gokulsrinivasagan commited on
Commit
bc548b6
1 Parent(s): 1c5ef95

End of training

Browse files
README.md CHANGED
@@ -3,9 +3,23 @@ license: mit
3
  base_model: openai-community/gpt2
4
  tags:
5
  - generated_from_trainer
 
 
 
 
6
  model-index:
7
  - name: gpt_train_2_768_new
8
- results: []
 
 
 
 
 
 
 
 
 
 
9
  ---
10
 
11
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -13,7 +27,10 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  # gpt_train_2_768_new
15
 
16
- This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on an unknown dataset.
 
 
 
17
 
18
  ## Model description
19
 
 
3
  base_model: openai-community/gpt2
4
  tags:
5
  - generated_from_trainer
6
+ datasets:
7
+ - gokuls/wiki_book_corpus_raw_dataset_tiny
8
+ metrics:
9
+ - accuracy
10
  model-index:
11
  - name: gpt_train_2_768_new
12
+ results:
13
+ - task:
14
+ name: Causal Language Modeling
15
+ type: text-generation
16
+ dataset:
17
+ name: gokuls/wiki_book_corpus_raw_dataset_tiny
18
+ type: gokuls/wiki_book_corpus_raw_dataset_tiny
19
+ metrics:
20
+ - name: Accuracy
21
+ type: accuracy
22
+ value: 0.3345458864628659
23
  ---
24
 
25
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
27
 
28
  # gpt_train_2_768_new
29
 
30
+ This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on the gokuls/wiki_book_corpus_raw_dataset_tiny dataset.
31
+ It achieves the following results on the evaluation set:
32
+ - Loss: 3.9456
33
+ - Accuracy: 0.3345
34
 
