aixk commited on
Commit
7f07402
·
1 Parent(s): f301426

FastPlus40m Backup checkpoint-1400 at global step 1400

Browse files
checkpoint-1400/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "FastPlus40mForCausalLM"
4
+ ],
5
+ "dtype": "float32",
6
+ "hidden_size": 512,
7
+ "initializer_range": 0.02,
8
+ "intermediate_size": 1536,
9
+ "kd_alpha": 0.4,
10
+ "kd_temperature": 2.5,
11
+ "max_position_embeddings": 512,
12
+ "model_type": "fastplus_40m",
13
+ "num_attention_heads": 8,
14
+ "num_hidden_layers": 11,
15
+ "tie_word_embeddings": true,
16
+ "transformers_version": "5.12.0",
17
+ "use_cache": false,
18
+ "vocab_size": 8160
19
+ }
checkpoint-1400/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "output_attentions": false,
4
+ "output_hidden_states": false,
5
+ "transformers_version": "5.12.0"
6
+ }
checkpoint-1400/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:30f2b68e060229613abb17e76031a49288a789a40c3dd3ea564b82edff87539c
3
+ size 159781520
checkpoint-1400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06c4c33a555bcd9faf6727a6e555271ab24584d58b845162ab27887349df6db2
3
+ size 46167627
checkpoint-1400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae07606ffe65644186d42f98b4116d1b49f8d756b484836f6bcfea7948f64fa9
3
+ size 5625
checkpoint-1400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1400/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "extra_special_tokens": [
6
+ "<|user|>",
7
+ "<|assistant|>"
8
+ ],
9
+ "model_max_length": 1000000000000000019884624838656,
10
+ "pad_token": "<pad>",
11
+ "tokenizer_class": "TokenizersBackend",
12
+ "unk_token": "<unk>"
13
+ }
checkpoint-1400/trainer_state.json ADDED
@@ -0,0 +1,1154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.726023401289988,
6
+ "eval_steps": 500,
7
+ "global_step": 1400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.005185881437785629,
14
+ "grad_norm": 258.9592590332031,
15
+ "learning_rate": 2.5499999999999996e-05,
16
+ "loss": 144.2029296875,
17
+ "original_ce_loss": 9.012,
18
+ "step": 10
19
+ },
20
+ {
21
+ "epoch": 0.010371762875571258,
22
+ "grad_norm": 257.445556640625,
23
+ "learning_rate": 6.8e-05,
24
+ "loss": 143.48330078125,
25
+ "original_ce_loss": 8.967,
26
+ "step": 20
27
+ },
28
+ {
29
+ "epoch": 0.015557644313356886,
30
+ "grad_norm": 433.229248046875,
31
+ "learning_rate": 0.00010625,
32
+ "loss": 140.6442138671875,
33
+ "original_ce_loss": 8.7896,
34
+ "step": 30
35
+ },
36
+ {
37
+ "epoch": 0.020743525751142515,
38
+ "grad_norm": 340.6360168457031,
39
+ "learning_rate": 0.00014450000000000002,
40
+ "loss": 128.65208740234374,
41
+ "original_ce_loss": 8.0401,
42
+ "step": 40
43
+ },
44
+ {
45
+ "epoch": 0.025929407188928142,
46
+ "grad_norm": 44.42266082763672,
47
+ "learning_rate": 0.000187,
48
+ "loss": 116.50721435546875,
49
+ "original_ce_loss": 7.281,
50
+ "step": 50
51
+ },
52
+ {
53
+ "epoch": 0.031115288626713773,
54
+ "grad_norm": 29.20110511779785,
55
+ "learning_rate": 0.0002295,
56
+ "loss": 110.56180419921876,
57
+ "original_ce_loss": 6.9094,
58
+ "step": 60
59
+ },
60
+ {
61
+ "epoch": 0.0363011700644994,
62
+ "grad_norm": 15.184632301330566,
63
+ "learning_rate": 0.000272,
64
+ "loss": 103.650634765625,
65
+ "original_ce_loss": 6.4775,
66
+ "step": 70
67
+ },
68
+ {
69
+ "epoch": 0.04148705150228503,
70
+ "grad_norm": 39.94831085205078,
71
+ "learning_rate": 0.0003145,
72
+ "loss": 100.78997192382812,
73
+ "original_ce_loss": 6.2987,
74
+ "step": 80
75
+ },
76
+ {
77
+ "epoch": 0.04667293294007066,
78
+ "grad_norm": NaN,
79
+ "learning_rate": 0.00035699999999999995,
80
+ "loss": 100.00459594726563,
81
+ "original_ce_loss": 6.2496,
82
+ "step": 90
83
+ },
84
+ {
85
+ "epoch": 0.051858814377856284,
86
+ "grad_norm": 38.79261779785156,
87
+ "learning_rate": 0.00039525,
88
+ "loss": 99.06669921875,
89
+ "original_ce_loss": 6.191,
90
+ "step": 100
91
+ },
92
+ {
93
+ "epoch": 0.057044695815641915,
94
+ "grad_norm": 37.9491081237793,
95
+ "learning_rate": 0.0004249971787418828,
96
+ "loss": 98.19793090820312,
