chansung commited on
Commit
b548f90
·
verified ·
1 Parent(s): 5055b99

Model save

Browse files
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: google/gemma-7b
3
+ datasets:
4
+ - generator
5
+ library_name: peft
6
+ license: gemma
7
+ tags:
8
+ - trl
9
+ - sft
10
+ - generated_from_trainer
11
+ model-index:
12
+ - name: gemma7b-gpt4o_100k_summarize-lora
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # gemma7b-gpt4o_100k_summarize-lora
20
+
21
+ This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 2.3887
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0003
43
+ - train_batch_size: 1
44
+ - eval_batch_size: 1
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 2
49
+ - total_train_batch_size: 16
50
+ - total_eval_batch_size: 8
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:-----:|:----:|:---------------:|
60
+ | 1.01 | 1.0 | 784 | 2.3887 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.13.0
66
+ - Transformers 4.45.1
67
+ - Pytorch 2.4.1+cu121
68
+ - Datasets 3.0.1
69
+ - Tokenizers 0.20.0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 5.976388653306348e+17,
4
+ "train_loss": 2.8103741881798725,
5
+ "train_runtime": 4265.0133,
6
+ "train_samples": 115376,
7
+ "train_samples_per_second": 2.94,
8
+ "train_steps_per_second": 0.184
9
+ }
runs/Sep29_13-28-30_5bc1778c8f32/events.out.tfevents.1727616560.5bc1778c8f32.8403.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:102d4f8e93d2431e5ea88cb3d91f7a853098cca72791789e740e221df0703d5b
3
- size 39729
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a98abc25bf14a651064d965f68fe04b6f7cceeea55be5fdc5ac802e9cbec66f6
3
+ size 40354
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 5.976388653306348e+17,
4
+ "train_loss": 2.8103741881798725,
5
+ "train_runtime": 4265.0133,
6
+ "train_samples": 115376,
7
+ "train_samples_per_second": 2.94,
8
+ "train_steps_per_second": 0.184
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1149 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 784,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0012755102040816326,
13
+ "grad_norm": 226.4242401123047,
14
+ "learning_rate": 3.7974683544303794e-06,
15
+ "loss": 48.9032,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.006377551020408163,
20
+ "grad_norm": 208.31333923339844,
21
+ "learning_rate": 1.89873417721519e-05,
22
+ "loss": 48.5429,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.012755102040816327,
27
+ "grad_norm": 113.34983825683594,
28
+ "learning_rate": 3.79746835443038e-05,
29
+ "loss": 39.6906,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.01913265306122449,
34
+ "grad_norm": 62.39444351196289,
35
+ "learning_rate": 5.696202531645569e-05,
36
+ "loss": 31.8222,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.025510204081632654,
41
+ "grad_norm": 16.312618255615234,
42
+ "learning_rate": 7.59493670886076e-05,
43
+ "loss": 24.5018,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.03188775510204082,
48
+ "grad_norm": 11.158844947814941,
49
+ "learning_rate": 9.493670886075949e-05,
50
+ "loss": 21.551,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.03826530612244898,
55
+ "grad_norm": 7.093152046203613,
56
+ "learning_rate": 0.00011392405063291139,
57
+ "loss": 20.5881,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.044642857142857144,
62
+ "grad_norm": 4.10010290145874,
63
+ "learning_rate": 0.00013291139240506327,
64
+ "loss": 19.0191,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.05102040816326531,
69
+ "grad_norm": 5.54206657409668,
70
+ "learning_rate": 0.0001518987341772152,
71
+ "loss": 18.1907,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.05739795918367347,
76
+ "grad_norm": 10.086193084716797,
77
+ "learning_rate": 0.00017088607594936708,
78
+ "loss": 17.5312,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.06377551020408163,
83
+ "grad_norm": 19.117507934570312,