35
  ## Model description
36
 
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.321127579192096,
3
+ "eval_accuracy": 0.3345458864628659,
4
+ "eval_loss": 3.945629119873047,
5
+ "eval_runtime": 233.188,
6
+ "eval_samples": 33767,
7
+ "eval_samples_per_second": 144.806,
8
+ "eval_steps_per_second": 1.51,
9
+ "perplexity": 51.70885895206038,
10
+ "total_flos": 5.363892569191219e+17,
11
+ "train_loss": 4.763968430426513,
12
+ "train_runtime": 93599.1484,
13
+ "train_samples": 660643,
14
+ "train_samples_per_second": 705.822,
15
+ "train_steps_per_second": 7.353
16
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.321127579192096,
3
+ "eval_accuracy": 0.3345458864628659,
4
+ "eval_loss": 3.945629119873047,
5
+ "eval_runtime": 233.188,
6
+ "eval_samples": 33767,
7
+ "eval_samples_per_second": 144.806,
8
+ "eval_steps_per_second": 1.51,
9
+ "perplexity": 51.70885895206038
10
+ }
logs/events.out.tfevents.1720019581.s_003_m.242502.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9e54bd39e5890b8512d1de62d32b836ffb20f55180a5bc6bc1b53432cd78069
3
+ size 369
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.321127579192096,
3
+ "total_flos": 5.363892569191219e+17,
4
+ "train_loss": 4.763968430426513,
5
+ "train_runtime": 93599.1484,
6
+ "train_samples": 660643,
7
+ "train_samples_per_second": 705.822,
8
+ "train_steps_per_second": 7.353
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,938 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 9.321127579192096,
5
+ "eval_steps": 1000000,
6
+ "global_step": 64148,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07265329845975008,
13
+ "grad_norm": 0.39395979046821594,
14
+ "learning_rate": 9.992734670154026e-06,
15
+ "loss": 8.1621,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.14530659691950015,
20
+ "grad_norm": 0.5780870318412781,
21
+ "learning_rate": 9.985469340308051e-06,
22
+ "loss": 7.0196,
23
+ "step": 1000
24
+ },
25
+ {
26
+ "epoch": 0.21795989537925023,
27
+ "grad_norm": 0.5540204048156738,
28
+ "learning_rate": 9.978204010462076e-06,
29
+ "loss": 6.6142,
30
+ "step": 1500
31
+ },
32
+ {
33
+ "epoch": 0.2906131938390003,
34
+ "grad_norm": 0.5443109273910522,
35
+ "learning_rate": 9.970938680616102e-06,
36
+ "loss": 6.3813,
37
+ "step": 2000
38
+ },
39
+ {
40
+ "epoch": 0.36326649229875035,
41
+ "grad_norm": 0.6196579337120056,
42
+ "learning_rate": 9.963673350770125e-06,
43
+ "loss": 6.2207,
44
+ "step": 2500
45
+ },
46
+ {
47
+ "epoch": 0.43591979075850046,
48
+ "grad_norm": 0.5789604187011719,
49
+ "learning_rate": 9.95640802092415e-06,
50
+ "loss": 6.0957,
51
+ "step": 3000
52
+ },
53
+ {
54
+ "epoch": 0.5085730892182505,
55
+ "grad_norm": 0.63157719373703,
56
+ "learning_rate": 9.949142691078175e-06,
57
+ "loss": 5.9962,
58
+ "step": 3500
59
+ },
60
+ {
61
+ "epoch": 0.5812263876780006,
62
+ "grad_norm": 0.5619252920150757,
63
+ "learning_rate": 9.9418773612322e-06,
64
+ "loss": 5.9116,
65
+ "step": 4000
66
+ },
67
+ {
68
+ "epoch": 0.6538796861377506,
69
+ "grad_norm": 0.6025997996330261,
70
+ "learning_rate": 9.934612031386226e-06,
71
+ "loss": 5.8332,
72
+ "step": 4500
73
+ },
74
+ {
75
+ "epoch": 0.7265329845975007,
76
+ "grad_norm": 0.6860383152961731,
77
+ "learning_rate": 9.927346701540251e-06,
78
+ "loss": 5.7695,
79
+ "step": 5000
80
+ },
81
+ {
82
+ "epoch": 0.7991862830572508,
83
+ "grad_norm": 0.7735297679901123,
84
+ "learning_rate": 9.920081371694276e-06,
85
+ "loss": 5.7068,
86
+ "step": 5500
87
+ },
88
+ {
89
+ "epoch": 0.8718395815170009,
90
+ "grad_norm": 0.6047775745391846,
91
+ "learning_rate": 9.912816041848301e-06,