97
+ "original_ce_loss": 6.1367,
98
+ "step": 110
99
+ },
100
+ {
101
+ "epoch": 0.062230577253427545,
102
+ "grad_norm": 29.972000122070312,
103
+ "learning_rate": 0.00042494702512599277,
104
+ "loss": 98.03455810546875,
105
+ "original_ce_loss": 6.1265,
106
+ "step": 120
107
+ },
108
+ {
109
+ "epoch": 0.06741645869121317,
110
+ "grad_norm": 327.20526123046875,
111
+ "learning_rate": 0.00042483419391697364,
112
+ "loss": 97.42244873046874,
113
+ "original_ce_loss": 6.0883,
114
+ "step": 130
115
+ },
116
+ {
117
+ "epoch": 0.0726023401289988,
118
+ "grad_norm": 26.064453125,
119
+ "learning_rate": 0.00042465871840307033,
120
+ "loss": 97.09953002929687,
121
+ "original_ce_loss": 6.0681,
122
+ "step": 140
123
+ },
124
+ {
125
+ "epoch": 0.07778822156678443,
126
+ "grad_norm": 12.600736618041992,
127
+ "learning_rate": 0.00042442065035428465,
128
+ "loss": 95.89002685546875,
129
+ "original_ce_loss": 5.9925,
130
+ "step": 150
131
+ },
132
+ {
133
+ "epoch": 0.08297410300457006,
134
+ "grad_norm": 16.495969772338867,
135
+ "learning_rate": 0.00042415293016226217,
136
+ "loss": 95.16254272460938,
137
+ "original_ce_loss": 5.947,
138
+ "step": 160
139
+ },
140
+ {
141
+ "epoch": 0.08815998444235569,
142
+ "grad_norm": 11.548162460327148,
143
+ "learning_rate": 0.0004237961450446541,
144
+ "loss": 94.430859375,
145
+ "original_ce_loss": 5.9013,
146
+ "step": 170
147
+ },
148
+ {
149
+ "epoch": 0.09334586588014132,
150
+ "grad_norm": 9.678289413452148,
151
+ "learning_rate": 0.0004233770218745316,
152
+ "loss": 93.07973022460938,
153
+ "original_ce_loss": 5.8169,
154
+ "step": 180
155
+ },
156
+ {
157
+ "epoch": 0.09853174731792694,
158
+ "grad_norm": 18.5349178314209,
159
+ "learning_rate": 0.0004230466099925828,
160
+ "loss": 92.07313232421875,
161
+ "original_ce_loss": 5.754,
162
+ "step": 190
163
+ },
164
+ {
165
+ "epoch": 0.10371762875571257,
166
+ "grad_norm": 14.110535621643066,
167
+ "learning_rate": 0.00042252180574663893,
168
+ "loss": 91.09517822265624,
169
+ "original_ce_loss": 5.6929,
170
+ "step": 200
171
+ },
172
+ {
173
+ "epoch": 0.1089035101934982,
174
+ "grad_norm": 636.2926025390625,
175
+ "learning_rate": 0.0004219350394125885,
176
+ "loss": 138.36116943359374,
177
+ "original_ce_loss": 8.6469,
178
+ "step": 210
179
+ },
180
+ {
181
+ "epoch": 0.11408939163128383,
182
+ "grad_norm": 123.25706481933594,
183
+ "learning_rate": 0.0004212864841023004,
184
+ "loss": 119.11766357421875,
185
+ "original_ce_loss": 7.4442,
186
+ "step": 220
187
+ },
188
+ {
189
+ "epoch": 0.11927527306906946,
190
+ "grad_norm": 43.103580474853516,
191
+ "learning_rate": 0.00042057633115705526,
192
+ "loss": 112.13214111328125,
193
+ "original_ce_loss": 7.0076,
194
+ "step": 230
195
+ },
196
+ {
197
+ "epoch": 0.12446115450685509,
198
+ "grad_norm": 26.395227432250977,
199
+ "learning_rate": 0.00041980479009109473,
200
+ "loss": 103.69727783203125,
201
+ "original_ce_loss": 6.4804,
202
+ "step": 240
203
+ },
204
+ {
205
+ "epoch": 0.12964703594464072,
206
+ "grad_norm": 11.802638053894043,
207
+ "learning_rate": 0.0004189720885298092,
208
+ "loss": 101.15250854492187,
209
+ "original_ce_loss": 6.3214,
210
+ "step": 250
211
+ },
212
+ {
213
+ "epoch": 0.13483291738242634,
214
+ "grad_norm": 88.59038543701172,
215
+ "learning_rate": 0.0004180784721425819,
216
+ "loss": 99.70953979492188,
217
+ "original_ce_loss": 6.2312,
218
+ "step": 260
219
+ },
220
+ {
221
+ "epoch": 0.14001879882021198,
222
+ "grad_norm": 32.838016510009766,
223
+ "learning_rate": 0.0004171242045703103,
224
+ "loss": 98.80262451171875,
225
+ "original_ce_loss": 6.1745,
226
+ "step": 270
227
+ },
228
+ {
229
+ "epoch": 0.1452046802579976,
230
+ "grad_norm": 20.591064453125,
231
+ "learning_rate": 0.0004161095673476243,
232
+ "loss": 97.41932373046875,
233
+ "original_ce_loss": 6.0881,
234
+ "step": 280
235
+ },
236
+ {
237
+ "epoch": 0.15039056169578324,
238
+ "grad_norm": 17.195009231567383,
239
+ "learning_rate": 0.0004150348598198265,
240
+ "loss": 95.6924560546875,
241
+ "original_ce_loss": 5.9802,