84
+ "learning_rate": 0.00018987341772151899,
85
+ "loss": 15.1473,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.07015306122448979,
90
+ "grad_norm": 30.6151180267334,
91
+ "learning_rate": 0.0002088607594936709,
92
+ "loss": 11.3497,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.07653061224489796,
97
+ "grad_norm": 21.005615234375,
98
+ "learning_rate": 0.00022784810126582277,
99
+ "loss": 6.1773,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.08290816326530612,
104
+ "grad_norm": 11.67768383026123,
105
+ "learning_rate": 0.0002468354430379747,
106
+ "loss": 3.0207,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.08928571428571429,
111
+ "grad_norm": 6.371170520782471,
112
+ "learning_rate": 0.00026582278481012653,
113
+ "loss": 2.478,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.09566326530612244,
118
+ "grad_norm": 2.4198906421661377,
119
+ "learning_rate": 0.00028481012658227844,
120
+ "loss": 2.1055,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.10204081632653061,
125
+ "grad_norm": 1.157726764678955,
126
+ "learning_rate": 0.00029999851070045734,
127
+ "loss": 1.8032,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.10841836734693877,
132
+ "grad_norm": 1.53150475025177,
133
+ "learning_rate": 0.00029994638832161753,
134
+ "loss": 1.6379,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.11479591836734694,
139
+ "grad_norm": 4.24367618560791,
140
+ "learning_rate": 0.0002998198305365145,
141
+ "loss": 1.5582,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.1211734693877551,
146
+ "grad_norm": 0.8776500821113586,
147
+ "learning_rate": 0.000299618900170137,
148
+ "loss": 1.5178,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.12755102040816327,
153
+ "grad_norm": 1.7082425355911255,
154
+ "learning_rate": 0.00029934369696702433,
155
+ "loss": 1.4593,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.13392857142857142,
160
+ "grad_norm": 2.545419931411743,
161
+ "learning_rate": 0.0002989943575417525,
162
+ "loss": 1.4022,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.14030612244897958,
167
+ "grad_norm": 0.9073415994644165,
168
+ "learning_rate": 0.0002985710553111161,
169
+ "loss": 1.391,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.14668367346938777,
174
+ "grad_norm": 0.7289413809776306,
175
+ "learning_rate": 0.0002980740004080424,
176
+ "loss": 1.3611,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.15306122448979592,
181
+ "grad_norm": 0.6918051838874817,
182
+ "learning_rate": 0.0002975034395772784,
183
+ "loss": 1.3474,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.15943877551020408,
188
+ "grad_norm": 1.0997295379638672,
189
+ "learning_rate": 0.00029685965605290337,
190
+ "loss": 1.3001,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.16581632653061223,
195
+ "grad_norm": 4.529347896575928,
196
+ "learning_rate": 0.0002961429694177276,
197
+ "loss": 1.3039,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.17219387755102042,
202
+ "grad_norm": 2.3405444622039795,
203
+ "learning_rate": 0.0002953537354446474,
204
+ "loss": 1.2888,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.17857142857142858,
209
+ "grad_norm": 2.904238224029541,
210
+ "learning_rate": 0.00029449234592003436,
211
+ "loss": 1.2796,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.18494897959183673,
216
+ "grad_norm": 1.2537798881530762,
217
+ "learning_rate": 0.0002935592284492476,
218
+ "loss": 1.2941,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.1913265306122449,
223
+ "grad_norm": 1.00596284866333,
224
+ "learning_rate": 0.0002925548462443644,
225
+ "loss": 1.2674,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.19770408163265307,
230
+ "grad_norm": 1.229261040687561,
231
+ "learning_rate": 0.0002914796978942358,
232
+ "loss": 1.2493,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.20408163265306123,
