Grogros commited on
Commit
2f57384
·
verified ·
1 Parent(s): 617223b

Training in progress, step 1500, checkpoint

Browse files
checkpoint-1500/config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 128000,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 128001,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2048,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 8192,
15
+ "max_position_embeddings": 131072,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 32,
19
+ "num_hidden_layers": 16,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": {
24
+ "factor": 32.0,
25
+ "high_freq_factor": 4.0,
26
+ "low_freq_factor": 1.0,
27
+ "original_max_position_embeddings": 8192,
28
+ "rope_type": "llama3"
29
+ },
30
+ "rope_theta": 500000.0,
31
+ "tie_word_embeddings": true,
32
+ "transformers_version": "4.57.1",
33
+ "use_cache": true,
34
+ "vocab_size": 128260
35
+ }
checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 128000,
4
+ "do_sample": true,
5
+ "eos_token_id": 128001,
6
+ "temperature": 0.6,
7
+ "top_p": 0.9,
8
+ "transformers_version": "4.57.1"
9
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e3b25bd359d451080b5023e006d6858ec1a4adab059eb90f34bea2d94241c12
3
+ size 2471661992
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb51eaa466e34d0e4e5ad446b11d3c4277349b318106a0283ef12e00a30ae929
3
+ size 3725003
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:076cae5a5e0780dc041430f690e0fce96ade47835dffe9dc8cdcd94071d183fd
3
+ size 14581
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f80cbe6e26c4a018c7d23cd0875893b9c33824b6db15b87e8678c1b3aef683be
3
+ size 1465
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,1084 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.6,
6
+ "eval_steps": 500,
7
+ "global_step": 1500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.004,
14
+ "grad_norm": 197.0,
15
+ "learning_rate": 7.2e-07,
16
+ "loss": 1.6492,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.008,
21
+ "grad_norm": 165.0,
22
+ "learning_rate": 1.52e-06,
23
+ "loss": 1.6789,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.012,
28
+ "grad_norm": 160.0,
29
+ "learning_rate": 2.3200000000000002e-06,
30
+ "loss": 1.5305,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.016,
35
+ "grad_norm": 205.0,
36
+ "learning_rate": 3.12e-06,
37
+ "loss": 1.6891,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.02,
42
+ "grad_norm": 106.5,
43
+ "learning_rate": 3.920000000000001e-06,
44
+ "loss": 1.4922,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.024,
49
+ "grad_norm": 29.625,
50
+ "learning_rate": 4.7200000000000005e-06,
51
+ "loss": 1.3617,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.028,
56
+ "grad_norm": 9.0625,
57
+ "learning_rate": 5.5200000000000005e-06,
58
+ "loss": 1.102,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.032,
63
+ "grad_norm": 21.25,
64
+ "learning_rate": 6.3200000000000005e-06,
65
+ "loss": 1.0555,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.036,
70
+ "grad_norm": 11.875,
71
+ "learning_rate": 7.1200000000000004e-06,
72
+ "loss": 0.9625,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.04,
77
+ "grad_norm": 205.0,
78
+ "learning_rate": 7.92e-06,
79
+ "loss": 0.9598,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.044,
84
+ "grad_norm": 4.40625,
85
+ "learning_rate": 8.720000000000001e-06,
86
+ "loss": 0.8262,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.048,
91
+ "grad_norm": 4.0625,
92
+ "learning_rate": 9.52e-06,
93
+ "loss": 0.8426,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.052,
98
+ "grad_norm": 140.0,
99
+ "learning_rate": 1.0320000000000001e-05,
100
+ "loss": 0.8352,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.056,
105
+ "grad_norm": 3.8125,
106
+ "learning_rate": 1.1120000000000002e-05,
107
+ "loss": 0.7488,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.06,
112
+ "grad_norm": 3.484375,
113
+ "learning_rate": 1.1920000000000001e-05,
114
+ "loss": 0.7242,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.064,
119
+ "grad_norm": 3.3125,
120
+ "learning_rate": 1.2720000000000002e-05,
121
+ "loss": 0.7527,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.068,
126
+ "grad_norm": 4.625,
127
+ "learning_rate": 1.3520000000000003e-05,
