loss": 1.3175,
486
+ "step": 330
487
+ },
488
+ {
489
+ "epoch": 0.4125615763546798,
490
+ "grad_norm": 0.9041763575954129,
491
+ "learning_rate": 3.410642635423967e-06,
492
+ "loss": 1.3192,
493
+ "step": 335
494
+ },
495
+ {
496
+ "epoch": 0.4187192118226601,
497
+ "grad_norm": 0.8556502970513064,
498
+ "learning_rate": 3.3630224224806616e-06,
499
+ "loss": 1.3183,
500
+ "step": 340
501
+ },
502
+ {
503
+ "epoch": 0.4248768472906404,
504
+ "grad_norm": 0.8626177859113207,
505
+ "learning_rate": 3.3150439992491667e-06,
506
+ "loss": 1.3105,
507
+ "step": 345
508
+ },
509
+ {
510
+ "epoch": 0.43103448275862066,
511
+ "grad_norm": 0.7967188042404322,
512
+ "learning_rate": 3.2667272798874038e-06,
513
+ "loss": 1.2897,
514
+ "step": 350
515
+ },
516
+ {
517
+ "epoch": 0.437192118226601,
518
+ "grad_norm": 0.8856403226277132,
519
+ "learning_rate": 3.2180923189681367e-06,
520
+ "loss": 1.2776,
521
+ "step": 355
522
+ },
523
+ {
524
+ "epoch": 0.4433497536945813,
525
+ "grad_norm": 0.9102748822081079,
526
+ "learning_rate": 3.169159303155017e-06,
527
+ "loss": 1.3281,
528
+ "step": 360
529
+ },
530
+ {
531
+ "epoch": 0.44950738916256155,
532
+ "grad_norm": 0.8702706004067514,
533
+ "learning_rate": 3.119948542823812e-06,
534
+ "loss": 1.2999,
535
+ "step": 365
536
+ },
537
+ {
538
+ "epoch": 0.45566502463054187,
539
+ "grad_norm": 0.8391791004394836,
540
+ "learning_rate": 3.0704804636322818e-06,
541
+ "loss": 1.3013,
542
+ "step": 370
543
+ },
544
+ {
545
+ "epoch": 0.4618226600985222,
546
+ "grad_norm": 0.9450784930448292,
547
+ "learning_rate": 3.0207755980422036e-06,
548
+ "loss": 1.3009,
549
+ "step": 375
550
+ },
551
+ {
552
+ "epoch": 0.46798029556650245,
553
+ "grad_norm": 0.8364655680922604,
554
+ "learning_rate": 2.970854576797073e-06,
555
+ "loss": 1.3081,
556
+ "step": 380
557
+ },
558
+ {
559
+ "epoch": 0.47413793103448276,
560
+ "grad_norm": 0.8637495993307496,
561
+ "learning_rate": 2.920738120359e-06,
562
+ "loss": 1.2983,
563
+ "step": 385
564
+ },
565
+ {
566
+ "epoch": 0.4802955665024631,
567
+ "grad_norm": 0.8350897912388913,
568
+ "learning_rate": 2.8704470303083808e-06,
569
+ "loss": 1.3022,
570
+ "step": 390
571
+ },
572
+ {
573
+ "epoch": 0.48645320197044334,
574
+ "grad_norm": 0.8003999962290238,
575
+ "learning_rate": 2.820002180709888e-06,
576
+ "loss": 1.2976,
577
+ "step": 395
578
+ },
579
+ {
580
+ "epoch": 0.49261083743842365,
581
+ "grad_norm": 0.8205844355868652,
582
+ "learning_rate": 2.769424509448379e-06,
583
+ "loss": 1.2806,
584
+ "step": 400
585
+ },
586
+ {
587
+ "epoch": 0.49261083743842365,
588
+ "eval_loss": 1.3155713081359863,
589
+ "eval_runtime": 145.3401,
590
+ "eval_samples_per_second": 159.006,