237
+ "grad_norm": 5.567780494689941,
238
+ "learning_rate": 0.0002903343171169803,
239
+ "loss": 1.2494,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.21045918367346939,
244
+ "grad_norm": 2.096792221069336,
245
+ "learning_rate": 0.00028911927249503864,
246
+ "loss": 1.2534,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.21683673469387754,
251
+ "grad_norm": 2.6790685653686523,
252
+ "learning_rate": 0.0002878351671929224,
253
+ "loss": 1.2544,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.22321428571428573,
258
+ "grad_norm": 1.817202091217041,
259
+ "learning_rate": 0.0002864826386577947,
260
+ "loss": 1.2283,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.22959183673469388,
265
+ "grad_norm": 1.452048420906067,
266
+ "learning_rate": 0.00028506235830303274,
267
+ "loss": 1.2416,
268
+ "step": 180
269
+ },
270
+ {
271
+ "epoch": 0.23596938775510204,
272
+ "grad_norm": 2.810974359512329,
273
+ "learning_rate": 0.00028357503117492944,
274
+ "loss": 1.2204,
275
+ "step": 185
276
+ },
277
+ {
278
+ "epoch": 0.2423469387755102,
279
+ "grad_norm": 0.8295219540596008,
280
+ "learning_rate": 0.0002820213956026987,
281
+ "loss": 1.2267,
282
+ "step": 190
283
+ },
284
+ {
285
+ "epoch": 0.24872448979591838,
286
+ "grad_norm": 1.347506046295166,
287
+ "learning_rate": 0.0002804022228319591,
288
+ "loss": 1.2054,
289
+ "step": 195
290
+ },
291
+ {
292
+ "epoch": 0.25510204081632654,
293
+ "grad_norm": 1.5803207159042358,
294
+ "learning_rate": 0.000278718316641877,
295
+ "loss": 1.1688,
296
+ "step": 200
297
+ },
298
+ {
299
+ "epoch": 0.2614795918367347,
300
+ "grad_norm": 0.6454965472221375,
301
+ "learning_rate": 0.00027697051294615964,
302
+ "loss": 1.2095,
303
+ "step": 205
304
+ },
305
+ {
306
+ "epoch": 0.26785714285714285,
307
+ "grad_norm": 0.8256592154502869,
308
+ "learning_rate": 0.0002751596793780968,
309
+ "loss": 1.1808,
310
+ "step": 210
311
+ },
312
+ {
313
+ "epoch": 0.274234693877551,
314
+ "grad_norm": 1.8195074796676636,
315
+ "learning_rate": 0.00027328671485985515,
316
+ "loss": 1.1963,
317
+ "step": 215
318
+ },
319
+ {
320
+ "epoch": 0.28061224489795916,
321
+ "grad_norm": 1.3873289823532104,
322
+ "learning_rate": 0.0002713525491562421,
323
+ "loss": 1.1795,
324
+ "step": 220
325
+ },
326
+ {
327
+ "epoch": 0.2869897959183674,
328
+ "grad_norm": 2.0349373817443848,
329
+ "learning_rate": 0.00026935814241315736,
330
+ "loss": 1.1802,
331
+ "step": 225
332
+ },
333
+ {
334
+ "epoch": 0.29336734693877553,
335
+ "grad_norm": 1.6414597034454346,
336
+ "learning_rate": 0.00026730448468096377,
337
+ "loss": 1.173,
338
+ "step": 230
339
+ },
340
+ {
341
+ "epoch": 0.2997448979591837,
342
+ "grad_norm": 1.2609707117080688,
343
+ "learning_rate": 0.00026519259542301317,
344
+ "loss": 1.1561,
345
+ "step": 235
346
+ },
347
+ {
348
+ "epoch": 0.30612244897959184,
349
+ "grad_norm": 4.516489028930664,
350
+ "learning_rate": 0.0002630235230095708,
351
+ "loss": 1.1573,
352
+ "step": 240
353
+ },
354
+ {
355
+ "epoch": 0.3125,
356
+ "grad_norm": 1.8223940134048462,
357
+ "learning_rate": 0.0002607983441973903,
358
+ "loss": 1.1619,
359
+ "step": 245
360
+ },
361
+ {
362
+ "epoch": 0.31887755102040816,
363
+ "grad_norm": 1.7611159086227417,
364
+ "learning_rate": 0.0002585181635951971,
365
+ "loss": 1.1654,
366
+ "step": 250
367
+ },
368
+ {
369
+ "epoch": 0.3252551020408163,
370
+ "grad_norm": 1.7671295404434204,
371
+ "learning_rate": 0.0002561841131153459,
372
+ "loss": 1.1701,
373
+ "step": 255
374
+ },
375
+ {
376
+ "epoch": 0.33163265306122447,
377
+ "grad_norm": 1.1895421743392944,
378
+ "learning_rate": 0.00025379735141192397,
379
+ "loss": 1.1517,
380
+ "step": 260
381
+ },
382
+ {
383
+ "epoch": 0.3380102040816326,
384
+ "grad_norm": 2.26123046875,
385
+ "learning_rate": 0.0002513590633055798,
386
+ "loss": 1.1304,
387
+ "step": 265
388
+ },
389