128
+ "loss": 0.6836,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.072,
133
+ "grad_norm": 4.125,
134
+ "learning_rate": 1.432e-05,
135
+ "loss": 0.7305,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.076,
140
+ "grad_norm": 3.671875,
141
+ "learning_rate": 1.5120000000000001e-05,
142
+ "loss": 0.6832,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.08,
147
+ "grad_norm": 5.21875,
148
+ "learning_rate": 1.5920000000000003e-05,
149
+ "loss": 0.6348,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.084,
154
+ "grad_norm": 3.171875,
155
+ "learning_rate": 1.672e-05,
156
+ "loss": 0.6645,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.088,
161
+ "grad_norm": 3.046875,
162
+ "learning_rate": 1.752e-05,
163
+ "loss": 0.6383,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.092,
168
+ "grad_norm": 3.125,
169
+ "learning_rate": 1.832e-05,
170
+ "loss": 0.6141,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.096,
175
+ "grad_norm": 2.453125,
176
+ "learning_rate": 1.912e-05,
177
+ "loss": 0.6262,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.1,
182
+ "grad_norm": 4.65625,
183
+ "learning_rate": 1.9920000000000002e-05,
184
+ "loss": 0.6109,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.104,
189
+ "grad_norm": 19.5,
190
+ "learning_rate": 1.9999210442038164e-05,
191
+ "loss": 0.6113,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.108,
196
+ "grad_norm": 2.546875,
197
+ "learning_rate": 1.9996481265944146e-05,
198
+ "loss": 0.5844,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.112,
203
+ "grad_norm": 2.5625,
204
+ "learning_rate": 1.9991803256020393e-05,
205
+ "loss": 0.6012,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.116,
210
+ "grad_norm": 3.625,
211
+ "learning_rate": 1.99851773242542e-05,
212
+ "loss": 0.6029,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.12,
217
+ "grad_norm": 2.203125,
218
+ "learning_rate": 1.99766047623841e-05,
219
+ "loss": 0.5723,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.124,
224
+ "grad_norm": 2.796875,
225
+ "learning_rate": 1.996608724164801e-05,
226
+ "loss": 0.5811,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.128,
231
+ "grad_norm": 2.125,
232
+ "learning_rate": 1.995362681245744e-05,
233
+ "loss": 0.5109,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.132,
238
+ "grad_norm": 89.0,
239
+ "learning_rate": 1.9939225903997748e-05,
240
+ "loss": 0.5428,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.136,
245
+ "grad_norm": 3.0625,
246
+ "learning_rate": 1.992288732375458e-05,
247
+ "loss": 0.498,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.14,
252
+ "grad_norm": 2.453125,
253
+ "learning_rate": 1.9904614256966514e-05,
254
+ "loss": 0.598,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.144,
259
+ "grad_norm": 1.9375,
260
+ "learning_rate": 1.9884410266004134e-05,
261
+ "loss": 0.5229,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.148,
266
+ "grad_norm": 2.3125,
267
+ "learning_rate": 1.986227928967551e-05,
268
+ "loss": 0.5152,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.152,
273
+ "grad_norm": 2.390625,
274
+ "learning_rate": 1.983822564245833e-05,
275
+ "loss": 0.543,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.156,
280
+ "grad_norm": 3.140625,
281
+ "learning_rate": 1.981225401365877e-05,
282
+ "loss": 0.5518,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.16,
287
+ "grad_norm": 3.109375,
288
+ "learning_rate": 1.9784369466497333e-05,
289
+ "loss": 0.5289,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.164,
294
+ "grad_norm": 2.34375,
295
+ "learning_rate": 1.9754577437121733e-05,
296
+ "loss": 0.5186,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.168,
301
+ "grad_norm": 2.296875,
302
+ "learning_rate": 1.9722883733547128e-05,
303
+ "loss": 0.5189,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.172,
308
+ "grad_norm": 1.9765625,
309
+ "learning_rate": 1.968929453452383e-05,
310
+ "loss": 0.5422,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.176,
315
+ "grad_norm": 2.296875,
316
+ "learning_rate": 1.965381638833274e-05,
317
+ "loss": 0.5281,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.18,
322
+ "grad_norm": 3.15625,
323
+ "learning_rate": 1.9616456211508756e-05,
324