242
+ "step": 290
243
+ },
244
+ {
245
+ "epoch": 0.15557644313356886,
246
+ "grad_norm": 14.183113098144531,
247
+ "learning_rate": 0.0004139003990545772,
248
+ "loss": 94.54354858398438,
249
+ "original_ce_loss": 5.9084,
250
+ "step": 300
251
+ },
252
+ {
253
+ "epoch": 0.16076232457135448,
254
+ "grad_norm": 14.340221405029297,
255
+ "learning_rate": 0.0004127065197483504,
256
+ "loss": 93.042724609375,
257
+ "original_ce_loss": 5.8146,
258
+ "step": 310
259
+ },
260
+ {
261
+ "epoch": 0.16594820600914012,
262
+ "grad_norm": 13.095779418945312,
263
+ "learning_rate": 0.00041145357412768993,
264
+ "loss": 92.05337524414062,
265
+ "original_ce_loss": 5.7527,
266
+ "step": 320
267
+ },
268
+ {
269
+ "epoch": 0.17113408744692574,
270
+ "grad_norm": 29.691593170166016,
271
+ "learning_rate": 0.0004101419318452931,
272
+ "loss": 90.80604248046875,
273
+ "original_ce_loss": 5.6748,
274
+ "step": 330
275
+ },
276
+ {
277
+ "epoch": 0.17631996888471138,
278
+ "grad_norm": 25.52593231201172,
279
+ "learning_rate": 0.0004089115876198953,
280
+ "loss": 89.53719482421874,
281
+ "original_ce_loss": 5.5955,
282
+ "step": 340
283
+ },
284
+ {
285
+ "epoch": 0.181505850322497,
286
+ "grad_norm": 16.21922492980957,
287
+ "learning_rate": 0.00040763430634235056,
288
+ "loss": 91.01376342773438,
289
+ "original_ce_loss": 5.6878,
290
+ "step": 350
291
+ },
292
+ {
293
+ "epoch": 0.18669173176028264,
294
+ "grad_norm": 21.000469207763672,
295
+ "learning_rate": 0.0004061604269221445,
296
+ "loss": 92.56256103515625,
297
+ "original_ce_loss": 5.7846,
298
+ "step": 360
299
+ },
300
+ {
301
+ "epoch": 0.19187761319806826,
302
+ "grad_norm": NaN,
303
+ "learning_rate": 0.00040462941246116,
304
+ "loss": 92.55601806640625,
305
+ "original_ce_loss": 5.7842,
306
+ "step": 370
307
+ },
308
+ {
309
+ "epoch": 0.19706349463585388,
310
+ "grad_norm": 24.319599151611328,
311
+ "learning_rate": 0.0004032030219624389,
312
+ "loss": 93.43309326171875,
313
+ "original_ce_loss": 5.839,
314
+ "step": 380
315
+ },
316
+ {
317
+ "epoch": 0.20224937607363952,
318
+ "grad_norm": 32.82670974731445,
319
+ "learning_rate": 0.0004015647091557703,
320
+ "loss": 94.34906005859375,
321
+ "original_ce_loss": 5.8963,
322
+ "step": 390
323
+ },
324
+ {
325
+ "epoch": 0.20743525751142514,
326
+ "grad_norm": 19.567174911499023,
327
+ "learning_rate": 0.0003998706171688167,
328
+ "loss": 98.32274169921875,
329
+ "original_ce_loss": 6.1446,
330
+ "step": 400
331
+ },
332
+ {
333
+ "epoch": 0.21262113894921078,
334
+ "grad_norm": 21.221345901489258,
335
+ "learning_rate": 0.0003984755178877064,
336
+ "loss": 99.81248779296875,
337
+ "original_ce_loss": 6.2377,
338
+ "step": 410
339
+ },
340
+ {
341
+ "epoch": 0.2178070203869964,
342
+ "grad_norm": 66.5442123413086,
343
+ "learning_rate": 0.0003974061340973119,
344
+ "loss": 100.27135009765625,
345
+ "original_ce_loss": 6.2664,
346
+ "step": 420
347
+ },
348
+ {
349
+ "epoch": 0.22299290182478204,
350
+ "grad_norm": 12.657095909118652,
351
+ "learning_rate": 0.00039576527110212465,
352
+ "loss": 99.00567626953125,
353
+ "original_ce_loss": 6.1873,
354
+ "step": 430
355
+ },
356
+ {
357
+ "epoch": 0.22817878326256766,
358
+ "grad_norm": 16.740495681762695,
359
+ "learning_rate": 0.00039389074236337047,
360
+ "loss": 93.76760864257812,
361
+ "original_ce_loss": 5.8599,
362
+ "step": 440
363
+ },
364
+ {
365
+ "epoch": 0.23336466470035327,
366
+ "grad_norm": 11.1776704788208,
367
+ "learning_rate": 0.00039215789495377897,
368
+ "loss": 91.48955078125,
369
+ "original_ce_loss": 5.7176,
370
+ "step": 450
371
+ },
372
+ {
373
+ "epoch": 0.23855054613813892,
374
+ "grad_norm": 8.501482009887695,
375
+ "learning_rate": 0.0003901821733222283,
376
+ "loss": 89.11729736328125,
377
+ "original_ce_loss": 5.5693,
378
+ "step": 460
379
+ },
380
+ {
381
+ "epoch": 0.24373642757592454,
382
+ "grad_norm": 7.447785377502441,
383
+ "learning_rate": 0.0003881540306649693,
384
+ "loss": 89.28330688476562,
385
+ "original_ce_loss": 5.5797,
386
+ "step": 470
387