+ {
390
+ "epoch": 0.34438775510204084,
391
+ "grad_norm": 2.0483951568603516,
392
+ "learning_rate": 0.00024887045919536207,
393
+ "loss": 1.1403,
394
+ "step": 270
395
+ },
396
+ {
397
+ "epoch": 0.350765306122449,
398
+ "grad_norm": 1.611738681793213,
399
+ "learning_rate": 0.00024633277445786143,
400
+ "loss": 1.1499,
401
+ "step": 275
402
+ },
403
+ {
404
+ "epoch": 0.35714285714285715,
405
+ "grad_norm": 2.4385955333709717,
406
+ "learning_rate": 0.00024374726883395273,
407
+ "loss": 1.1424,
408
+ "step": 280
409
+ },
410
+ {
411
+ "epoch": 0.3635204081632653,
412
+ "grad_norm": 1.5102245807647705,
413
+ "learning_rate": 0.00024111522580344265,
414
+ "loss": 1.1135,
415
+ "step": 285
416
+ },
417
+ {
418
+ "epoch": 0.36989795918367346,
419
+ "grad_norm": 0.8719022870063782,
420
+ "learning_rate": 0.00023843795194793336,
421
+ "loss": 1.1236,
422
+ "step": 290
423
+ },
424
+ {
425
+ "epoch": 0.3762755102040816,
426
+ "grad_norm": 0.9534187316894531,
427
+ "learning_rate": 0.0002357167763022175,
428
+ "loss": 1.1253,
429
+ "step": 295
430
+ },
431
+ {
432
+ "epoch": 0.3826530612244898,
433
+ "grad_norm": 1.8618340492248535,
434
+ "learning_rate": 0.00023295304969452748,
435
+ "loss": 1.1296,
436
+ "step": 300
437
+ },
438
+ {
439
+ "epoch": 0.38903061224489793,
440
+ "grad_norm": 0.7363711595535278,
441
+ "learning_rate": 0.00023014814407596628,
442
+ "loss": 1.1298,
443
+ "step": 305
444
+ },
445
+ {
446
+ "epoch": 0.39540816326530615,
447
+ "grad_norm": 8.200729370117188,
448
+ "learning_rate": 0.00022730345183945248,
449
+ "loss": 1.1155,
450
+ "step": 310
451
+ },
452
+ {
453
+ "epoch": 0.4017857142857143,
454
+ "grad_norm": 1.8119416236877441,
455
+ "learning_rate": 0.0002244203851285173,
456
+ "loss": 1.1161,
457
+ "step": 315
458
+ },
459
+ {
460
+ "epoch": 0.40816326530612246,
461
+ "grad_norm": 3.4057934284210205,
462
+ "learning_rate": 0.000221500375136298,
463
+ "loss": 1.1257,
464
+ "step": 320
465
+ },
466
+ {
467
+ "epoch": 0.4145408163265306,
468
+ "grad_norm": 0.9975801706314087,
469
+ "learning_rate": 0.00021854487139507398,
470
+ "loss": 1.1356,
471
+ "step": 325
472
+ },
473
+ {
474
+ "epoch": 0.42091836734693877,
475
+ "grad_norm": 1.9435389041900635,
476
+ "learning_rate": 0.00021555534105669976,
477
+ "loss": 1.1142,
478
+ "step": 330
479
+ },
480
+ {
481
+ "epoch": 0.4272959183673469,
482
+ "grad_norm": 1.204506516456604,
483
+ "learning_rate": 0.00021253326816429076,
484
+ "loss": 1.1174,
485
+ "step": 335
486
+ },
487
+ {
488
+ "epoch": 0.4336734693877551,
489
+ "grad_norm": 2.5877513885498047,
490
+ "learning_rate": 0.0002094801529155244,
491
+ "loss": 1.0873,
492
+ "step": 340
493
+ },
494
+ {
495
+ "epoch": 0.44005102040816324,
496
+ "grad_norm": 0.7226624488830566,
497
+ "learning_rate": 0.00020639751091792188,
498
+ "loss": 1.1018,
499
+ "step": 345
500
+ },
501
+ {
502
+ "epoch": 0.44642857142857145,
503
+ "grad_norm": 1.762408971786499,
504
+ "learning_rate": 0.00020328687243648,
505
+ "loss": 1.1056,
506
+ "step": 350
507
+ },
508
+ {
509
+ "epoch": 0.4528061224489796,
510
+ "grad_norm": 4.760770320892334,
511
+ "learning_rate": 0.00020014978163402725,
512
+ "loss": 1.0828,
513
+ "step": 355
514
+ },
515
+ {
516
+ "epoch": 0.45918367346938777,
517
+ "grad_norm": 0.723813533782959,
518
+ "learning_rate": 0.0001969877958046804,
519
+ "loss": 1.1015,
520
+ "step": 360
521
+ },
522
+ {
523
+ "epoch": 0.4655612244897959,
524
+ "grad_norm": 1.1869585514068604,
525
+ "learning_rate": 0.0001938024846007828,
526
+ "loss": 1.0955,
527
+ "step": 365
528
+ },
529
+ {
530
+ "epoch": 0.4719387755102041,
531
+ "grad_norm": 0.7096392512321472,
532
+ "learning_rate": 0.00019059542925370783,
533
+ "loss": 1.0949,
534
+ "step": 370
535
+ },
536
+ {
537
+ "epoch": 0.47831632653061223,
538
+ "grad_norm": 2.3162615299224854,
539
+ "learning_rate": 0.0001873682217889143,
540
+ "loss": 1.0943,
