Grogros commited on
Commit
e9d0667
·
verified ·
1 Parent(s): 3a888d5

Training in progress, step 500, checkpoint

Browse files
checkpoint-500/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
+ "eos_token_id": 128001,
9
+ "head_dim": 64,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "max_position_embeddings": 131072,
15
+ "mlp_bias": false,
16
+ "model_type": "llama",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 16,
19
+ "num_key_value_heads": 8,
20
+ "pretraining_tp": 1,
21
+ "rms_norm_eps": 1e-05,
22
+ "rope_scaling": {
23
+ "factor": 32.0,
24
+ "high_freq_factor": 4.0,
25
+ "low_freq_factor": 1.0,
26
+ "original_max_position_embeddings": 8192,
27
+ "rope_type": "llama3"
28
+ },
29
+ "rope_theta": 500000.0,
30
+ "tie_word_embeddings": true,
31
+ "torch_dtype": "bfloat16",
32
+ "transformers_version": "4.51.3",
33
+ "use_cache": true,
34
+ "vocab_size": 128256
35
+ }
checkpoint-500/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": "4.51.3"
9
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78da6001b47d391c055a5818b828948ffc98a54a7db85b35fb23fde2847bb04f
3
+ size 2471645608
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad0f56f0c7cc6823787f88f2dbd2271210bc923111bc246032b3a35eb5a421ff
3
+ size 3724602
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
3
+ size 14244
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed756641eefe71867a63e333dcdc6966062a1e1ec00d402455186493e7261e86
3
+ size 1064
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,384 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.5,
6
+ "eval_steps": 500,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.01,
14
+ "grad_norm": 2416.0,
15
+ "learning_rate": 9.000000000000001e-07,
16
+ "loss": 14.5271,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.02,
21
+ "grad_norm": 2624.0,
22
+ "learning_rate": 1.9000000000000002e-06,
23
+ "loss": 10.9556,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.03,
28
+ "grad_norm": 9.0,
29
+ "learning_rate": 2.9e-06,
30
+ "loss": 1.4421,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.04,
35
+ "grad_norm": 4.96875,
36
+ "learning_rate": 3.900000000000001e-06,
37
+ "loss": 0.6727,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.05,
42
+ "grad_norm": 4.96875,
43
+ "learning_rate": 4.9000000000000005e-06,
44
+ "loss": 0.6174,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.06,
49
+ "grad_norm": 8.75,
50
+ "learning_rate": 5.9e-06,
51
+ "loss": 0.6219,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.07,
56
+ "grad_norm": 1.859375,
57
+ "learning_rate": 6.9e-06,
58
+ "loss": 0.5935,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.08,
63
+ "grad_norm": 1.9765625,
64
+ "learning_rate": 7.9e-06,
65
+ "loss": 0.6067,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.09,
70
+ "grad_norm": 1.9765625,
71
+ "learning_rate": 8.900000000000001e-06,
72
+ "loss": 0.5768,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.1,
77
+ "grad_norm": 2.0,
78
+ "learning_rate": 9.9e-06,
79
+ "loss": 0.57,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.11,
84
+ "grad_norm": 1.828125,
85
+ "learning_rate": 9.997532801828659e-06,
86
+ "loss": 0.5519,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.12,
91
+ "grad_norm": 1.796875,
92
+ "learning_rate": 9.989007341460251e-06,
93
+ "loss": 0.5569,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.13,
98
+ "grad_norm": 1.6171875,
99
+ "learning_rate": 9.974403544143942e-06,
100
+ "loss": 0.5436,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.14,
105
+ "grad_norm": 1.546875,
106
+ "learning_rate": 9.953739202357219e-06,
107
+ "loss": 0.5502,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.15,
112
+ "grad_norm": 1.6640625,
113
+ "learning_rate": 9.927039492417452e-06,
114
+ "loss": 0.5441,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.16,
119
+ "grad_norm": 1.578125,
120
+ "learning_rate": 9.894336943808426e-06,
121
+ "loss": 0.5649,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.17,
126
+ "grad_norm": 1.5546875,
127
+ "learning_rate": 9.85567139954818e-06,
128
+ "loss": 0.5599,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.18,
133
+ "grad_norm": 1.71875,
134
+ "learning_rate": 9.811089967646427e-06,
135
+ "loss": 0.5816,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.19,
140
+ "grad_norm": 1.6796875,
141
+ "learning_rate": 9.760646963710694e-06,
142
+ "loss": 0.5367,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.2,
147
+ "grad_norm": 1.6484375,
148
+ "learning_rate": 9.704403844771128e-06,
149
+ "loss": 0.5808,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.21,
154
+ "grad_norm": 1.6015625,
155
+ "learning_rate": 9.642429134404568e-06,
156
+ "loss": 0.5369,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.22,
161
+ "grad_norm": 1.75,
162
+ "learning_rate": 9.574798339249124e-06,
163
+ "loss": 0.5587,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.23,
168
+ "grad_norm": 1.6640625,
169