+ "loss": 0.5316,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.184,
329
+ "grad_norm": 3.953125,
330
+ "learning_rate": 1.9577221287492368e-05,
331
+ "loss": 0.5521,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.188,
336
+ "grad_norm": 2.125,
337
+ "learning_rate": 1.9536119265209763e-05,
338
+ "loss": 0.5502,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.192,
343
+ "grad_norm": 2.28125,
344
+ "learning_rate": 1.9493158157581617e-05,
345
+ "loss": 0.5279,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.196,
350
+ "grad_norm": 2.265625,
351
+ "learning_rate": 1.9448346339960984e-05,
352
+ "loss": 0.5557,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.2,
357
+ "grad_norm": 2.296875,
358
+ "learning_rate": 1.9401692548500504e-05,
359
+ "loss": 0.4969,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.204,
364
+ "grad_norm": 2.0,
365
+ "learning_rate": 1.935320587844926e-05,
366
+ "loss": 0.5225,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.208,
371
+ "grad_norm": 2.25,
372
+ "learning_rate": 1.9302895782379648e-05,
373
+ "loss": 0.4932,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.212,
378
+ "grad_norm": 2.3125,
379
+ "learning_rate": 1.925077206834458e-05,
380
+ "loss": 0.4939,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.216,
385
+ "grad_norm": 3.09375,
386
+ "learning_rate": 1.9196844897965393e-05,
387
+ "loss": 0.4953,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.22,
392
+ "grad_norm": 2.078125,
393
+ "learning_rate": 1.914112478445079e-05,
394
+ "loss": 0.4768,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.224,
399
+ "grad_norm": 2.1875,
400
+ "learning_rate": 1.9083622590547313e-05,
401
+ "loss": 0.5273,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.228,
406
+ "grad_norm": 2.046875,
407
+ "learning_rate": 1.9024349526421596e-05,
408
+ "loss": 0.5279,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.232,
413
+ "grad_norm": 2.3125,
414
+ "learning_rate": 1.896331714747493e-05,
415
+ "loss": 0.543,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.236,
420
+ "grad_norm": 1.828125,
421
+ "learning_rate": 1.8900537352090523e-05,
422
+ "loss": 0.4842,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.24,
427
+ "grad_norm": 2.09375,
428
+ "learning_rate": 1.8836022379313884e-05,
429
+ "loss": 0.4787,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.244,
434
+ "grad_norm": 1.9609375,
435
+ "learning_rate": 1.8769784806466768e-05,
436
+ "loss": 0.5025,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.248,
441
+ "grad_norm": 1.84375,
442
+ "learning_rate": 1.870183754669526e-05,
443
+ "loss": 0.4691,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.252,
448
+ "grad_norm": 1.7421875,
449
+ "learning_rate": 1.863219384645227e-05,
450
+ "loss": 0.5076,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.256,
455
+ "grad_norm": 2.375,
456
+ "learning_rate": 1.8560867282915164e-05,
457
+ "loss": 0.5377,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.26,
462
+ "grad_norm": 1.9921875,
463
+ "learning_rate": 1.848787176133882e-05,
464
+ "loss": 0.5049,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.264,
469
+ "grad_norm": 1.875,
470
+ "learning_rate": 1.8413221512344805e-05,
471
+ "loss": 0.5322,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.268,
476
+ "grad_norm": 2.203125,
477
+ "learning_rate": 1.8336931089147076e-05,
478
+ "loss": 0.4994,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.272,
483
+ "grad_norm": 2.234375,
484
+ "learning_rate": 1.8259015364714786e-05,
485
+ "loss": 0.4979,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.276,
490
+ "grad_norm": 2.171875,
491
+ "learning_rate": 1.8179489528872808e-05,
492
+ "loss": 0.5281,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.28,
497
+ "grad_norm": 2.28125,
498
+ "learning_rate": 1.80983690853404e-05,
499
+ "loss": 0.5242,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.284,
504
+ "grad_norm": 1.8671875,
505
+ "learning_rate": 1.8015669848708768e-05,
506
+ "loss": 0.5252,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.288,
511
+ "grad_norm": 2.546875,
512
+ "learning_rate": 1.793140794135795e-05,
513
+ "loss": 0.5342,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.292,
518
+ "grad_norm": 2.46875,