541
+ "step": 375
542
+ },
543
+ {
544
+ "epoch": 0.4846938775510204,
545
+ "grad_norm": 0.7215905785560608,
546
+ "learning_rate": 0.0001841224642356436,
547
+ "loss": 1.0957,
548
+ "step": 380
549
+ },
550
+ {
551
+ "epoch": 0.49107142857142855,
552
+ "grad_norm": 1.3136285543441772,
553
+ "learning_rate": 0.0001808597678316509,
554
+ "loss": 1.0865,
555
+ "step": 385
556
+ },
557
+ {
558
+ "epoch": 0.49744897959183676,
559
+ "grad_norm": 2.323671817779541,
560
+ "learning_rate": 0.00017758175222336496,
561
+ "loss": 1.0752,
562
+ "step": 390
563
+ },
564
+ {
565
+ "epoch": 0.5038265306122449,
566
+ "grad_norm": 1.1084301471710205,
567
+ "learning_rate": 0.00017429004466187377,
568
+ "loss": 1.0892,
569
+ "step": 395
570
+ },
571
+ {
572
+ "epoch": 0.5102040816326531,
573
+ "grad_norm": 2.052607297897339,
574
+ "learning_rate": 0.00017098627919513528,
575
+ "loss": 1.1036,
576
+ "step": 400
577
+ },
578
+ {
579
+ "epoch": 0.5165816326530612,
580
+ "grad_norm": 1.0028424263000488,
581
+ "learning_rate": 0.00016767209585681394,
582
+ "loss": 1.0913,
583
+ "step": 405
584
+ },
585
+ {
586
+ "epoch": 0.5229591836734694,
587
+ "grad_norm": 0.9611196517944336,
588
+ "learning_rate": 0.00016434913985214597,
589
+ "loss": 1.0825,
590
+ "step": 410
591
+ },
592
+ {
593
+ "epoch": 0.5293367346938775,
594
+ "grad_norm": 2.9150185585021973,
595
+ "learning_rate": 0.0001610190607412373,
596
+ "loss": 1.085,
597
+ "step": 415
598
+ },
599
+ {
600
+ "epoch": 0.5357142857142857,
601
+ "grad_norm": 2.879117250442505,
602
+ "learning_rate": 0.00015768351162019975,
603
+ "loss": 1.0889,
604
+ "step": 420
605
+ },
606
+ {
607
+ "epoch": 0.5420918367346939,
608
+ "grad_norm": 1.0958130359649658,
609
+ "learning_rate": 0.0001543441483005318,
610
+ "loss": 1.0833,
611
+ "step": 425
612
+ },
613
+ {
614
+ "epoch": 0.548469387755102,
615
+ "grad_norm": 0.9718190431594849,
616
+ "learning_rate": 0.00015100262848715133,
617
+ "loss": 1.0813,
618
+ "step": 430
619
+ },
620
+ {
621
+ "epoch": 0.5548469387755102,
622
+ "grad_norm": 1.2847260236740112,
623
+ "learning_rate": 0.000147660610955489,
624
+ "loss": 1.0684,
625
+ "step": 435
626
+ },
627
+ {
628
+ "epoch": 0.5612244897959183,
629
+ "grad_norm": 1.039225459098816,
630
+ "learning_rate": 0.00014431975472804936,
631
+ "loss": 1.0797,
632
+ "step": 440
633
+ },
634
+ {
635
+ "epoch": 0.5676020408163265,
636
+ "grad_norm": 1.2738494873046875,
637
+ "learning_rate": 0.00014098171825084995,
638
+ "loss": 1.0839,
639
+ "step": 445
640
+ },
641
+ {
642
+ "epoch": 0.5739795918367347,
643
+ "grad_norm": 1.06371009349823,
644
+ "learning_rate": 0.00013764815857014636,
645
+ "loss": 1.0649,
646
+ "step": 450
647
+ },
648
+ {
649
+ "epoch": 0.5803571428571429,
650
+ "grad_norm": 1.394904375076294,
651
+ "learning_rate": 0.000134320730509852,
652
+ "loss": 1.0798,
653
+ "step": 455
654
+ },
655
+ {
656
+ "epoch": 0.5867346938775511,
657
+ "grad_norm": 0.6322318315505981,
658
+ "learning_rate": 0.00013100108585006146,
659
+ "loss": 1.0702,
660
+ "step": 460
661
+ },
662
+ {
663
+ "epoch": 0.5931122448979592,
664
+ "grad_norm": 1.5734529495239258,
665
+ "learning_rate": 0.0001276908725070845,
666
+ "loss": 1.0571,
667
+ "step": 465
668
+ },
669
+ {
670
+ "epoch": 0.5994897959183674,
671
+ "grad_norm": 1.5619932413101196,
672
+ "learning_rate": 0.00012439173371539818,
673
+ "loss": 1.0714,
674
+ "step": 470
675
+ },
676
+ {
677
+ "epoch": 0.6058673469387755,
678
+ "grad_norm": 1.8657833337783813,
679
+ "learning_rate": 0.00012110530721192319,
680
+ "loss": 1.0727,
681
+ "step": 475
682
+ },
683
+ {
684
+ "epoch": 0.6122448979591837,
685
+ "grad_norm": 2.461625337600708,
686
+ "learning_rate": 0.00011783322442302915,
687
+ "loss": 1.0472,
688
+ "step": 480
689
+ },
690
+ {
691
+ "epoch": 0.6186224489795918,
692
+ "grad_norm": 1.1668888330459595,
693
+ "learning_rate": 0.00011457710965467266,