591
+ "eval_steps_per_second": 2.491,
592
+ "step": 400
593
+ },
594
+ {
595
+ "epoch": 0.4987684729064039,
596
+ "grad_norm": 0.7401769827680776,
597
+ "learning_rate": 2.7187350095383197e-06,
598
+ "loss": 1.2898,
599
+ "step": 405
600
+ },
601
+ {
602
+ "epoch": 0.5049261083743842,
603
+ "grad_norm": 0.7932407776274949,
604
+ "learning_rate": 2.6679547204103174e-06,
605
+ "loss": 1.2932,
606
+ "step": 410
607
+ },
608
+ {
609
+ "epoch": 0.5110837438423645,
610
+ "grad_norm": 0.746144620956253,
611
+ "learning_rate": 2.617104719178395e-06,
612
+ "loss": 1.2883,
613
+ "step": 415
614
+ },
615
+ {
616
+ "epoch": 0.5172413793103449,
617
+ "grad_norm": 0.8276505969533728,
618
+ "learning_rate": 2.566206111891621e-06,
619
+ "loss": 1.3198,
620
+ "step": 420
621
+ },
622
+ {
623
+ "epoch": 0.5233990147783252,
624
+ "grad_norm": 0.7534012683391147,
625
+ "learning_rate": 2.5152800247737293e-06,
626
+ "loss": 1.3157,
627
+ "step": 425
628
+ },
629
+ {
630
+ "epoch": 0.5295566502463054,
631
+ "grad_norm": 0.7409765493416938,
632
+ "learning_rate": 2.4643475954543657e-06,
633
+ "loss": 1.2899,
634
+ "step": 430
635
+ },
636
+ {
637
+ "epoch": 0.5357142857142857,
638
+ "grad_norm": 0.7291610138959701,
639
+ "learning_rate": 2.413429964195603e-06,
640
+ "loss": 1.3,
641
+ "step": 435
642
+ },
643
+ {
644
+ "epoch": 0.541871921182266,
645
+ "grad_norm": 0.8169273699015402,
646
+ "learning_rate": 2.362548265117355e-06,
647
+ "loss": 1.309,
648
+ "step": 440
649
+ },
650
+ {
651
+ "epoch": 0.5480295566502463,
652
+ "grad_norm": 0.7758419122658194,
653
+ "learning_rate": 2.3117236174253556e-06,
654
+ "loss": 1.2992,
655
+ "step": 445
656
+ },
657
+ {
658
+ "epoch": 0.5541871921182266,
659
+ "grad_norm": 0.8040315305923249,
660
+ "learning_rate": 2.260977116645306e-06,
661
+ "loss": 1.2885,
662
+ "step": 450
663
+ },
664
+ {
665
+ "epoch": 0.5603448275862069,
666
+ "grad_norm": 0.7548655355086679,
667
+ "learning_rate": 2.2103298258668755e-06,
668
+ "loss": 1.3115,
669
+ "step": 455
670
+ },
671
+ {
672
+ "epoch": 0.5665024630541872,
673
+ "grad_norm": 0.7357816983542891,
674
+ "learning_rate": 2.1598027670011427e-06,
675
+ "loss": 1.3013,
676
+ "step": 460
677
+ },
678
+ {
679
+ "epoch": 0.5726600985221675,
680
+ "grad_norm": 0.7152977794562079,
681
+ "learning_rate": 2.109416912055145e-06,
682
+ "loss": 1.2661,
683
+ "step": 465
684
+ },
685
+ {
686
+ "epoch": 0.5788177339901478,
687
+ "grad_norm": 0.7449902853108209,
688
+ "learning_rate": 2.0591931744271343e-06,
689
+ "loss": 1.29,
690
+ "step": 470
691
+ },
692
+ {
693
+ "epoch": 0.5849753694581281,
694
+ "grad_norm": 0.6971165726597337,
695
+ "learning_rate": 2.009152400226159e-06,