92
+ "loss": 5.65,
93
+ "step": 6000
94
+ },
95
+ {
96
+ "epoch": 0.9444928799767509,
97
+ "grad_norm": 0.8214231133460999,
98
+ "learning_rate": 9.905550712002325e-06,
99
+ "loss": 5.6052,
100
+ "step": 6500
101
+ },
102
+ {
103
+ "epoch": 1.017146178436501,
104
+ "grad_norm": 0.8017473816871643,
105
+ "learning_rate": 9.89828538215635e-06,
106
+ "loss": 5.5515,
107
+ "step": 7000
108
+ },
109
+ {
110
+ "epoch": 1.0897994768962511,
111
+ "grad_norm": 0.7803598642349243,
112
+ "learning_rate": 9.891020052310377e-06,
113
+ "loss": 5.5058,
114
+ "step": 7500
115
+ },
116
+ {
117
+ "epoch": 1.1624527753560012,
118
+ "grad_norm": 0.7102193236351013,
119
+ "learning_rate": 9.8837547224644e-06,
120
+ "loss": 5.4684,
121
+ "step": 8000
122
+ },
123
+ {
124
+ "epoch": 1.2351060738157513,
125
+ "grad_norm": 0.7121894359588623,
126
+ "learning_rate": 9.876489392618425e-06,
127
+ "loss": 5.4271,
128
+ "step": 8500
129
+ },
130
+ {
131
+ "epoch": 1.3077593722755014,
132
+ "grad_norm": 0.7547861337661743,
133
+ "learning_rate": 9.86922406277245e-06,
134
+ "loss": 5.397,
135
+ "step": 9000
136
+ },
137
+ {
138
+ "epoch": 1.3804126707352513,
139
+ "grad_norm": 0.7697557210922241,
140
+ "learning_rate": 9.861958732926476e-06,
141
+ "loss": 5.3563,
142
+ "step": 9500
143
+ },
144
+ {
145
+ "epoch": 1.4530659691950014,
146
+ "grad_norm": 0.8195068836212158,
147
+ "learning_rate": 9.854693403080501e-06,
148
+ "loss": 5.3228,
149
+ "step": 10000
150
+ },
151
+ {
152
+ "epoch": 1.5257192676547515,
153
+ "grad_norm": 0.7637478113174438,
154
+ "learning_rate": 9.847428073234524e-06,
155
+ "loss": 5.2946,
156
+ "step": 10500
157
+ },
158
+ {
159
+ "epoch": 1.5983725661145016,
160
+ "grad_norm": 0.8972837328910828,
161
+ "learning_rate": 9.840162743388551e-06,
162
+ "loss": 5.2588,
163
+ "step": 11000
164
+ },
165
+ {
166
+ "epoch": 1.6710258645742515,
167
+ "grad_norm": 0.9061026573181152,
168
+ "learning_rate": 9.832897413542576e-06,
169
+ "loss": 5.2321,
170
+ "step": 11500
171
+ },
172
+ {
173
+ "epoch": 1.7436791630340016,
174
+ "grad_norm": 0.7589295506477356,
175
+ "learning_rate": 9.8256320836966e-06,
176
+ "loss": 5.2065,
177
+ "step": 12000
178
+ },
179
+ {
180
+ "epoch": 1.8163324614937517,
181
+ "grad_norm": 0.8363276720046997,
182
+ "learning_rate": 9.818366753850625e-06,
183
+ "loss": 5.1819,
184
+ "step": 12500
185
+ },
186
+ {
187
+ "epoch": 1.8889857599535018,
188
+ "grad_norm": 0.8336675763130188,
189
+ "learning_rate": 9.81110142400465e-06,
190
+ "loss": 5.1575,
191
+ "step": 13000
192
+ },
193
+ {
194
+ "epoch": 1.961639058413252,
195
+ "grad_norm": 0.8027577996253967,
196
+ "learning_rate": 9.803836094158675e-06,
197
+ "loss": 5.1314,
198
+ "step": 13500
199
+ },
200
+ {
201
+ "epoch": 2.034292356873002,
202
+ "grad_norm": 0.7707503437995911,
203
+ "learning_rate": 9.7965707643127e-06,
204
+ "loss": 5.1088,
205
+ "step": 14000
206
+ },
207
+ {
208
+ "epoch": 2.106945655332752,
209
+ "grad_norm": 0.8998398184776306,
210
+ "learning_rate": 9.789305434466726e-06,
211
+ "loss": 5.0808,
212
+ "step": 14500
213
+ },
214
+ {
215
+ "epoch": 2.1795989537925022,
216
+ "grad_norm": 0.8320822715759277,
217
+ "learning_rate": 9.782040104620751e-06,
218
+ "loss": 5.0634,
219
+ "step": 15000
220
+ },
221
+ {
222
+ "epoch": 2.2522522522522523,
223
+ "grad_norm": 0.8281899094581604,
224
+ "learning_rate": 9.774774774774776e-06,
225
+ "loss": 5.0444,
226
+ "step": 15500
227
+ },
228
+ {
229
+ "epoch": 2.3249055507120024,
230
+ "grad_norm": 0.8026254177093506,
231
+ "learning_rate": 9.767509444928801e-06,
232
+ "loss": 5.0179,