694
+ "loss": 1.0482,
695
+ "step": 485
696
+ },
697
+ {
698
+ "epoch": 0.625,
699
+ "grad_norm": 1.11457359790802,
700
+ "learning_rate": 0.00011133857928606992,
701
+ "loss": 1.0618,
702
+ "step": 490
703
+ },
704
+ {
705
+ "epoch": 0.6313775510204082,
706
+ "grad_norm": 1.9202951192855835,
707
+ "learning_rate": 0.00010811924096730416,
708
+ "loss": 1.0494,
709
+ "step": 495
710
+ },
711
+ {
712
+ "epoch": 0.6377551020408163,
713
+ "grad_norm": 0.7689357995986938,
714
+ "learning_rate": 0.00010492069282126685,
715
+ "loss": 1.0443,
716
+ "step": 500
717
+ },
718
+ {
719
+ "epoch": 0.6441326530612245,
720
+ "grad_norm": 2.5137813091278076,
721
+ "learning_rate": 0.00010174452265032769,
722
+ "loss": 1.0689,
723
+ "step": 505
724
+ },
725
+ {
726
+ "epoch": 0.6505102040816326,
727
+ "grad_norm": 0.6587682962417603,
728
+ "learning_rate": 9.859230714812812e-05,
729
+ "loss": 1.0482,
730
+ "step": 510
731
+ },
732
+ {
733
+ "epoch": 0.6568877551020408,
734
+ "grad_norm": 1.251999020576477,
735
+ "learning_rate": 9.546561111688958e-05,
736
+ "loss": 1.0439,
737
+ "step": 515
738
+ },
739
+ {
740
+ "epoch": 0.6632653061224489,
741
+ "grad_norm": 0.9404116272926331,
742
+ "learning_rate": 9.236598669062443e-05,
743
+ "loss": 1.042,
744
+ "step": 520
745
+ },
746
+ {
747
+ "epoch": 0.6696428571428571,
748
+ "grad_norm": 2.4428858757019043,
749
+ "learning_rate": 8.929497256463586e-05,
750
+ "loss": 1.0399,
751
+ "step": 525
752
+ },
753
+ {
754
+ "epoch": 0.6760204081632653,
755
+ "grad_norm": 0.9080251455307007,
756
+ "learning_rate": 8.625409323168858e-05,
757
+ "loss": 1.0655,
758
+ "step": 530
759
+ },
760
+ {
761
+ "epoch": 0.6823979591836735,
762
+ "grad_norm": 0.9034568071365356,
763
+ "learning_rate": 8.32448582252305e-05,
764
+ "loss": 1.0342,
765
+ "step": 535
766
+ },
767
+ {
768
+ "epoch": 0.6887755102040817,
769
+ "grad_norm": 0.7829791307449341,
770
+ "learning_rate": 8.026876137003956e-05,
771
+ "loss": 1.0468,
772
+ "step": 540
773
+ },
774
+ {
775
+ "epoch": 0.6951530612244898,
776
+ "grad_norm": 0.9003449082374573,
777
+ "learning_rate": 7.732728004066927e-05,
778
+ "loss": 1.0516,
779
+ "step": 545
780
+ },
781
+ {
782
+ "epoch": 0.701530612244898,
783
+ "grad_norm": 1.5128637552261353,
784
+ "learning_rate": 7.442187442805997e-05,
785
+ "loss": 1.0486,
786
+ "step": 550
787
+ },
788
+ {
789
+ "epoch": 0.7079081632653061,
790
+ "grad_norm": 0.843595027923584,
791
+ "learning_rate": 7.155398681468041e-05,
792
+ "loss": 1.035,
793
+ "step": 555
794
+ },
795
+ {
796
+ "epoch": 0.7142857142857143,
797
+ "grad_norm": 1.120501160621643,
798
+ "learning_rate": 6.872504085855947e-05,
799
+ "loss": 1.0468,
800
+ "step": 560
801
+ },
802
+ {
803
+ "epoch": 0.7206632653061225,
804
+ "grad_norm": 0.6691803932189941,
805
+ "learning_rate": 6.593644088656277e-05,
806
+ "loss": 1.032,
807
+ "step": 565
808
+ },
809
+ {
810
+ "epoch": 0.7270408163265306,
811
+ "grad_norm": 1.066981315612793,
812
+ "learning_rate": 6.318957119726636e-05,
813
+ "loss": 1.0366,
814
+ "step": 570
815
+ },
816
+ {
817
+ "epoch": 0.7334183673469388,
818
+ "grad_norm": 0.9314639568328857,
819
+ "learning_rate": 6.048579537377189e-05,
820
+ "loss": 1.0258,
821
+ "step": 575
822
+ },
823
+ {
824
+ "epoch": 0.7397959183673469,
825
+ "grad_norm": 2.091320753097534,
826
+ "learning_rate": 5.782645560680571e-05,
827
+ "loss": 1.0328,
828
+ "step": 580
829
+ },
830
+ {
831
+ "epoch": 0.7461734693877551,
832
+ "grad_norm": 0.6987316012382507,
833
+ "learning_rate": 5.5212872028437564e-05,
834
+ "loss": 1.0588,
835
+ "step": 585
836
+ },
837
+ {
838
+ "epoch": 0.7525510204081632,
839
+ "grad_norm": 0.9516999125480652,
840
+ "learning_rate": 5.264634205674884e-05,
841
+ "loss": 1.036,
842
+ "step": 590
843
+ },
844
+ {
845
+ "epoch": 0.7589285714285714,
846
+ "grad_norm": 0.9213479161262512,
847
+ "learning_rate": 5.012813975177708e-05,