696
+ "loss": 1.3252,
697
+ "step": 475
698
+ },
699
+ {
700
+ "epoch": 0.5911330049261084,
701
+ "grad_norm": 0.707778592285853,
702
+ "learning_rate": 1.959315359619575e-06,
703
+ "loss": 1.3133,
704
+ "step": 480
705
+ },
706
+ {
707
+ "epoch": 0.5972906403940886,
708
+ "grad_norm": 0.7260676752365064,
709
+ "learning_rate": 1.909702738212082e-06,
710
+ "loss": 1.2915,
711
+ "step": 485
712
+ },
713
+ {
714
+ "epoch": 0.603448275862069,
715
+ "grad_norm": 0.7872640671088219,
716
+ "learning_rate": 1.860335128459853e-06,
717
+ "loss": 1.2664,
718
+ "step": 490
719
+ },
720
+ {
721
+ "epoch": 0.6096059113300493,
722
+ "grad_norm": 0.7299570473994874,
723
+ "learning_rate": 1.8112330211233345e-06,
724
+ "loss": 1.3039,
725
+ "step": 495
726
+ },
727
+ {
728
+ "epoch": 0.6157635467980296,
729
+ "grad_norm": 0.6628505663907466,
730
+ "learning_rate": 1.7624167967622435e-06,
731
+ "loss": 1.2999,
732
+ "step": 500
733
+ },
734
+ {
735
+ "epoch": 0.6219211822660099,
736
+ "grad_norm": 0.7138940590397398,
737
+ "learning_rate": 1.7139067172763182e-06,
738
+ "loss": 1.2828,
739
+ "step": 505
740
+ },
741
+ {
742
+ "epoch": 0.6280788177339901,
743
+ "grad_norm": 0.7291370506888436,
744
+ "learning_rate": 1.6657229174953165e-06,
745
+ "loss": 1.2994,
746
+ "step": 510
747
+ },
748
+ {
749
+ "epoch": 0.6342364532019704,
750
+ "grad_norm": 0.7164829111988928,
751
+ "learning_rate": 1.6178853968217507e-06,
752
+ "loss": 1.2856,
753
+ "step": 515
754
+ },
755
+ {
756
+ "epoch": 0.6403940886699507,
757
+ "grad_norm": 0.7191968032340825,
758
+ "learning_rate": 1.5704140109298446e-06,
759
+ "loss": 1.2698,
760
+ "step": 520
761
+ },
762
+ {
763
+ "epoch": 0.646551724137931,
764
+ "grad_norm": 0.6251465309812664,
765
+ "learning_rate": 1.5233284635241335e-06,
766
+ "loss": 1.2933,
767
+ "step": 525
768
+ },
769
+ {
770
+ "epoch": 0.6527093596059114,
771
+ "grad_norm": 0.6568883358981559,
772
+ "learning_rate": 1.4766482981611538e-06,
773
+ "loss": 1.329,
774
+ "step": 530
775
+ },
776
+ {
777
+ "epoch": 0.6588669950738916,
778
+ "grad_norm": 0.6556665876586427,
779
+ "learning_rate": 1.4303928901375975e-06,
780
+ "loss": 1.3021,
781
+ "step": 535
782
+ },
783
+ {
784
+ "epoch": 0.6650246305418719,
785
+ "grad_norm": 0.7081746689649916,
786
+ "learning_rate": 1.3845814384483069e-06,
787
+ "loss": 1.2891,
788
+ "step": 540
789
+ },
790
+ {
791
+ "epoch": 0.6711822660098522,
792
+ "grad_norm": 0.701781476378995,
793
+ "learning_rate": 1.3392329578174495e-06,
794
+ "loss": 1.2794,
795
+ "step": 545
796
+ },
797
+ {
798
+ "epoch": 0.6773399014778325,
799
+ "grad_norm": 0.6824594950934834,
800
+ "learning_rate": 1.2943662708061677e-06,
801