+ },
388
+ {
389
+ "epoch": 0.24892230901371018,
390
+ "grad_norm": 9.327811241149902,
391
+ "learning_rate": 0.0003860740653386992,
392
+ "loss": 89.13573608398437,
393
+ "original_ce_loss": 5.5705,
394
+ "step": 480
395
+ },
396
+ {
397
+ "epoch": 0.2541081904514958,
398
+ "grad_norm": 15.746177673339844,
399
+ "learning_rate": 0.0003839428909891979,
400
+ "loss": 89.25917358398438,
401
+ "original_ce_loss": 5.5782,
402
+ "step": 490
403
+ },
404
+ {
405
+ "epoch": 0.25929407188928144,
406
+ "grad_norm": 23.92910385131836,
407
+ "learning_rate": 0.00038176113637028637,
408
+ "loss": 89.31339721679687,
409
+ "original_ce_loss": 5.5816,
410
+ "step": 500
411
+ },
412
+ {
413
+ "epoch": 0.2644799533270671,
414
+ "grad_norm": 17.797733306884766,
415
+ "learning_rate": 0.0003795294451583268,
416
+ "loss": 89.24127197265625,
417
+ "original_ce_loss": 5.5771,
418
+ "step": 510
419
+ },
420
+ {
421
+ "epoch": 0.2696658347648527,
422
+ "grad_norm": 28.356403350830078,
423
+ "learning_rate": 0.0003772484757623211,
424
+ "loss": 90.0321044921875,
425
+ "original_ce_loss": 5.6265,
426
+ "step": 520
427
+ },
428
+ {
429
+ "epoch": 0.2748517162026383,
430
+ "grad_norm": 35.42921829223633,
431
+ "learning_rate": 0.0003749189011296624,
432
+ "loss": 90.63424072265624,
433
+ "original_ce_loss": 5.6641,
434
+ "step": 530
435
+ },
436
+ {
437
+ "epoch": 0.28003759764042396,
438
+ "grad_norm": 59.50849151611328,
439
+ "learning_rate": 0.00037254140854759755,
440
+ "loss": 92.27958984375,
441
+ "original_ce_loss": 5.767,
442
+ "step": 540
443
+ },
444
+ {
445
+ "epoch": 0.28522347907820955,
446
+ "grad_norm": 57.38438034057617,
447
+ "learning_rate": 0.0003701166994404587,
448
+ "loss": 95.16227416992187,
449
+ "original_ce_loss": 5.9472,
450
+ "step": 550
451
+ },
452
+ {
453
+ "epoch": 0.2904093605159952,
454
+ "grad_norm": 79.11128997802734,
455
+ "learning_rate": 0.0003676454891627244,
456
+ "loss": 100.294775390625,
457
+ "original_ce_loss": 6.2679,
458
+ "step": 560
459
+ },
460
+ {
461
+ "epoch": 0.29559524195378084,
462
+ "grad_norm": 90.20303344726562,
463
+ "learning_rate": 0.0003651285067879708,
464
+ "loss": 104.52440185546875,
465
+ "original_ce_loss": 6.5323,
466
+ "step": 570
467
+ },
468
+ {
469
+ "epoch": 0.3007811233915665,
470
+ "grad_norm": 205.7957000732422,
471
+ "learning_rate": 0.0003625664948937748,
472
+ "loss": 103.786083984375,
473
+ "original_ce_loss": 6.4861,
474
+ "step": 580
475
+ },
476
+ {
477
+ "epoch": 0.3059670048293521,
478
+ "grad_norm": 204.43067932128906,
479
+ "learning_rate": 0.00035996020934263463,
480
+ "loss": 103.99923095703124,
481
+ "original_ce_loss": 6.4995,
482
+ "step": 590
483
+ },
484
+ {
485
+ "epoch": 0.3111528862671377,
486
+ "grad_norm": 162.4797821044922,
487
+ "learning_rate": 0.00035731041905896917,
488
+ "loss": 103.354345703125,
489
+ "original_ce_loss": 6.4592,
490
+ "step": 600
491
+ },
492
+ {
493
+ "epoch": 0.31633876770492336,
494
+ "grad_norm": 137.0195770263672,
495
+ "learning_rate": 0.00035461790580226484,
496
+ "loss": 104.15499267578124,
497
+ "original_ce_loss": 6.5092,
498
+ "step": 610
499
+ },
500
+ {
501
+ "epoch": 0.32152464914270895,
502
+ "grad_norm": 255.99755859375,
503
+ "learning_rate": 0.00035188346393643576,
504
+ "loss": 103.5878173828125,
505
+ "original_ce_loss": 6.4738,
506
+ "step": 620
507
+ },
508
+ {
509
+ "epoch": 0.3267105305804946,
510
+ "grad_norm": 132.10855102539062,
511
+ "learning_rate": 0.0003491079001954646,
512
+ "loss": 103.6423095703125,
513
+ "original_ce_loss": 6.4772,
514
+ "step": 630
515
+ },
516
+ {
517
+ "epoch": 0.33189641201828024,
518
+ "grad_norm": 15410.90234375,
519
+ "learning_rate": 0.0003462920334453946,
520
+ "loss": 103.73336181640624,
521
+ "original_ce_loss": 6.4829,
522
+ "step": 640
523
+ },
524
+ {
525
+ "epoch": 0.3370822934560659,
526
+ "grad_norm": 518.74267578125,
527
+ "learning_rate": 0.00034343669444274194,
528
+ "loss": 106.26522216796874,
529
+ "original_ce_loss": 6.6411,
530
+ "step": 650
531
+ },
532
+ {
533