233
+ "step": 16000
234
+ },
235
+ {
236
+ "epoch": 2.3975588491717525,
237
+ "grad_norm": 0.776054322719574,
238
+ "learning_rate": 9.760244115082825e-06,
239
+ "loss": 5.0009,
240
+ "step": 16500
241
+ },
242
+ {
243
+ "epoch": 2.4702121476315027,
244
+ "grad_norm": 0.8994656801223755,
245
+ "learning_rate": 9.75297878523685e-06,
246
+ "loss": 4.9812,
247
+ "step": 17000
248
+ },
249
+ {
250
+ "epoch": 2.5428654460912528,
251
+ "grad_norm": 0.8583637475967407,
252
+ "learning_rate": 9.745713455390875e-06,
253
+ "loss": 4.9628,
254
+ "step": 17500
255
+ },
256
+ {
257
+ "epoch": 2.615518744551003,
258
+ "grad_norm": 0.893622636795044,
259
+ "learning_rate": 9.7384481255449e-06,
260
+ "loss": 4.9464,
261
+ "step": 18000
262
+ },
263
+ {
264
+ "epoch": 2.688172043010753,
265
+ "grad_norm": 0.8372634053230286,
266
+ "learning_rate": 9.731182795698925e-06,
267
+ "loss": 4.9304,
268
+ "step": 18500
269
+ },
270
+ {
271
+ "epoch": 2.7608253414705026,
272
+ "grad_norm": 0.9035200476646423,
273
+ "learning_rate": 9.72391746585295e-06,
274
+ "loss": 4.9127,
275
+ "step": 19000
276
+ },
277
+ {
278
+ "epoch": 2.8334786399302527,
279
+ "grad_norm": 0.8660902380943298,
280
+ "learning_rate": 9.716652136006976e-06,
281
+ "loss": 4.8998,
282
+ "step": 19500
283
+ },
284
+ {
285
+ "epoch": 2.906131938390003,
286
+ "grad_norm": 0.9463688731193542,
287
+ "learning_rate": 9.709386806161001e-06,
288
+ "loss": 4.8778,
289
+ "step": 20000
290
+ },
291
+ {
292
+ "epoch": 2.978785236849753,
293
+ "grad_norm": 0.8671707510948181,
294
+ "learning_rate": 9.702121476315024e-06,
295
+ "loss": 4.8658,
296
+ "step": 20500
297
+ },
298
+ {
299
+ "epoch": 3.051438535309503,
300
+ "grad_norm": 0.9653003215789795,
301
+ "learning_rate": 9.69485614646905e-06,
302
+ "loss": 4.853,
303
+ "step": 21000
304
+ },
305
+ {
306
+ "epoch": 3.124091833769253,
307
+ "grad_norm": 0.8984787464141846,
308
+ "learning_rate": 9.687590816623077e-06,
309
+ "loss": 4.8364,
310
+ "step": 21500
311
+ },
312
+ {
313
+ "epoch": 3.1967451322290033,
314
+ "grad_norm": 0.8803947567939758,
315
+ "learning_rate": 9.6803254867771e-06,
316
+ "loss": 4.8177,
317
+ "step": 22000
318
+ },
319
+ {
320
+ "epoch": 3.2693984306887534,
321
+ "grad_norm": 0.8390781283378601,
322
+ "learning_rate": 9.673060156931125e-06,
323
+ "loss": 4.8077,
324
+ "step": 22500
325
+ },
326
+ {
327
+ "epoch": 3.3420517291485035,
328
+ "grad_norm": 0.9399817585945129,
329
+ "learning_rate": 9.66579482708515e-06,
330
+ "loss": 4.7942,
331
+ "step": 23000
332
+ },
333
+ {
334
+ "epoch": 3.4147050276082536,
335
+ "grad_norm": 0.864647626876831,
336
+ "learning_rate": 9.658529497239176e-06,
337
+ "loss": 4.7819,
338
+ "step": 23500
339
+ },
340
+ {
341
+ "epoch": 3.4873583260680037,
342
+ "grad_norm": 0.863230288028717,
343
+ "learning_rate": 9.6512641673932e-06,
344
+ "loss": 4.7675,
345
+ "step": 24000
346
+ },
347
+ {
348
+ "epoch": 3.5600116245277533,
349
+ "grad_norm": 0.9053711891174316,
350
+ "learning_rate": 9.643998837547224e-06,
351
+ "loss": 4.7562,
352
+ "step": 24500
353
+ },
354
+ {
355
+ "epoch": 3.6326649229875034,
356
+ "grad_norm": 0.9143590331077576,
357
+ "learning_rate": 9.636733507701251e-06,
358
+ "loss": 4.7408,
359
+ "step": 25000
360
+ },
361
+ {
362
+ "epoch": 3.7053182214472535,
363
+ "grad_norm": 0.8819567561149597,
364
+ "learning_rate": 9.629468177855276e-06,
365
+ "loss": 4.7347,
366
+ "step": 25500
367
+ },
368
+ {
369
+ "epoch": 3.7779715199070036,
370
+ "grad_norm": 0.9433513283729553,
371
+ "learning_rate": 9.6222028480093e-06,
372
+ "loss": 4.7227,
373
+ "step": 26000