848
+ "loss": 1.0275,
849
+ "step": 595
850
+ },
851
+ {
852
+ "epoch": 0.7653061224489796,
853
+ "grad_norm": 1.1120542287826538,
854
+ "learning_rate": 4.765951518305524e-05,
855
+ "loss": 1.0286,
856
+ "step": 600
857
+ },
858
+ {
859
+ "epoch": 0.7716836734693877,
860
+ "grad_norm": 1.146470308303833,
861
+ "learning_rate": 4.524169380906042e-05,
862
+ "loss": 1.0306,
863
+ "step": 605
864
+ },
865
+ {
866
+ "epoch": 0.7780612244897959,
867
+ "grad_norm": 0.7507238388061523,
868
+ "learning_rate": 4.287587586887991e-05,
869
+ "loss": 1.0358,
870
+ "step": 610
871
+ },
872
+ {
873
+ "epoch": 0.7844387755102041,
874
+ "grad_norm": 0.7958881258964539,
875
+ "learning_rate": 4.056323578639599e-05,
876
+ "loss": 1.0522,
877
+ "step": 615
878
+ },
879
+ {
880
+ "epoch": 0.7908163265306123,
881
+ "grad_norm": 1.0853825807571411,
882
+ "learning_rate": 3.83049215872865e-05,
883
+ "loss": 1.0318,
884
+ "step": 620
885
+ },
886
+ {
887
+ "epoch": 0.7971938775510204,
888
+ "grad_norm": 0.6897711157798767,
889
+ "learning_rate": 3.610205432912914e-05,
890
+ "loss": 1.0253,
891
+ "step": 625
892
+ },
893
+ {
894
+ "epoch": 0.8035714285714286,
895
+ "grad_norm": 1.08404541015625,
896
+ "learning_rate": 3.395572754489349e-05,
897
+ "loss": 1.0536,
898
+ "step": 630
899
+ },
900
+ {
901
+ "epoch": 0.8099489795918368,
902
+ "grad_norm": 0.9480798244476318,
903
+ "learning_rate": 3.1867006700096625e-05,
904
+ "loss": 1.0363,
905
+ "step": 635
906
+ },
907
+ {
908
+ "epoch": 0.8163265306122449,
909
+ "grad_norm": 1.013339877128601,
910
+ "learning_rate": 2.9836928663891517e-05,
911
+ "loss": 1.0286,
912
+ "step": 640
913
+ },
914
+ {
915
+ "epoch": 0.8227040816326531,
916
+ "grad_norm": 0.8233193755149841,
917
+ "learning_rate": 2.7866501194351253e-05,
918
+ "loss": 1.047,
919
+ "step": 645
920
+ },
921
+ {
922
+ "epoch": 0.8290816326530612,
923
+ "grad_norm": 0.8080177903175354,
924
+ "learning_rate": 2.595670243820461e-05,
925
+ "loss": 1.0383,
926
+ "step": 650
927
+ },
928
+ {
929
+ "epoch": 0.8354591836734694,
930
+ "grad_norm": 0.6761140823364258,
931
+ "learning_rate": 2.4108480445270346e-05,
932
+ "loss": 1.032,
933
+ "step": 655
934
+ },
935
+ {
936
+ "epoch": 0.8418367346938775,
937
+ "grad_norm": 1.0132781267166138,
938
+ "learning_rate": 2.2322752697833045e-05,
939
+ "loss": 1.0381,
940
+ "step": 660
941
+ },
942
+ {
943
+ "epoch": 0.8482142857142857,
944
+ "grad_norm": 2.3679418563842773,
945
+ "learning_rate": 2.0600405655192166e-05,
946
+ "loss": 1.035,
947
+ "step": 665
948
+ },
949
+ {
950
+ "epoch": 0.8545918367346939,
951
+ "grad_norm": 1.0122100114822388,
952
+ "learning_rate": 1.8942294313611974e-05,
953
+ "loss": 1.0523,
954
+ "step": 670
955
+ },
956
+ {
957
+ "epoch": 0.860969387755102,
958
+ "grad_norm": 0.797031819820404,
959
+ "learning_rate": 1.7349241781889654e-05,
960
+ "loss": 1.0225,
961
+ "step": 675
962
+ },
963
+ {
964
+ "epoch": 0.8673469387755102,
965
+ "grad_norm": 0.6979250907897949,
966
+ "learning_rate": 1.5822038872753166e-05,
967
+ "loss": 1.023,
968
+ "step": 680
969
+ },
970
+ {
971
+ "epoch": 0.8737244897959183,
972
+ "grad_norm": 0.5788841843605042,
973
+ "learning_rate": 1.4361443710291149e-05,
974
+ "loss": 1.0388,
975
+ "step": 685
976
+ },
977
+ {
978
+ "epoch": 0.8801020408163265,
979
+ "grad_norm": 0.7020506858825684,
980
+ "learning_rate": 1.2968181353609852e-05,
981
+ "loss": 1.0148,
982
+ "step": 690
983
+ },
984
+ {
985
+ "epoch": 0.8864795918367347,
986
+ "grad_norm": 0.6345797181129456,
987
+ "learning_rate": 1.1642943436904111e-05,
988
+ "loss": 1.0269,
989
+ "step": 695
990
+ },
991
+ {
992
+ "epoch": 0.8928571428571429,
993
+ "grad_norm": 0.7500460147857666,
994
+ "learning_rate": 1.038638782612073e-05,
995
+ "loss": 1.0271,
996
+ "step": 700
997
+ },
998
+ {
999
+ "epoch": 0.8992346938775511,
1000
+ "grad_norm": 0.6654278635978699,
1001