+ "loss": 1.2819,
802
+ "step": 550
803
+ },
804
+ {
805
+ "epoch": 0.6834975369458128,
806
+ "grad_norm": 0.6822805531132496,
807
+ "learning_rate": 1.2500000000000007e-06,
808
+ "loss": 1.2888,
809
+ "step": 555
810
+ },
811
+ {
812
+ "epoch": 0.6896551724137931,
813
+ "grad_norm": 0.7121381471617909,
814
+ "learning_rate": 1.2061525602792994e-06,
815
+ "loss": 1.298,
816
+ "step": 560
817
+ },
818
+ {
819
+ "epoch": 0.6958128078817734,
820
+ "grad_norm": 0.7156140232396447,
821
+ "learning_rate": 1.1628421511758623e-06,
822
+ "loss": 1.2879,
823
+ "step": 565
824
+ },
825
+ {
826
+ "epoch": 0.7019704433497537,
827
+ "grad_norm": 0.6633266964663513,
828
+ "learning_rate": 1.1200867493189418e-06,
829
+ "loss": 1.2756,
830
+ "step": 570
831
+ },
832
+ {
833
+ "epoch": 0.708128078817734,
834
+ "grad_norm": 0.7027669823672923,
835
+ "learning_rate": 1.0779041009737814e-06,
836
+ "loss": 1.2593,
837
+ "step": 575
838
+ },
839
+ {
840
+ "epoch": 0.7142857142857143,
841
+ "grad_norm": 0.6782260773686245,
842
+ "learning_rate": 1.036311714675766e-06,
843
+ "loss": 1.2915,
844
+ "step": 580
845
+ },
846
+ {
847
+ "epoch": 0.7204433497536946,
848
+ "grad_norm": 0.6853410422742611,
849
+ "learning_rate": 9.953268539632373e-07,
850
+ "loss": 1.2555,
851
+ "step": 585
852
+ },
853
+ {
854
+ "epoch": 0.7266009852216748,
855
+ "grad_norm": 0.6562363208181954,
856
+ "learning_rate": 9.549665302120146e-07,
857
+ "loss": 1.3124,
858
+ "step": 590
859
+ },
860
+ {
861
+ "epoch": 0.7327586206896551,
862
+ "grad_norm": 0.6576311894970724,
863
+ "learning_rate": 9.152474955745622e-07,
864
+ "loss": 1.2861,
865
+ "step": 595
866
+ },
867
+ {
868
+ "epoch": 0.7389162561576355,
869
+ "grad_norm": 0.6870118556621342,
870
+ "learning_rate": 8.761862360267637e-07,
871
+ "loss": 1.2827,
872
+ "step": 600
873
+ },
874
+ {
875
+ "epoch": 0.7389162561576355,
876
+ "eval_loss": 1.2938079833984375,
877
+ "eval_runtime": 145.3232,
878
+ "eval_samples_per_second": 159.025,
879
+ "eval_steps_per_second": 2.491,
880
+ "step": 600
881
+ },
882
+ {
883
+ "epoch": 0.7450738916256158,
884
+ "grad_norm": 0.6499954525297986,
885
+ "learning_rate": 8.377989645251719e-07,
886
+ "loss": 1.2891,
887
+ "step": 605
888
+ },
889
+ {
890
+ "epoch": 0.7512315270935961,
891
+ "grad_norm": 0.670520502123996,
892
+ "learning_rate": 8.001016142775788e-07,
893
+ "loss": 1.2926,
894
+ "step": 610
895
+ },
896
+ {
897
+ "epoch": 0.7573891625615764,
898
+ "grad_norm": 0.6283287368331439,
899
+ "learning_rate": 7.631098321297043e-07,
900
+ "loss": 1.2568,
901
+ "step": 615
902
+ },
903
+ {
904
+ "epoch": 0.7635467980295566,
905
+ "grad_norm": 0.6284975248447452,
906
+ "learning_rate": 7.26838972070737e-07,