374
+ },
375
+ {
376
+ "epoch": 3.8506248183667537,
377
+ "grad_norm": 0.9881005883216858,
378
+ "learning_rate": 9.614937518163325e-06,
379
+ "loss": 4.7096,
380
+ "step": 26500
381
+ },
382
+ {
383
+ "epoch": 3.923278116826504,
384
+ "grad_norm": 0.890139639377594,
385
+ "learning_rate": 9.60767218831735e-06,
386
+ "loss": 4.7,
387
+ "step": 27000
388
+ },
389
+ {
390
+ "epoch": 3.995931415286254,
391
+ "grad_norm": 0.9619746804237366,
392
+ "learning_rate": 9.600406858471375e-06,
393
+ "loss": 4.6879,
394
+ "step": 27500
395
+ },
396
+ {
397
+ "epoch": 4.068584713746004,
398
+ "grad_norm": 0.8994712233543396,
399
+ "learning_rate": 9.5931415286254e-06,
400
+ "loss": 4.6762,
401
+ "step": 28000
402
+ },
403
+ {
404
+ "epoch": 4.141238012205754,
405
+ "grad_norm": 0.9636611938476562,
406
+ "learning_rate": 9.585876198779426e-06,
407
+ "loss": 4.6643,
408
+ "step": 28500
409
+ },
410
+ {
411
+ "epoch": 4.213891310665504,
412
+ "grad_norm": 0.8346700072288513,
413
+ "learning_rate": 9.57861086893345e-06,
414
+ "loss": 4.6541,
415
+ "step": 29000
416
+ },
417
+ {
418
+ "epoch": 4.286544609125254,
419
+ "grad_norm": 0.8618379831314087,
420
+ "learning_rate": 9.571345539087476e-06,
421
+ "loss": 4.6473,
422
+ "step": 29500
423
+ },
424
+ {
425
+ "epoch": 4.3591979075850045,
426
+ "grad_norm": 0.9965615272521973,
427
+ "learning_rate": 9.564080209241501e-06,
428
+ "loss": 4.636,
429
+ "step": 30000
430
+ },
431
+ {
432
+ "epoch": 4.431851206044755,
433
+ "grad_norm": 0.9595836400985718,
434
+ "learning_rate": 9.556814879395525e-06,
435
+ "loss": 4.6262,
436
+ "step": 30500
437
+ },
438
+ {
439
+ "epoch": 4.504504504504505,
440
+ "grad_norm": 0.918498694896698,
441
+ "learning_rate": 9.54954954954955e-06,
442
+ "loss": 4.6139,
443
+ "step": 31000
444
+ },
445
+ {
446
+ "epoch": 4.577157802964255,
447
+ "grad_norm": 0.9323708415031433,
448
+ "learning_rate": 9.542284219703575e-06,
449
+ "loss": 4.6138,
450
+ "step": 31500
451
+ },
452
+ {
453
+ "epoch": 4.649811101424005,
454
+ "grad_norm": 0.9415541291236877,
455
+ "learning_rate": 9.5350188898576e-06,
456
+ "loss": 4.6008,
457
+ "step": 32000
458
+ },
459
+ {
460
+ "epoch": 4.722464399883755,
461
+ "grad_norm": 0.9424421787261963,
462
+ "learning_rate": 9.527753560011625e-06,
463
+ "loss": 4.5893,
464
+ "step": 32500
465
+ },
466
+ {
467
+ "epoch": 4.795117698343505,
468
+ "grad_norm": 0.942669689655304,
469
+ "learning_rate": 9.52048823016565e-06,
470
+ "loss": 4.5866,
471
+ "step": 33000
472
+ },
473
+ {
474
+ "epoch": 4.867770996803255,
475
+ "grad_norm": 0.9807618856430054,
476
+ "learning_rate": 9.513222900319676e-06,
477
+ "loss": 4.5736,
478
+ "step": 33500
479
+ },
480
+ {
481
+ "epoch": 4.940424295263005,
482
+ "grad_norm": 0.960895299911499,
483
+ "learning_rate": 9.5059575704737e-06,
484
+ "loss": 4.5619,
485
+ "step": 34000
486
+ },
487
+ {
488
+ "epoch": 5.013077593722755,
489
+ "grad_norm": 0.9801125526428223,
490
+ "learning_rate": 9.498692240627724e-06,
491
+ "loss": 4.5551,
492
+ "step": 34500
493
+ },
494
+ {
495
+ "epoch": 5.0857308921825055,
496
+ "grad_norm": 0.9244375228881836,
497
+ "learning_rate": 9.49142691078175e-06,
498
+ "loss": 4.5408,
499
+ "step": 35000
500
+ },
501
+ {
502
+ "epoch": 5.158384190642255,
503
+ "grad_norm": 0.9349271059036255,
504
+ "learning_rate": 9.484161580935776e-06,
505
+ "loss": 4.5358,
506
+ "step": 35500
507
+ },
508
+ {
509
+ "epoch": 5.231037489102005,
510
+ "grad_norm": 0.9496700763702393,
511
+ "learning_rate": 9.4768962510898e-06,
512
+ "loss": 4.5284,
513
+ "step": 36000
514
+ },
515
+ {
516