+ "epoch": 0.3422681748938515,
534
+ "grad_norm": 1601.3238525390625,
535
+ "learning_rate": 0.0003405427255894003,
536
+ "loss": 105.530419921875,
537
+ "original_ce_loss": 6.5952,
538
+ "step": 660
539
+ },
540
+ {
541
+ "epoch": 0.3474540563316371,
542
+ "grad_norm": 377.1868591308594,
543
+ "learning_rate": 0.00033761098068410935,
544
+ "loss": 106.2079345703125,
545
+ "original_ce_loss": 6.6375,
546
+ "step": 670
547
+ },
548
+ {
549
+ "epoch": 0.35263993776942276,
550
+ "grad_norm": NaN,
551
+ "learning_rate": 0.00033613121207849256,
552
+ "loss": 164.1599853515625,
553
+ "original_ce_loss": NaN,
554
+ "step": 680
555
+ },
556
+ {
557
+ "epoch": 0.35782581920720835,
558
+ "grad_norm": NaN,
559
+ "learning_rate": 0.00033613121207849256,
560
+ "loss": 304.5229248046875,
561
+ "original_ce_loss": NaN,
562
+ "step": 690
563
+ },
564
+ {
565
+ "epoch": 0.363011700644994,
566
+ "grad_norm": NaN,
567
+ "learning_rate": 0.00033613121207849256,
568
+ "loss": 642.434521484375,
569
+ "original_ce_loss": NaN,
570
+ "step": 700
571
+ },
572
+ {
573
+ "epoch": 0.36819758208277964,
574
+ "grad_norm": NaN,
575
+ "learning_rate": 0.00033613121207849256,
576
+ "loss": 303.0937744140625,
577
+ "original_ce_loss": NaN,
578
+ "step": 710
579
+ },
580
+ {
581
+ "epoch": 0.3733834635205653,
582
+ "grad_norm": NaN,
583
+ "learning_rate": 0.00033613121207849256,
584
+ "loss": 937.6494140625,
585
+ "original_ce_loss": NaN,
586
+ "step": 720
587
+ },
588
+ {
589
+ "epoch": 0.3785693449583509,
590
+ "grad_norm": NaN,
591
+ "learning_rate": 0.00033613121207849256,
592
+ "loss": 376.8843505859375,
593
+ "original_ce_loss": NaN,
594
+ "step": 730
595
+ },
596
+ {
597
+ "epoch": 0.3837552263961365,
598
+ "grad_norm": NaN,
599
+ "learning_rate": 0.00033613121207849256,
600
+ "loss": 326.5380126953125,
601
+ "original_ce_loss": NaN,
602
+ "step": 740
603
+ },
604
+ {
605
+ "epoch": 0.38894110783392216,
606
+ "grad_norm": NaN,
607
+ "learning_rate": 0.00033613121207849256,
608
+ "loss": 538.42578125,
609
+ "original_ce_loss": NaN,
610
+ "step": 750
611
+ },
612
+ {
613
+ "epoch": 0.39412698927170775,
614
+ "grad_norm": NaN,
615
+ "learning_rate": 0.00033613121207849256,
616
+ "loss": 371.935546875,
617
+ "original_ce_loss": NaN,
618
+ "step": 760
619
+ },
620
+ {
621
+ "epoch": 0.3993128707094934,
622
+ "grad_norm": NaN,
623
+ "learning_rate": 0.00033613121207849256,
624
+ "loss": 192.04744873046874,
625
+ "original_ce_loss": NaN,
626
+ "step": 770
627
+ },
628
+ {
629
+ "epoch": 0.40449875214727904,
630
+ "grad_norm": NaN,
631
+ "learning_rate": 0.00033613121207849256,
632
+ "loss": 270.1376708984375,
633
+ "original_ce_loss": NaN,
634
+ "step": 780
635
+ },
636
+ {
637
+ "epoch": 0.4096846335850647,
638
+ "grad_norm": NaN,
639
+ "learning_rate": 0.00033613121207849256,
640
+ "loss": 303.8236328125,
641
+ "original_ce_loss": NaN,
642
+ "step": 790
643
+ },
644
+ {
645
+ "epoch": 0.4148705150228503,
646
+ "grad_norm": NaN,
647
+ "learning_rate": 0.00033613121207849256,
648
+ "loss": 316.044873046875,
649
+ "original_ce_loss": NaN,
650
+ "step": 800
651
+ },
652
+ {
653
+ "epoch": 0.4200563964606359,
654
+ "grad_norm": NaN,
655
+ "learning_rate": 0.00033613121207849256,
656
+ "loss": 217.78037109375,
657
+ "original_ce_loss": NaN,
658
+ "step": 810
659
+ },
660
+ {
661
+ "epoch": 0.42524227789842156,
662
+ "grad_norm": NaN,
663
+ "learning_rate": 0.00033613121207849256,
664
+ "loss": 270.949853515625,
665
+ "original_ce_loss": NaN,
666
+ "step": 820
667
+ },
668
+ {
669
+ "epoch": 0.43042815933620715,
670
+ "grad_norm": NaN,
671
+ "learning_rate": 0.00033613121207849256,
672
+ "loss": 1181.424609375,
673
+ "original_ce_loss": NaN,
674
+ "step": 830
675
+ },
676
+ {
677
+ "epoch": 0.4356140407739928,
678
+ "grad_norm": NaN,
679
+ "learning_rate": 0.00033613121207849256,
680
+ "loss": 303.05693359375,
681
+ "original_ce_loss": NaN,
682
+ "step": 840
683
+ },
684
+ {
685
+ "epoch": 0.44079992221177844,
686
+ "grad_norm": NaN,
687
+ "learning_rate": 0.00033613121207849256,
688
+ "loss": 324.5959716796875,