+ "learning_rate": 9.199138292385006e-06,
1002
+ "loss": 1.0241,
1003
+ "step": 705
1004
+ },
1005
+ {
1006
+ "epoch": 0.9056122448979592,
1007
+ "grad_norm": 0.6786679029464722,
1008
+ "learning_rate": 8.081784202352404e-06,
1009
+ "loss": 1.0301,
1010
+ "step": 710
1011
+ },
1012
+ {
1013
+ "epoch": 0.9119897959183674,
1014
+ "grad_norm": 0.712835967540741,
1015
+ "learning_rate": 7.034880225638878e-06,
1016
+ "loss": 1.0355,
1017
+ "step": 715
1018
+ },
1019
+ {
1020
+ "epoch": 0.9183673469387755,
1021
+ "grad_norm": 0.691426694393158,
1022
+ "learning_rate": 6.058946059475595e-06,
1023
+ "loss": 1.0374,
1024
+ "step": 720
1025
+ },
1026
+ {
1027
+ "epoch": 0.9247448979591837,
1028
+ "grad_norm": 0.6497547626495361,
1029
+ "learning_rate": 5.154466170724031e-06,
1030
+ "loss": 1.03,
1031
+ "step": 725
1032
+ },
1033
+ {
1034
+ "epoch": 0.9311224489795918,
1035
+ "grad_norm": 0.7832013964653015,
1036
+ "learning_rate": 4.32188955538022e-06,
1037
+ "loss": 1.0222,
1038
+ "step": 730
1039
+ },
1040
+ {
1041
+ "epoch": 0.9375,
1042
+ "grad_norm": 1.1313321590423584,
1043
+ "learning_rate": 3.5616295156870754e-06,
1044
+ "loss": 1.0005,
1045
+ "step": 735
1046
+ },
1047
+ {
1048
+ "epoch": 0.9438775510204082,
1049
+ "grad_norm": 0.7574149966239929,
1050
+ "learning_rate": 2.874063454965558e-06,
1051
+ "loss": 1.0293,
1052
+ "step": 740
1053
+ },
1054
+ {
1055
+ "epoch": 0.9502551020408163,
1056
+ "grad_norm": 0.6553170680999756,
1057
+ "learning_rate": 2.2595326902666412e-06,
1058
+ "loss": 1.027,
1059
+ "step": 745
1060
+ },
1061
+ {
1062
+ "epoch": 0.9566326530612245,
1063
+ "grad_norm": 0.7936610579490662,
1064
+ "learning_rate": 1.7183422829368409e-06,
1065
+ "loss": 1.0301,
1066
+ "step": 750
1067
+ },
1068
+ {
1069
+ "epoch": 0.9630102040816326,
1070
+ "grad_norm": 0.6665956974029541,
1071
+ "learning_rate": 1.2507608871817566e-06,
1072
+ "loss": 1.0299,
1073
+ "step": 755
1074
+ },
1075
+ {
1076
+ "epoch": 0.9693877551020408,
1077
+ "grad_norm": 0.6638586521148682,
1078
+ "learning_rate": 8.570206167024807e-07,
1079
+ "loss": 1.0108,
1080
+ "step": 760
1081
+ },
1082
+ {
1083
+ "epoch": 0.9757653061224489,
1084
+ "grad_norm": 0.6911438703536987,
1085
+ "learning_rate": 5.37316929471121e-07,
1086
+ "loss": 1.0219,
1087
+ "step": 765
1088
+ },
1089
+ {
1090
+ "epoch": 0.9821428571428571,
1091
+ "grad_norm": 0.6120124459266663,
1092
+ "learning_rate": 2.9180853070303044e-07,
1093
+ "loss": 1.0279,
1094
+ "step": 770
1095
+ },
1096
+ {
1097
+ "epoch": 0.9885204081632653,
1098
+ "grad_norm": 0.6324964165687561,
1099
+ "learning_rate": 1.2061729407324416e-07,
1100
+ "loss": 1.0428,
1101
+ "step": 775
1102
+ },
1103
+ {
1104
+ "epoch": 0.9948979591836735,
1105
+ "grad_norm": 0.5991183519363403,
1106
+ "learning_rate": 2.3828201216774533e-08,
1107
+ "loss": 1.01,
1108
+ "step": 780
1109
+ },
1110
+ {
1111
+ "epoch": 1.0,
1112
+ "eval_loss": 2.388667106628418,
1113
+ "eval_runtime": 1.6706,
1114
+ "eval_samples_per_second": 5.986,
1115
+ "eval_steps_per_second": 1.197,
1116
+ "step": 784
1117
+ },
1118
+ {
1119
+ "epoch": 1.0,
1120
+ "step": 784,
1121
+ "total_flos": 5.976388653306348e+17,
1122
+ "train_loss": 2.8103741881798725,
1123
+ "train_runtime": 4265.0133,
1124
+ "train_samples_per_second": 2.94,
1125
+ "train_steps_per_second": 0.184
1126
+ }
1127
+ ],
1128
+ "logging_steps": 5,
1129
+ "max_steps": 784,
1130
+ "num_input_tokens_seen": 0,
1131
+ "num_train_epochs": 1,
1132
+ "save_steps": 100,
1133
+ "stateful_callbacks": {
1134
+ "TrainerControl": {
1135
+ "args": {
1136
+ "should_epoch_stop": false,
1137
+ "should_evaluate": false,
1138
+ "should_log": false,
1139
+ "should_save": true,
1140
+ "should_training_stop": true
1141
+ },
1142
+ "attributes": {}
1143
+ }
1144
+ },
1145
+ "total_flos": 5.976388653306348e+17,
1146
+ "train_batch_size": 1,
1147
+ "trial_name": null,
1148
+ "trial_params": null
1149
+ }