+ "epoch": 5.303690787561755,
517
+ "grad_norm": 0.8824469447135925,
518
+ "learning_rate": 9.469630921243825e-06,
519
+ "loss": 4.5225,
520
+ "step": 36500
521
+ },
522
+ {
523
+ "epoch": 5.376344086021505,
524
+ "grad_norm": 0.9746178984642029,
525
+ "learning_rate": 9.46236559139785e-06,
526
+ "loss": 4.5084,
527
+ "step": 37000
528
+ },
529
+ {
530
+ "epoch": 5.448997384481255,
531
+ "grad_norm": 0.9742151498794556,
532
+ "learning_rate": 9.455100261551875e-06,
533
+ "loss": 4.5036,
534
+ "step": 37500
535
+ },
536
+ {
537
+ "epoch": 5.521650682941005,
538
+ "grad_norm": 1.0122525691986084,
539
+ "learning_rate": 9.4478349317059e-06,
540
+ "loss": 4.4951,
541
+ "step": 38000
542
+ },
543
+ {
544
+ "epoch": 5.594303981400755,
545
+ "grad_norm": 0.9454242587089539,
546
+ "learning_rate": 9.440569601859924e-06,
547
+ "loss": 4.4844,
548
+ "step": 38500
549
+ },
550
+ {
551
+ "epoch": 5.6669572798605055,
552
+ "grad_norm": 0.9594370722770691,
553
+ "learning_rate": 9.43330427201395e-06,
554
+ "loss": 4.4776,
555
+ "step": 39000
556
+ },
557
+ {
558
+ "epoch": 5.739610578320256,
559
+ "grad_norm": 0.9644564986228943,
560
+ "learning_rate": 9.426038942167976e-06,
561
+ "loss": 4.4748,
562
+ "step": 39500
563
+ },
564
+ {
565
+ "epoch": 5.812263876780006,
566
+ "grad_norm": 0.9564418792724609,
567
+ "learning_rate": 9.418773612322e-06,
568
+ "loss": 4.4592,
569
+ "step": 40000
570
+ },
571
+ {
572
+ "epoch": 5.884917175239756,
573
+ "grad_norm": 0.9269986152648926,
574
+ "learning_rate": 9.411508282476025e-06,
575
+ "loss": 4.455,
576
+ "step": 40500
577
+ },
578
+ {
579
+ "epoch": 5.957570473699506,
580
+ "grad_norm": 0.9712278842926025,
581
+ "learning_rate": 9.40424295263005e-06,
582
+ "loss": 4.4455,
583
+ "step": 41000
584
+ },
585
+ {
586
+ "epoch": 6.030223772159256,
587
+ "grad_norm": 0.9372689723968506,
588
+ "learning_rate": 9.396977622784075e-06,
589
+ "loss": 4.4318,
590
+ "step": 41500
591
+ },
592
+ {
593
+ "epoch": 6.102877070619006,
594
+ "grad_norm": 0.942115843296051,
595
+ "learning_rate": 9.3897122929381e-06,
596
+ "loss": 4.4264,
597
+ "step": 42000
598
+ },
599
+ {
600
+ "epoch": 6.175530369078756,
601
+ "grad_norm": 1.0541362762451172,
602
+ "learning_rate": 9.382446963092125e-06,
603
+ "loss": 4.4163,
604
+ "step": 42500
605
+ },
606
+ {
607
+ "epoch": 6.248183667538506,
608
+ "grad_norm": 0.9615015387535095,
609
+ "learning_rate": 9.37518163324615e-06,
610
+ "loss": 4.4096,
611
+ "step": 43000
612
+ },
613
+ {
614
+ "epoch": 6.320836965998256,
615
+ "grad_norm": 1.0179786682128906,
616
+ "learning_rate": 9.367916303400176e-06,
617
+ "loss": 4.4034,
618
+ "step": 43500
619
+ },
620
+ {
621
+ "epoch": 6.3934902644580065,
622
+ "grad_norm": 0.9816853404045105,
623
+ "learning_rate": 9.360650973554201e-06,
624
+ "loss": 4.3956,
625
+ "step": 44000
626
+ },
627
+ {
628
+ "epoch": 6.466143562917757,
629
+ "grad_norm": 1.0108842849731445,
630
+ "learning_rate": 9.353385643708224e-06,
631
+ "loss": 4.3811,
632
+ "step": 44500
633
+ },
634
+ {
635
+ "epoch": 6.538796861377507,
636
+ "grad_norm": 1.0273536443710327,
637
+ "learning_rate": 9.34612031386225e-06,
638
+ "loss": 4.3748,
639
+ "step": 45000
640
+ },
641
+ {
642
+ "epoch": 6.611450159837257,
643
+ "grad_norm": 0.9984000325202942,
644
+ "learning_rate": 9.338854984016275e-06,
645
+ "loss": 4.3711,
646
+ "step": 45500
647
+ },
648
+ {
649
+ "epoch": 6.684103458297007,
650
+ "grad_norm": 0.9320149421691895,
651
+ "learning_rate": 9.3315896541703e-06,
652
+ "loss": 4.3621,
653
+ "step": 46000
654
+ },
655
+ {
656
+ "epoch": 6.756756756756757,
657