689
+ "original_ce_loss": NaN,
690
+ "step": 850
691
+ },
692
+ {
693
+ "epoch": 0.4459858036495641,
694
+ "grad_norm": NaN,
695
+ "learning_rate": 0.00033613121207849256,
696
+ "loss": 764.512548828125,
697
+ "original_ce_loss": NaN,
698
+ "step": 860
699
+ },
700
+ {
701
+ "epoch": 0.45117168508734967,
702
+ "grad_norm": NaN,
703
+ "learning_rate": 0.00033613121207849256,
704
+ "loss": 257.5296142578125,
705
+ "original_ce_loss": NaN,
706
+ "step": 870
707
+ },
708
+ {
709
+ "epoch": 0.4563575665251353,
710
+ "grad_norm": NaN,
711
+ "learning_rate": 0.00033613121207849256,
712
+ "loss": 933.280078125,
713
+ "original_ce_loss": NaN,
714
+ "step": 880
715
+ },
716
+ {
717
+ "epoch": 0.46154344796292096,
718
+ "grad_norm": NaN,
719
+ "learning_rate": 0.00033613121207849256,
720
+ "loss": 206.591650390625,
721
+ "original_ce_loss": NaN,
722
+ "step": 890
723
+ },
724
+ {
725
+ "epoch": 0.46672932940070655,
726
+ "grad_norm": NaN,
727
+ "learning_rate": 0.00033613121207849256,
728
+ "loss": 342.59130859375,
729
+ "original_ce_loss": NaN,
730
+ "step": 900
731
+ },
732
+ {
733
+ "epoch": 0.4719152108384922,
734
+ "grad_norm": NaN,
735
+ "learning_rate": 0.00033613121207849256,
736
+ "loss": 449.882177734375,
737
+ "original_ce_loss": NaN,
738
+ "step": 910
739
+ },
740
+ {
741
+ "epoch": 0.47710109227627784,
742
+ "grad_norm": NaN,
743
+ "learning_rate": 0.00033613121207849256,
744
+ "loss": 162.45423583984376,
745
+ "original_ce_loss": NaN,
746
+ "step": 920
747
+ },
748
+ {
749
+ "epoch": 0.4822869737140635,
750
+ "grad_norm": NaN,
751
+ "learning_rate": 0.00033613121207849256,
752
+ "loss": 230.4999755859375,
753
+ "original_ce_loss": NaN,
754
+ "step": 930
755
+ },
756
+ {
757
+ "epoch": 0.48747285515184907,
758
+ "grad_norm": NaN,
759
+ "learning_rate": 0.00033613121207849256,
760
+ "loss": 516.80693359375,
761
+ "original_ce_loss": NaN,
762
+ "step": 940
763
+ },
764
+ {
765
+ "epoch": 0.4926587365896347,
766
+ "grad_norm": NaN,
767
+ "learning_rate": 0.00033613121207849256,
768
+ "loss": 365.7063720703125,
769
+ "original_ce_loss": NaN,
770
+ "step": 950
771
+ },
772
+ {
773
+ "epoch": 0.49784461802742036,
774
+ "grad_norm": NaN,
775
+ "learning_rate": 0.00033613121207849256,
776
+ "loss": 281.401904296875,
777
+ "original_ce_loss": NaN,
778
+ "step": 960
779
+ },
780
+ {
781
+ "epoch": 0.503030499465206,
782
+ "grad_norm": NaN,
783
+ "learning_rate": 0.00033613121207849256,
784
+ "loss": 399.17626953125,
785
+ "original_ce_loss": NaN,
786
+ "step": 970
787
+ },
788
+ {
789
+ "epoch": 0.5082163809029916,
790
+ "grad_norm": NaN,
791
+ "learning_rate": 0.00033613121207849256,
792
+ "loss": 277.8823486328125,
793
+ "original_ce_loss": NaN,
794
+ "step": 980
795
+ },
796
+ {
797
+ "epoch": 0.5134022623407772,
798
+ "grad_norm": NaN,
799
+ "learning_rate": 0.00033613121207849256,
800
+ "loss": 210.3106689453125,
801
+ "original_ce_loss": NaN,
802
+ "step": 990
803
+ },
804
+ {
805
+ "epoch": 0.5185881437785629,
806
+ "grad_norm": NaN,
807
+ "learning_rate": 0.00033613121207849256,
808
+ "loss": 233.843701171875,
809
+ "original_ce_loss": NaN,
810
+ "step": 1000
811
+ },
812
+ {
813
+ "epoch": 0.5237740252163485,
814
+ "grad_norm": NaN,
815
+ "learning_rate": 0.00033613121207849256,
816
+ "loss": 379.18935546875,
817
+ "original_ce_loss": NaN,
818
+ "step": 1010
819
+ },
820
+ {
821
+ "epoch": 0.5289599066541342,
822
+ "grad_norm": NaN,
823
+ "learning_rate": 0.00033613121207849256,
824
+ "loss": 375.970751953125,
825
+ "original_ce_loss": NaN,
826
+ "step": 1020
827
+ },
828
+ {
829
+ "epoch": 0.5341457880919197,
830
+ "grad_norm": NaN,
831
+ "learning_rate": 0.00033613121207849256,
832
+ "loss": 295.0241455078125,
833
+ "original_ce_loss": NaN,
834
+ "step": 1030
835
+ },
836
+ {
837
+ "epoch": 0.5393316695297053,
838
+ "grad_norm": NaN,
839
+ "learning_rate": 0.00033613121207849256,
840
+ "loss": 931.68974609375,
841
+ "original_ce_loss": NaN,
842
+ "step": 1040
843
+ },
844
+ {
845
+ "epoch": 0.544517550967491,
846
+ "grad_norm": NaN,