519
+ "learning_rate": 1.7845599790313735e-05,
520
+ "loss": 0.5254,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.296,
525
+ "grad_norm": 2.09375,
526
+ "learning_rate": 1.7758262124045195e-05,
527
+ "loss": 0.473,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.3,
532
+ "grad_norm": 3.03125,
533
+ "learning_rate": 1.7669411969203417e-05,
534
+ "loss": 0.5459,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.304,
539
+ "grad_norm": 2.484375,
540
+ "learning_rate": 1.7579066647302134e-05,
541
+ "loss": 0.535,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.308,
546
+ "grad_norm": 2.3125,
547
+ "learning_rate": 1.7487243771340862e-05,
548
+ "loss": 0.5197,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.312,
553
+ "grad_norm": 2.265625,
554
+ "learning_rate": 1.7393961242371203e-05,
555
+ "loss": 0.483,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.316,
560
+ "grad_norm": 2.1875,
561
+ "learning_rate": 1.7299237246007018e-05,
562
+ "loss": 0.5004,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.32,
567
+ "grad_norm": 2.171875,
568
+ "learning_rate": 1.720309024887907e-05,
569
+ "loss": 0.5025,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.324,
574
+ "grad_norm": 2.140625,
575
+ "learning_rate": 1.710553899503496e-05,
576
+ "loss": 0.5244,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.328,
581
+ "grad_norm": 2.21875,
582
+ "learning_rate": 1.700660250228492e-05,
583
+ "loss": 0.4688,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.332,
588
+ "grad_norm": 1.984375,
589
+ "learning_rate": 1.690630005849423e-05,
590
+ "loss": 0.4752,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.336,
595
+ "grad_norm": 2.078125,
596
+ "learning_rate": 1.6804651217823055e-05,
597
+ "loss": 0.5152,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.34,
602
+ "grad_norm": 2.171875,
603
+ "learning_rate": 1.6701675796914284e-05,
604
+ "loss": 0.4875,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.344,
609
+ "grad_norm": 13.625,
610
+ "learning_rate": 1.6597393871030264e-05,
611
+ "loss": 0.4646,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.348,
616
+ "grad_norm": 1.7734375,
617
+ "learning_rate": 1.649182577013906e-05,
618
+ "loss": 0.4816,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.352,
623
+ "grad_norm": 3.421875,
624
+ "learning_rate": 1.6384992074951124e-05,
625
+ "loss": 0.5014,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.356,
630
+ "grad_norm": 4.3125,
631
+ "learning_rate": 1.6276913612907005e-05,
632
+ "loss": 0.4709,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.36,
637
+ "grad_norm": 3.109375,
638
+ "learning_rate": 1.6167611454117027e-05,
639
+ "loss": 0.5275,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.364,
644
+ "grad_norm": 2.296875,
645
+ "learning_rate": 1.6057106907253617e-05,
646
+ "loss": 0.4465,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.368,
651
+ "grad_norm": 2.890625,
652
+ "learning_rate": 1.5945421515397135e-05,
653
+ "loss": 0.4494,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.372,
658
+ "grad_norm": 13.4375,
659
+ "learning_rate": 1.5832577051836016e-05,
660
+ "loss": 0.5127,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.376,
665
+ "grad_norm": 7.15625,
666
+ "learning_rate": 1.5718595515822027e-05,
667
+ "loss": 0.4617,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.38,
672
+ "grad_norm": 2.859375,
673
+ "learning_rate": 1.5603499128281447e-05,
674
+ "loss": 0.4672,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.384,
679
+ "grad_norm": 5.21875,
680
+ "learning_rate": 1.5487310327483087e-05,
681
+ "loss": 0.4311,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.388,
686
+ "grad_norm": 2.09375,
687
+ "learning_rate": 1.5370051764663872e-05,
688
+ "loss": 0.4787,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.392,
693
+ "grad_norm": 2.046875,
694
+ "learning_rate": 1.5251746299612959e-05,
695
+ "loss": 0.4402,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.396,
700
+ "grad_norm": 2.625,
701
+ "learning_rate": 1.5132416996215171e-05,
702
+ "loss": 0.4746,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.4,
707
+ "grad_norm": 1.796875,
708
+ "learning_rate": 1.5012087117954643e-05,
709