+ "learning_rate": 9.501593857010968e-06,
170
+ "loss": 0.5843,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.24,
175
+ "grad_norm": 1.6875,
176
+ "learning_rate": 9.42290487607542e-06,
177
+ "loss": 0.573,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.25,
182
+ "grad_norm": 1.6953125,
183
+ "learning_rate": 9.338827266844643e-06,
184
+ "loss": 0.5658,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.26,
189
+ "grad_norm": 1.796875,
190
+ "learning_rate": 9.24946346493432e-06,
191
+ "loss": 0.5526,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.27,
196
+ "grad_norm": 1.5390625,
197
+ "learning_rate": 9.154922346371641e-06,
198
+ "loss": 0.5527,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.28,
203
+ "grad_norm": 1.671875,
204
+ "learning_rate": 9.055319094946633e-06,
205
+ "loss": 0.5505,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.29,
210
+ "grad_norm": 1.578125,
211
+ "learning_rate": 8.950775061878453e-06,
212
+ "loss": 0.5642,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.3,
217
+ "grad_norm": 1.7578125,
218
+ "learning_rate": 8.841417617967618e-06,
219
+ "loss": 0.5479,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.31,
224
+ "grad_norm": 1.546875,
225
+ "learning_rate": 8.727379998414311e-06,
226
+ "loss": 0.5347,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.32,
231
+ "grad_norm": 1.6015625,
232
+ "learning_rate": 8.608801140491811e-06,
233
+ "loss": 0.5488,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.33,
238
+ "grad_norm": 1.75,
239
+ "learning_rate": 8.485825514272824e-06,
240
+ "loss": 0.5565,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.34,
245
+ "grad_norm": 1.6328125,
246
+ "learning_rate": 8.358602946614952e-06,
247
+ "loss": 0.5399,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.35,
252
+ "grad_norm": 1.5546875,
253
+ "learning_rate": 8.227288438619754e-06,
254
+ "loss": 0.5388,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.36,
259
+ "grad_norm": 1.5546875,
260
+ "learning_rate": 8.092041976787772e-06,
261
+ "loss": 0.5432,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.37,
266
+ "grad_norm": 1.6328125,
267
+ "learning_rate": 7.953028338099628e-06,
268
+ "loss": 0.5681,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.38,
273
+ "grad_norm": 1.8515625,
274
+ "learning_rate": 7.810416889260653e-06,
275
+ "loss": 0.5573,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.39,
280
+ "grad_norm": 1.59375,
281
+ "learning_rate": 7.66438138035365e-06,
282
+ "loss": 0.5602,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.4,
287
+ "grad_norm": 1.609375,
288
+ "learning_rate": 7.515099733151177e-06,
289
+ "loss": 0.561,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.41,
294
+ "grad_norm": 1.640625,
295
+ "learning_rate": 7.362753824345271e-06,
296
+ "loss": 0.5423,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.42,
301
+ "grad_norm": 1.484375,
302
+ "learning_rate": 7.207529263958727e-06,
303
+ "loss": 0.5475,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.43,
308
+ "grad_norm": 1.6640625,
309
+ "learning_rate": 7.049615169207864e-06,
310
+ "loss": 0.5531,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.44,
315
+ "grad_norm": 1.5390625,
316
+ "learning_rate": 6.889203934092337e-06,
317
+ "loss": 0.557,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.45,
322
+ "grad_norm": 1.6953125,
323
+ "learning_rate": 6.7264909949926735e-06,
324
+ "loss": 0.5688,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.46,
329
+ "grad_norm": 1.5,
330
+ "learning_rate": 6.561674592561164e-06,
331
+ "loss": 0.5524,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.47,
336
+ "grad_norm": 1.6171875,
337
+ "learning_rate": 6.3949555301961474e-06,
338
+ "loss": 0.5625,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.48,
343
+ "grad_norm": 1.5859375,
344
+ "learning_rate": 6.2265369293940135e-06,
345
+ "loss": 0.5302,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.49,
350
+ "grad_norm": 1.7578125,
351
+ "learning_rate": 6.056623982276945e-06,
352
+ "loss": 0.5375,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.5,
357
+ "grad_norm": 1.6796875,
358
+ "learning_rate": 5.885423701597918e-06,
359
+ "loss": 0.5344,
360
+ "step": 500
361
+ }
362
+ ],
363
+ "logging_steps": 10,
364
+ "max_steps": 1000,
365
+ "num_input_tokens_seen": 0,
366
+ "num_train_epochs": 9223372036854775807,
367
+ "save_steps": 500,
368
+ "stateful_callbacks": {
369
+ "TrainerControl": {
370
+ "args": {
371
+ "should_epoch_stop": false,
372
+ "should_evaluate": false,
373
+ "should_log": false,
374
+ "should_save": true,
375
+ "should_training_stop": false
376
+ },
377
+ "attributes": {}
378
+ }
379
+ },
380
+ "total_flos": 1.91328310788096e+17,
381
+ "train_batch_size": 32,
382
+ "trial_name": null,
383
+ "trial_params": null
384
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8d9759becb77ac102a67be27839f1ea17c0f5678241b4fb13ecf5637fb79622
3
+ size 5304