+ "grad_norm": 1.1042180061340332,
658
+ "learning_rate": 9.324324324324325e-06,
659
+ "loss": 4.3516,
660
+ "step": 46500
661
+ },
662
+ {
663
+ "epoch": 6.829410055216507,
664
+ "grad_norm": 0.9890114665031433,
665
+ "learning_rate": 9.31705899447835e-06,
666
+ "loss": 4.3464,
667
+ "step": 47000
668
+ },
669
+ {
670
+ "epoch": 6.902063353676257,
671
+ "grad_norm": 1.034157156944275,
672
+ "learning_rate": 9.309793664632375e-06,
673
+ "loss": 4.3374,
674
+ "step": 47500
675
+ },
676
+ {
677
+ "epoch": 6.974716652136007,
678
+ "grad_norm": 0.9817051887512207,
679
+ "learning_rate": 9.3025283347864e-06,
680
+ "loss": 4.3284,
681
+ "step": 48000
682
+ },
683
+ {
684
+ "epoch": 7.047369950595757,
685
+ "grad_norm": 1.0558934211730957,
686
+ "learning_rate": 9.295263004940424e-06,
687
+ "loss": 4.3154,
688
+ "step": 48500
689
+ },
690
+ {
691
+ "epoch": 7.1200232490555075,
692
+ "grad_norm": 0.9428159594535828,
693
+ "learning_rate": 9.28799767509445e-06,
694
+ "loss": 4.3088,
695
+ "step": 49000
696
+ },
697
+ {
698
+ "epoch": 7.192676547515257,
699
+ "grad_norm": 1.0073195695877075,
700
+ "learning_rate": 9.280732345248476e-06,
701
+ "loss": 4.2983,
702
+ "step": 49500
703
+ },
704
+ {
705
+ "epoch": 7.265329845975007,
706
+ "grad_norm": 0.9252042174339294,
707
+ "learning_rate": 9.2734670154025e-06,
708
+ "loss": 4.2912,
709
+ "step": 50000
710
+ },
711
+ {
712
+ "epoch": 7.337983144434757,
713
+ "grad_norm": 1.0058454275131226,
714
+ "learning_rate": 9.266201685556525e-06,
715
+ "loss": 4.2814,
716
+ "step": 50500
717
+ },
718
+ {
719
+ "epoch": 7.410636442894507,
720
+ "grad_norm": 1.0393654108047485,
721
+ "learning_rate": 9.25893635571055e-06,
722
+ "loss": 4.2693,
723
+ "step": 51000
724
+ },
725
+ {
726
+ "epoch": 7.483289741354257,
727
+ "grad_norm": 0.9429093599319458,
728
+ "learning_rate": 9.251671025864575e-06,
729
+ "loss": 4.2604,
730
+ "step": 51500
731
+ },
732
+ {
733
+ "epoch": 7.555943039814007,
734
+ "grad_norm": 0.9522086381912231,
735
+ "learning_rate": 9.2444056960186e-06,
736
+ "loss": 4.2499,
737
+ "step": 52000
738
+ },
739
+ {
740
+ "epoch": 7.628596338273757,
741
+ "grad_norm": 0.9854516386985779,
742
+ "learning_rate": 9.237140366172624e-06,
743
+ "loss": 4.2411,
744
+ "step": 52500
745
+ },
746
+ {
747
+ "epoch": 7.7012496367335075,
748
+ "grad_norm": 0.9321109056472778,
749
+ "learning_rate": 9.22987503632665e-06,
750
+ "loss": 4.2257,
751
+ "step": 53000
752
+ },
753
+ {
754
+ "epoch": 7.773902935193258,
755
+ "grad_norm": 0.9426267147064209,
756
+ "learning_rate": 9.222609706480676e-06,
757
+ "loss": 4.2225,
758
+ "step": 53500
759
+ },
760
+ {
761
+ "epoch": 7.846556233653008,
762
+ "grad_norm": 0.9784730672836304,
763
+ "learning_rate": 9.2153443766347e-06,
764
+ "loss": 4.2077,
765
+ "step": 54000
766
+ },
767
+ {
768
+ "epoch": 7.919209532112758,
769
+ "grad_norm": 0.9224970936775208,
770
+ "learning_rate": 9.208079046788724e-06,
771
+ "loss": 4.2029,
772
+ "step": 54500
773
+ },
774
+ {
775
+ "epoch": 7.991862830572508,
776
+ "grad_norm": 0.9461036920547485,
777
+ "learning_rate": 9.20081371694275e-06,
778
+ "loss": 4.1932,
779
+ "step": 55000
780
+ },
781
+ {
782
+ "epoch": 8.064516129032258,
783
+ "grad_norm": 0.9398289322853088,
784
+ "learning_rate": 9.193548387096775e-06,
785
+ "loss": 4.182,
786
+ "step": 55500
787
+ },
788
+ {
789
+ "epoch": 8.137169427492008,
790
+ "grad_norm": 0.9409332275390625,
791
+ "learning_rate": 9.1862830572508e-06,
792
+ "loss": 4.1696,
793
+ "step": 56000
794
+ },
795
+ {
796
+ "epoch": 8.209822725951758,
797
+ "grad_norm": 0.9343535304069519,
798