+ "loss": 0.4604,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.404,
714
+ "grad_norm": 2.5,
715
+ "learning_rate": 1.4890780123379565e-05,
716
+ "loss": 0.474,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.408,
721
+ "grad_norm": 2.640625,
722
+ "learning_rate": 1.4768519661528879e-05,
723
+ "loss": 0.457,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.412,
728
+ "grad_norm": 2.03125,
729
+ "learning_rate": 1.464532956732188e-05,
730
+ "loss": 0.4199,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.416,
735
+ "grad_norm": 1.9609375,
736
+ "learning_rate": 1.4521233856911507e-05,
737
+ "loss": 0.4207,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.42,
742
+ "grad_norm": 2.109375,
743
+ "learning_rate": 1.43962567230024e-05,
744
+ "loss": 0.4141,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.424,
749
+ "grad_norm": 1.8359375,
750
+ "learning_rate": 1.4270422530134433e-05,
751
+ "loss": 0.4348,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.428,
756
+ "grad_norm": 1.5859375,
757
+ "learning_rate": 1.4143755809932843e-05,
758
+ "loss": 0.4832,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.432,
763
+ "grad_norm": 2.046875,
764
+ "learning_rate": 1.4016281256325702e-05,
765
+ "loss": 0.4232,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.436,
770
+ "grad_norm": 1.9140625,
771
+ "learning_rate": 1.388802372072981e-05,
772
+ "loss": 0.4176,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.44,
777
+ "grad_norm": 2.15625,
778
+ "learning_rate": 1.3759008207205869e-05,
779
+ "loss": 0.4455,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.444,
784
+ "grad_norm": 2.140625,
785
+ "learning_rate": 1.3629259867583864e-05,
786
+ "loss": 0.4148,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.448,
791
+ "grad_norm": 2.578125,
792
+ "learning_rate": 1.349880399655969e-05,
793
+ "loss": 0.4314,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.452,
798
+ "grad_norm": 1.9921875,
799
+ "learning_rate": 1.3367666026763884e-05,
800
+ "loss": 0.4383,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.456,
805
+ "grad_norm": 2.4375,
806
+ "learning_rate": 1.3235871523803496e-05,
807
+ "loss": 0.4256,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.46,
812
+ "grad_norm": 1.8359375,
813
+ "learning_rate": 1.3103446181278015e-05,
814
+ "loss": 0.432,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.464,
819
+ "grad_norm": 1.7890625,
820
+ "learning_rate": 1.297041581577035e-05,
821
+ "loss": 0.4328,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.468,
826
+ "grad_norm": 1.875,
827
+ "learning_rate": 1.2836806361813846e-05,
828
+ "loss": 0.434,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.472,
833
+ "grad_norm": 1.953125,
834
+ "learning_rate": 1.270264386683628e-05,
835
+ "loss": 0.442,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.476,
840
+ "grad_norm": 1.65625,
841
+ "learning_rate": 1.256795448608188e-05,
842
+ "loss": 0.4393,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.48,
847
+ "grad_norm": 1.6484375,
848
+ "learning_rate": 1.2432764477512294e-05,
849
+ "loss": 0.4393,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.484,
854
+ "grad_norm": 2.234375,
855
+ "learning_rate": 1.2297100196687557e-05,
856
+ "loss": 0.4252,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.488,
861
+ "grad_norm": 1.6484375,
862
+ "learning_rate": 1.2160988091628023e-05,
863
+ "loss": 0.423,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.492,
868
+ "grad_norm": 2.328125,
869
+ "learning_rate": 1.202445469765826e-05,
870
+ "loss": 0.408,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.496,
875
+ "grad_norm": 2.171875,
876
+ "learning_rate": 1.1887526632233954e-05,
877
+ "loss": 0.435,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.5,
882
+ "grad_norm": 1.734375,
883
+ "learning_rate": 1.1750230589752763e-05,
884
+ "loss": 0.4137,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.504,
889
+ "grad_norm": 1.4765625,
890
+ "learning_rate": 1.1612593336350209e-05,
891
+ "loss": 0.4123,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.508,
896
+ "grad_norm": 2.265625,
897
+ "learning_rate": 1.1474641704681551e-05,
898
+ "loss": 0.4148,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.512,
903
+ "grad_norm": 1.921875,