907
+ "loss": 1.2775,
908
+ "step": 620
909
+ },
910
+ {
911
+ "epoch": 0.7697044334975369,
912
+ "grad_norm": 0.660893437889715,
913
+ "learning_rate": 6.913040888604319e-07,
914
+ "loss": 1.2773,
915
+ "step": 625
916
+ },
917
+ {
918
+ "epoch": 0.7758620689655172,
919
+ "grad_norm": 0.6534843100770891,
920
+ "learning_rate": 6.565199317804119e-07,
921
+ "loss": 1.3018,
922
+ "step": 630
923
+ },
924
+ {
925
+ "epoch": 0.7820197044334976,
926
+ "grad_norm": 0.5870111446293268,
927
+ "learning_rate": 6.22500938512256e-07,
928
+ "loss": 1.2605,
929
+ "step": 635
930
+ },
931
+ {
932
+ "epoch": 0.7881773399014779,
933
+ "grad_norm": 0.6650501710983314,
934
+ "learning_rate": 5.892612291449285e-07,
935
+ "loss": 1.2832,
936
+ "step": 640
937
+ },
938
+ {
939
+ "epoch": 0.7943349753694581,
940
+ "grad_norm": 0.6867995221706822,
941
+ "learning_rate": 5.568146003140224e-07,
942
+ "loss": 1.28,
943
+ "step": 645
944
+ },
945
+ {
946
+ "epoch": 0.8004926108374384,
947
+ "grad_norm": 0.5896681713601163,
948
+ "learning_rate": 5.251745194752622e-07,
949
+ "loss": 1.2587,
950
+ "step": 650
951
+ },
952
+ {
953
+ "epoch": 0.8066502463054187,
954
+ "grad_norm": 0.6717699104805078,
955
+ "learning_rate": 4.943541193146359e-07,
956
+ "loss": 1.2477,
957
+ "step": 655
958
+ },
959
+ {
960
+ "epoch": 0.812807881773399,
961
+ "grad_norm": 0.7052198718541833,
962
+ "learning_rate": 4.6436619229748043e-07,
963
+ "loss": 1.2522,
964
+ "step": 660
965
+ },
966
+ {
967
+ "epoch": 0.8189655172413793,
968
+ "grad_norm": 0.6158705019186339,
969
+ "learning_rate": 4.352231853587763e-07,
970
+ "loss": 1.256,
971
+ "step": 665
972
+ },
973
+ {
974
+ "epoch": 0.8251231527093597,
975
+ "grad_norm": 0.6548358452202809,
976
+ "learning_rate": 4.069371947368619e-07,
977
+ "loss": 1.2963,
978
+ "step": 670
979
+ },
980
+ {
981
+ "epoch": 0.8312807881773399,
982
+ "grad_norm": 0.6535540639408235,
983
+ "learning_rate": 3.795199609527117e-07,
984
+ "loss": 1.2823,
985
+ "step": 675
986
+ },
987
+ {
988
+ "epoch": 0.8374384236453202,
989
+ "grad_norm": 0.6989573890022164,
990
+ "learning_rate": 3.529828639368568e-07,
991
+ "loss": 1.2919,
992
+ "step": 680
993
+ },
994
+ {
995
+ "epoch": 0.8435960591133005,
996
+ "grad_norm": 0.6428403916541138,
997
+ "learning_rate": 3.273369183059782e-07,
998
+ "loss": 1.2985,
999
+ "step": 685
1000
+ },
1001
+ {
1002
+ "epoch": 0.8497536945812808,
1003
+ "grad_norm": 0.6934651562859432,
1004
+ "learning_rate": 3.025927687911229e-07,
1005
+ "loss": 1.3011,
1006
+ "step": 690
1007
+ },
1008
+ {
1009
+ "epoch": 0.8559113300492611,
1010
+ "grad_norm": 0.6591962271180641,
1011
+ "learning_rate": 2.7876068581945195e-07,