847
+ "learning_rate": 0.00033613121207849256,
848
+ "loss": 245.52197265625,
849
+ "original_ce_loss": NaN,
850
+ "step": 1050
851
+ },
852
+ {
853
+ "epoch": 0.5497034324052766,
854
+ "grad_norm": NaN,
855
+ "learning_rate": 0.00033613121207849256,
856
+ "loss": 308.66875,
857
+ "original_ce_loss": NaN,
858
+ "step": 1060
859
+ },
860
+ {
861
+ "epoch": 0.5548893138430623,
862
+ "grad_norm": NaN,
863
+ "learning_rate": 0.00033613121207849256,
864
+ "loss": 485.547607421875,
865
+ "original_ce_loss": NaN,
866
+ "step": 1070
867
+ },
868
+ {
869
+ "epoch": 0.5600751952808479,
870
+ "grad_norm": NaN,
871
+ "learning_rate": 0.00033613121207849256,
872
+ "loss": 296.8137939453125,
873
+ "original_ce_loss": NaN,
874
+ "step": 1080
875
+ },
876
+ {
877
+ "epoch": 0.5652610767186336,
878
+ "grad_norm": NaN,
879
+ "learning_rate": 0.00033613121207849256,
880
+ "loss": 410.13212890625,
881
+ "original_ce_loss": NaN,
882
+ "step": 1090
883
+ },
884
+ {
885
+ "epoch": 0.5704469581564191,
886
+ "grad_norm": NaN,
887
+ "learning_rate": 0.00033613121207849256,
888
+ "loss": 382.29248046875,
889
+ "original_ce_loss": NaN,
890
+ "step": 1100
891
+ },
892
+ {
893
+ "epoch": 0.5756328395942047,
894
+ "grad_norm": NaN,
895
+ "learning_rate": 0.00033613121207849256,
896
+ "loss": 296.0078125,
897
+ "original_ce_loss": NaN,
898
+ "step": 1110
899
+ },
900
+ {
901
+ "epoch": 0.5808187210319904,
902
+ "grad_norm": NaN,
903
+ "learning_rate": 0.00033613121207849256,
904
+ "loss": 305.527001953125,
905
+ "original_ce_loss": NaN,
906
+ "step": 1120
907
+ },
908
+ {
909
+ "epoch": 0.586004602469776,
910
+ "grad_norm": NaN,
911
+ "learning_rate": 0.00033613121207849256,
912
+ "loss": 283.384033203125,
913
+ "original_ce_loss": NaN,
914
+ "step": 1130
915
+ },
916
+ {
917
+ "epoch": 0.5911904839075617,
918
+ "grad_norm": NaN,
919
+ "learning_rate": 0.00033613121207849256,
920
+ "loss": 339.2777099609375,
921
+ "original_ce_loss": NaN,
922
+ "step": 1140
923
+ },
924
+ {
925
+ "epoch": 0.5963763653453473,
926
+ "grad_norm": NaN,
927
+ "learning_rate": 0.00033613121207849256,
928
+ "loss": 577.176318359375,
929
+ "original_ce_loss": NaN,
930
+ "step": 1150
931
+ },
932
+ {
933
+ "epoch": 0.601562246783133,
934
+ "grad_norm": NaN,
935
+ "learning_rate": 0.00033613121207849256,
936
+ "loss": 369.7349609375,
937
+ "original_ce_loss": NaN,
938
+ "step": 1160
939
+ },
940
+ {
941
+ "epoch": 0.6067481282209185,
942
+ "grad_norm": NaN,
943
+ "learning_rate": 0.00033613121207849256,
944
+ "loss": 921.526953125,
945
+ "original_ce_loss": NaN,
946
+ "step": 1170
947
+ },
948
+ {
949
+ "epoch": 0.6119340096587041,
950
+ "grad_norm": NaN,
951
+ "learning_rate": 0.00033613121207849256,
952
+ "loss": 164.85584716796876,
953
+ "original_ce_loss": NaN,
954
+ "step": 1180
955
+ },
956
+ {
957
+ "epoch": 0.6171198910964898,
958
+ "grad_norm": NaN,
959
+ "learning_rate": 0.00033613121207849256,
960
+ "loss": 554.365576171875,
961
+ "original_ce_loss": NaN,
962
+ "step": 1190
963
+ },
964
+ {
965
+ "epoch": 0.6223057725342754,
966
+ "grad_norm": NaN,
967
+ "learning_rate": 0.00033613121207849256,
968
+ "loss": 426.81318359375,
969
+ "original_ce_loss": NaN,
970
+ "step": 1200
971
+ },
972
+ {
973
+ "epoch": 0.6274916539720611,
974
+ "grad_norm": NaN,
975
+ "learning_rate": 0.00033613121207849256,
976
+ "loss": 304.1646484375,
977
+ "original_ce_loss": NaN,
978
+ "step": 1210
979
+ },
980
+ {
981
+ "epoch": 0.6326775354098467,
982
+ "grad_norm": NaN,
983
+ "learning_rate": 0.00033613121207849256,
984
+ "loss": 595.540185546875,
985
+ "original_ce_loss": NaN,
986
+ "step": 1220
987
+ },
988
+ {
989
+ "epoch": 0.6378634168476324,
990
+ "grad_norm": NaN,
991
+ "learning_rate": 0.00033613121207849256,
992
+ "loss": 610.712060546875,
993
+ "original_ce_loss": NaN,
994
+ "step": 1230
995
+ },
996
+ {
997
+ "epoch": 0.6430492982854179,
998
+ "grad_norm": NaN,
999
+ "learning_rate": 0.00033613121207849256,
1000
+ "loss": 317.8940185546875,
1001
+ "original_ce_loss": NaN,
1002
+ "step": 1240
1003
+ },
1004