904
+ "learning_rate": 1.1336402588690727e-05,
905
+ "loss": 0.4498,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.516,
910
+ "grad_norm": 1.9453125,
911
+ "learning_rate": 1.1197902938367297e-05,
912
+ "loss": 0.4564,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.52,
917
+ "grad_norm": 2.359375,
918
+ "learning_rate": 1.105916975449252e-05,
919
+ "loss": 0.4127,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.524,
924
+ "grad_norm": 1.7265625,
925
+ "learning_rate": 1.0920230083375474e-05,
926
+ "loss": 0.407,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.528,
931
+ "grad_norm": 1.765625,
932
+ "learning_rate": 1.0781111011580336e-05,
933
+ "loss": 0.3959,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.532,
938
+ "grad_norm": 1.875,
939
+ "learning_rate": 1.0641839660645806e-05,
940
+ "loss": 0.4164,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.536,
945
+ "grad_norm": 1.9375,
946
+ "learning_rate": 1.0502443181797696e-05,
947
+ "loss": 0.3768,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.54,
952
+ "grad_norm": 1.7890625,
953
+ "learning_rate": 1.036294875065576e-05,
954
+ "loss": 0.4422,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.544,
959
+ "grad_norm": 2.046875,
960
+ "learning_rate": 1.0223383561935738e-05,
961
+ "loss": 0.4512,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.548,
966
+ "grad_norm": 1.984375,
967
+ "learning_rate": 1.0083774824147707e-05,
968
+ "loss": 0.4141,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.552,
973
+ "grad_norm": 1.6171875,
974
+ "learning_rate": 9.944149754291719e-06,
975
+ "loss": 0.4156,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.556,
980
+ "grad_norm": 1.609375,
981
+ "learning_rate": 9.80453557255179e-06,
982
+ "loss": 0.4055,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.56,
987
+ "grad_norm": 1.890625,
988
+ "learning_rate": 9.664959496989286e-06,
989
+ "loss": 0.4254,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.564,
994
+ "grad_norm": 1.5234375,
995
+ "learning_rate": 9.525448738236691e-06,
996
+ "loss": 0.4359,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.568,
1001
+ "grad_norm": 1.8203125,
1002
+ "learning_rate": 9.386030494192847e-06,
1003
+ "loss": 0.4273,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.572,
1008
+ "grad_norm": 1.984375,
1009
+ "learning_rate": 9.246731944720675e-06,
1010
+ "loss": 0.4553,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.576,
1015
+ "grad_norm": 1.9375,
1016
+ "learning_rate": 9.107580246348395e-06,
1017
+ "loss": 0.4437,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.58,
1022
+ "grad_norm": 1.75,
1023
+ "learning_rate": 8.968602526975329e-06,
1024
+ "loss": 0.4039,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.584,
1029
+ "grad_norm": 1.8203125,
1030
+ "learning_rate": 8.829825880583228e-06,
1031
+ "loss": 0.4287,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.588,
1036
+ "grad_norm": 1.6953125,
1037
+ "learning_rate": 8.69127736195428e-06,
1038
+ "loss": 0.4467,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.592,
1043
+ "grad_norm": 1.5859375,
1044
+ "learning_rate": 8.552983981396709e-06,
1045
+ "loss": 0.4336,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.596,
1050
+ "grad_norm": 1.8984375,
1051
+ "learning_rate": 8.414972699479076e-06,
1052
+ "loss": 0.3941,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.6,
1057
+ "grad_norm": 1.796875,
1058
+ "learning_rate": 8.277270421774234e-06,
1059
+ "loss": 0.4219,
1060
+ "step": 1500
1061
+ }
1062
+ ],
1063
+ "logging_steps": 10,
1064
+ "max_steps": 2500,
1065
+ "num_input_tokens_seen": 0,
1066
+ "num_train_epochs": 9223372036854775807,
1067
+ "save_steps": 500,
1068
+ "stateful_callbacks": {
1069
+ "TrainerControl": {
1070
+ "args": {
1071
+ "should_epoch_stop": false,
1072
+ "should_evaluate": false,
1073
+ "should_log": false,
1074
+ "should_save": true,
1075
+ "should_training_stop": false
1076
+ },
1077
+ "attributes": {}
1078
+ }
1079
+ },
1080
+ "total_flos": 1.43496233091072e+17,
1081
+ "train_batch_size": 32,
1082
+ "trial_name": null,
1083
+ "trial_params": null
1084
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e12bfa8fe909cca94ca597ca0c5eb98c290d61afb2b4b00bd6474df50472db76
3
+ size 5841