1012
+ "loss": 1.265,
1013
+ "step": 695
1014
+ },
1015
+ {
1016
+ "epoch": 0.8620689655172413,
1017
+ "grad_norm": 0.590195214936619,
1018
+ "learning_rate": 2.558505612513479e-07,
1019
+ "loss": 1.2809,
1020
+ "step": 700
1021
+ },
1022
+ {
1023
+ "epoch": 0.8682266009852216,
1024
+ "grad_norm": 0.6115453386704671,
1025
+ "learning_rate": 2.3387190427464817e-07,
1026
+ "loss": 1.2664,
1027
+ "step": 705
1028
+ },
1029
+ {
1030
+ "epoch": 0.874384236453202,
1031
+ "grad_norm": 0.7043010144385191,
1032
+ "learning_rate": 2.1283383745771853e-07,
1033
+ "loss": 1.2946,
1034
+ "step": 710
1035
+ },
1036
+ {
1037
+ "epoch": 0.8805418719211823,
1038
+ "grad_norm": 0.6609378479773281,
1039
+ "learning_rate": 1.9274509296299315e-07,
1040
+ "loss": 1.2976,
1041
+ "step": 715
1042
+ },
1043
+ {
1044
+ "epoch": 0.8866995073891626,
1045
+ "grad_norm": 0.6582696629201034,
1046
+ "learning_rate": 1.736140089225613e-07,
1047
+ "loss": 1.2638,
1048
+ "step": 720
1049
+ },
1050
+ {
1051
+ "epoch": 0.8928571428571429,
1052
+ "grad_norm": 0.6835787335143644,
1053
+ "learning_rate": 1.554485259773017e-07,
1054
+ "loss": 1.2738,
1055
+ "step": 725
1056
+ },
1057
+ {
1058
+ "epoch": 0.8990147783251231,
1059
+ "grad_norm": 0.662487948050227,
1060
+ "learning_rate": 1.3825618398100232e-07,
1061
+ "loss": 1.28,
1062
+ "step": 730
1063
+ },
1064
+ {
1065
+ "epoch": 0.9051724137931034,
1066
+ "grad_norm": 0.687432804005716,
1067
+ "learning_rate": 1.2204411887083072e-07,
1068
+ "loss": 1.2783,
1069
+ "step": 735
1070
+ },
1071
+ {
1072
+ "epoch": 0.9113300492610837,
1073
+ "grad_norm": 0.7021729312071676,
1074
+ "learning_rate": 1.068190597054583e-07,
1075
+ "loss": 1.2964,
1076
+ "step": 740
1077
+ },
1078
+ {
1079
+ "epoch": 0.9174876847290641,
1080
+ "grad_norm": 0.6515658253201944,
1081
+ "learning_rate": 9.25873258720636e-08,
1082
+ "loss": 1.2762,
1083
+ "step": 745
1084
+ },
1085
+ {
1086
+ "epoch": 0.9236453201970444,
1087
+ "grad_norm": 0.6550075584930779,
1088
+ "learning_rate": 7.935482446337628e-08,
1089
+ "loss": 1.2812,
1090
+ "step": 750
1091
+ },
1092
+ {
1093
+ "epoch": 0.9298029556650246,
1094
+ "grad_norm": 0.6848726797074379,
1095
+ "learning_rate": 6.712704782585206e-08,
1096
+ "loss": 1.2735,
1097
+ "step": 755
1098
+ },
1099
+ {
1100
+ "epoch": 0.9359605911330049,
1101
+ "grad_norm": 0.6302612182103497,
1102
+ "learning_rate": 5.590907127999173e-08,
1103
+ "loss": 1.2659,
1104
+ "step": 760
1105
+ },
1106
+ {
1107
+ "epoch": 0.9421182266009852,
1108
+ "grad_norm": 0.6297453401962552,
1109
+ "learning_rate": 4.570555101375357e-08,
1110
+ "loss": 1.2616,
1111
+ "step": 765
1112
+ },
1113
+ {
1114
+ "epoch": 0.9482758620689655,