+ {
1005
+ "epoch": 0.6482351797232035,
1006
+ "grad_norm": NaN,
1007
+ "learning_rate": 0.00033613121207849256,
1008
+ "loss": 259.523388671875,
1009
+ "original_ce_loss": NaN,
1010
+ "step": 1250
1011
+ },
1012
+ {
1013
+ "epoch": 0.6534210611609892,
1014
+ "grad_norm": NaN,
1015
+ "learning_rate": 0.00033613121207849256,
1016
+ "loss": 260.00859375,
1017
+ "original_ce_loss": NaN,
1018
+ "step": 1260
1019
+ },
1020
+ {
1021
+ "epoch": 0.6586069425987748,
1022
+ "grad_norm": NaN,
1023
+ "learning_rate": 0.00033613121207849256,
1024
+ "loss": 182.1218994140625,
1025
+ "original_ce_loss": NaN,
1026
+ "step": 1270
1027
+ },
1028
+ {
1029
+ "epoch": 0.6637928240365605,
1030
+ "grad_norm": NaN,
1031
+ "learning_rate": 0.00033613121207849256,
1032
+ "loss": 273.693896484375,
1033
+ "original_ce_loss": NaN,
1034
+ "step": 1280
1035
+ },
1036
+ {
1037
+ "epoch": 0.6689787054743461,
1038
+ "grad_norm": NaN,
1039
+ "learning_rate": 0.00033613121207849256,
1040
+ "loss": 562.66689453125,
1041
+ "original_ce_loss": NaN,
1042
+ "step": 1290
1043
+ },
1044
+ {
1045
+ "epoch": 0.6741645869121318,
1046
+ "grad_norm": NaN,
1047
+ "learning_rate": 0.00033613121207849256,
1048
+ "loss": 1035.09306640625,
1049
+ "original_ce_loss": NaN,
1050
+ "step": 1300
1051
+ },
1052
+ {
1053
+ "epoch": 0.6793504683499173,
1054
+ "grad_norm": NaN,
1055
+ "learning_rate": 0.00033613121207849256,
1056
+ "loss": 1410.64736328125,
1057
+ "original_ce_loss": NaN,
1058
+ "step": 1310
1059
+ },
1060
+ {
1061
+ "epoch": 0.684536349787703,
1062
+ "grad_norm": NaN,
1063
+ "learning_rate": 0.00033613121207849256,
1064
+ "loss": 413.46591796875,
1065
+ "original_ce_loss": NaN,
1066
+ "step": 1320
1067
+ },
1068
+ {
1069
+ "epoch": 0.6897222312254886,
1070
+ "grad_norm": NaN,
1071
+ "learning_rate": 0.0003358341605965657,
1072
+ "loss": 286168.875,
1073
+ "original_ce_loss": NaN,
1074
+ "step": 1330
1075
+ },
1076
+ {
1077
+ "epoch": 0.6949081126632742,
1078
+ "grad_norm": NaN,
1079
+ "learning_rate": 0.0003358341605965657,
1080
+ "loss": 0.0,
1081
+ "original_ce_loss": NaN,
1082
+ "step": 1340
1083
+ },
1084
+ {
1085
+ "epoch": 0.7000939941010599,
1086
+ "grad_norm": NaN,
1087
+ "learning_rate": 0.0003358341605965657,
1088
+ "loss": 0.0,
1089
+ "original_ce_loss": NaN,
1090
+ "step": 1350
1091
+ },
1092
+ {
1093
+ "epoch": 0.7052798755388455,
1094
+ "grad_norm": NaN,
1095
+ "learning_rate": 0.0003358341605965657,
1096
+ "loss": 0.0,
1097
+ "original_ce_loss": NaN,
1098
+ "step": 1360
1099
+ },
1100
+ {
1101
+ "epoch": 0.7104657569766312,
1102
+ "grad_norm": NaN,
1103
+ "learning_rate": 0.0003358341605965657,
1104
+ "loss": 0.0,
1105
+ "original_ce_loss": NaN,
1106
+ "step": 1370
1107
+ },
1108
+ {
1109
+ "epoch": 0.7156516384144167,
1110
+ "grad_norm": NaN,
1111
+ "learning_rate": 0.0003358341605965657,
1112
+ "loss": 0.0,
1113
+ "original_ce_loss": NaN,
1114
+ "step": 1380
1115
+ },
1116
+ {
1117
+ "epoch": 0.7208375198522023,
1118
+ "grad_norm": NaN,
1119
+ "learning_rate": 0.0003358341605965657,
1120
+ "loss": 0.0,
1121
+ "original_ce_loss": NaN,
1122
+ "step": 1390
1123
+ },
1124
+ {
1125
+ "epoch": 0.726023401289988,
1126
+ "grad_norm": NaN,
1127
+ "learning_rate": 0.0003358341605965657,
1128
+ "loss": 0.0,
1129
+ "original_ce_loss": NaN,
1130
+ "step": 1400
1131
+ }
1132
+ ],
1133
+ "logging_steps": 10,
1134
+ "max_steps": 1929,
1135
+ "num_input_tokens_seen": 0,
1136
+ "num_train_epochs": 1,
1137
+ "save_steps": 200,
1138
+ "stateful_callbacks": {
1139
+ "TrainerControl": {
1140
+ "args": {
1141
+ "should_epoch_stop": false,
1142
+ "should_evaluate": false,
1143
+ "should_log": false,
1144
+ "should_save": true,
1145
+ "should_training_stop": false
1146
+ },
1147
+ "attributes": {}
1148
+ }
1149
+ },
1150
+ "total_flos": 1.4808501728600064e+16,
1151
+ "train_batch_size": 16,
1152
+ "trial_name": null,
1153
+ "trial_params": null
1154
+ }
checkpoint-1400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a5b94426045bba4934b9dcb4f7d04d1ceb183c5a625b4741fd6de5948d5dade
3
+ size 5201