+ "learning_rate": 9.179017727404825e-06,
799
+ "loss": 4.1627,
800
+ "step": 56500
801
+ },
802
+ {
803
+ "epoch": 8.282476024411508,
804
+ "grad_norm": 0.96622234582901,
805
+ "learning_rate": 9.17175239755885e-06,
806
+ "loss": 4.1569,
807
+ "step": 57000
808
+ },
809
+ {
810
+ "epoch": 8.355129322871258,
811
+ "grad_norm": 0.9742193818092346,
812
+ "learning_rate": 9.164487067712876e-06,
813
+ "loss": 4.1494,
814
+ "step": 57500
815
+ },
816
+ {
817
+ "epoch": 8.427782621331009,
818
+ "grad_norm": 0.9126195907592773,
819
+ "learning_rate": 9.1572217378669e-06,
820
+ "loss": 4.1413,
821
+ "step": 58000
822
+ },
823
+ {
824
+ "epoch": 8.500435919790759,
825
+ "grad_norm": 0.978380024433136,
826
+ "learning_rate": 9.149956408020924e-06,
827
+ "loss": 4.1355,
828
+ "step": 58500
829
+ },
830
+ {
831
+ "epoch": 8.573089218250509,
832
+ "grad_norm": 0.943013608455658,
833
+ "learning_rate": 9.14269107817495e-06,
834
+ "loss": 4.1277,
835
+ "step": 59000
836
+ },
837
+ {
838
+ "epoch": 8.645742516710259,
839
+ "grad_norm": 0.9417549967765808,
840
+ "learning_rate": 9.135425748328975e-06,
841
+ "loss": 4.1254,
842
+ "step": 59500
843
+ },
844
+ {
845
+ "epoch": 8.718395815170009,
846
+ "grad_norm": 0.9427275061607361,
847
+ "learning_rate": 9.128160418483e-06,
848
+ "loss": 4.1183,
849
+ "step": 60000
850
+ },
851
+ {
852
+ "epoch": 8.791049113629759,
853
+ "grad_norm": 0.972618818283081,
854
+ "learning_rate": 9.120895088637025e-06,
855
+ "loss": 4.1094,
856
+ "step": 60500
857
+ },
858
+ {
859
+ "epoch": 8.86370241208951,
860
+ "grad_norm": 0.9143289923667908,
861
+ "learning_rate": 9.11362975879105e-06,
862
+ "loss": 4.1035,
863
+ "step": 61000
864
+ },
865
+ {
866
+ "epoch": 8.93635571054926,
867
+ "grad_norm": 0.961862325668335,
868
+ "learning_rate": 9.106364428945075e-06,
869
+ "loss": 4.1035,
870
+ "step": 61500
871
+ },
872
+ {
873
+ "epoch": 9.00900900900901,
874
+ "grad_norm": 0.9252375960350037,
875
+ "learning_rate": 9.0990990990991e-06,
876
+ "loss": 4.0923,
877
+ "step": 62000
878
+ },
879
+ {
880
+ "epoch": 9.08166230746876,
881
+ "grad_norm": 0.9321468472480774,
882
+ "learning_rate": 9.091833769253124e-06,
883
+ "loss": 4.0855,
884
+ "step": 62500
885
+ },
886
+ {
887
+ "epoch": 9.15431560592851,
888
+ "grad_norm": 0.9290264844894409,
889
+ "learning_rate": 9.084568439407149e-06,
890
+ "loss": 4.0847,
891
+ "step": 63000
892
+ },
893
+ {
894
+ "epoch": 9.22696890438826,
895
+ "grad_norm": 1.0008461475372314,
896
+ "learning_rate": 9.077303109561176e-06,
897
+ "loss": 4.0791,
898
+ "step": 63500
899
+ },
900
+ {
901
+ "epoch": 9.29962220284801,
902
+ "grad_norm": 1.0144729614257812,
903
+ "learning_rate": 9.0700377797152e-06,
904
+ "loss": 4.0724,
905
+ "step": 64000
906
+ },
907
+ {
908
+ "epoch": 9.321127579192096,
909
+ "step": 64148,
910
+ "total_flos": 5.363892569191219e+17,
911
+ "train_loss": 4.763968430426513,
912
+ "train_runtime": 93599.1484,
913
+ "train_samples_per_second": 705.822,
914
+ "train_steps_per_second": 7.353
915
+ }
916
+ ],
917
+ "logging_steps": 500,
918
+ "max_steps": 688200,
919
+ "num_input_tokens_seen": 0,
920
+ "num_train_epochs": 100,
921
+ "save_steps": 1000000,
922
+ "stateful_callbacks": {
923
+ "TrainerControl": {
924
+ "args": {
925
+ "should_epoch_stop": false,
926
+ "should_evaluate": false,
927
+ "should_log": false,
928
+ "should_save": false,
929
+ "should_training_stop": false
930
+ },
931
+ "attributes": {}
932
+ }
933
+ },
934
+ "total_flos": 5.363892569191219e+17,
935
+ "train_batch_size": 96,
936
+ "trial_name": null,
937
+ "trial_params": null
938
+ }