1115
+ "grad_norm": 0.6875738096430432,
1116
+ "learning_rate": 3.652072214993335e-08,
1117
+ "loss": 1.2823,
1118
+ "step": 770
1119
+ },
1120
+ {
1121
+ "epoch": 0.9544334975369458,
1122
+ "grad_norm": 0.6691771389169094,
1123
+ "learning_rate": 2.8358396988315272e-08,
1124
+ "loss": 1.2823,
1125
+ "step": 775
1126
+ },
1127
+ {
1128
+ "epoch": 0.9605911330049262,
1129
+ "grad_norm": 0.6162495535352837,
1130
+ "learning_rate": 2.122196342331767e-08,
1131
+ "loss": 1.2605,
1132
+ "step": 780
1133
+ },
1134
+ {
1135
+ "epoch": 0.9667487684729064,
1136
+ "grad_norm": 0.6765789960335756,
1137
+ "learning_rate": 1.511438353779898e-08,
1138
+ "loss": 1.2758,
1139
+ "step": 785
1140
+ },
1141
+ {
1142
+ "epoch": 0.9729064039408867,
1143
+ "grad_norm": 0.6926059474276152,
1144
+ "learning_rate": 1.0038192373600652e-08,
1145
+ "loss": 1.2591,
1146
+ "step": 790
1147
+ },
1148
+ {
1149
+ "epoch": 0.979064039408867,
1150
+ "grad_norm": 0.6413165173071386,
1151
+ "learning_rate": 5.995496879339924e-09,
1152
+ "loss": 1.2875,
1153
+ "step": 795
1154
+ },
1155
+ {
1156
+ "epoch": 0.9852216748768473,
1157
+ "grad_norm": 0.6372683334165604,
1158
+ "learning_rate": 2.9879750358896565e-09,
1159
+ "loss": 1.2807,
1160
+ "step": 800
1161
+ },
1162
+ {
1163
+ "epoch": 0.9852216748768473,
1164
+ "eval_loss": 1.2902166843414307,
1165
+ "eval_runtime": 145.3653,
1166
+ "eval_samples_per_second": 158.979,
1167
+ "eval_steps_per_second": 2.49,
1168
+ "step": 800
1169
+ },
1170
+ {
1171
+ "epoch": 0.9913793103448276,
1172
+ "grad_norm": 0.5963019194626616,
1173
+ "learning_rate": 1.0168751599085035e-09,
1174
+ "loss": 1.2782,
1175
+ "step": 805
1176
+ },
1177
+ {
1178
+ "epoch": 0.9975369458128078,
1179
+ "grad_norm": 0.6176142023053375,
1180
+ "learning_rate": 8.301538570676393e-11,
1181
+ "loss": 1.2582,
1182
+ "step": 810
1183
+ },
1184
+ {
1185
+ "epoch": 1.0,
1186
+ "step": 812,
1187
+ "total_flos": 277556883980288.0,
1188
+ "train_loss": 1.458759990231744,
1189
+ "train_runtime": 5428.2383,
1190
+ "train_samples_per_second": 38.293,
1191
+ "train_steps_per_second": 0.15
1192
+ }
1193
+ ],
1194
+ "logging_steps": 5,
1195
+ "max_steps": 812,
1196
+ "num_input_tokens_seen": 0,
1197
+ "num_train_epochs": 1,
1198
+ "save_steps": 1000000,
1199
+ "stateful_callbacks": {
1200
+ "TrainerControl": {
1201
+ "args": {
1202
+ "should_epoch_stop": false,
1203
+ "should_evaluate": false,
1204
+ "should_log": false,
1205
+ "should_save": true,
1206
+ "should_training_stop": true
1207
+ },
1208
+ "attributes": {}
1209
+ }
1210
+ },
1211
+ "total_flos": 277556883980288.0,
1212
+ "train_batch_size": 8,
1213
+ "trial_name": null,
1214
+ "trial_params": null
1215
+ }