Grogros commited on
Commit
ca21b48
·
verified ·
1 Parent(s): e42da4d

Training in progress, step 500, checkpoint

Browse files
checkpoint-500/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "meta-llama/Llama-3.2-1B-Instruct",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 128000,
9
+ "eos_token_id": [
10
+ 128001,
11
+ 128008,
12
+ 128009
13
+ ],
14
+ "head_dim": 64,
15
+ "hidden_act": "silu",
16
+ "hidden_size": 2048,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 8192,
19
+ "max_position_embeddings": 131072,
20
+ "mlp_bias": false,
21
+ "model_type": "llama",
22
+ "num_attention_heads": 32,
23
+ "num_hidden_layers": 16,
24
+ "num_key_value_heads": 8,
25
+ "pretraining_tp": 1,
26
+ "rms_norm_eps": 1e-05,
27
+ "rope_scaling": {
28
+ "factor": 32.0,
29
+ "high_freq_factor": 4.0,
30
+ "low_freq_factor": 1.0,
31
+ "original_max_position_embeddings": 8192,
32
+ "rope_type": "llama3"
33
+ },
34
+ "rope_theta": 500000.0,
35
+ "tie_word_embeddings": true,
36
+ "torch_dtype": "float32",
37
+ "transformers_version": "4.47.1",
38
+ "use_cache": true,
39
+ "vocab_size": 128256
40
+ }
checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 128000,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 128001,
6
+ 128008,
7
+ 128009
8
+ ],
9
+ "temperature": 0.6,
10
+ "top_p": 0.9,
11
+ "transformers_version": "4.47.1"
12
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d42adc38d2c6af11df04b7a24eb9807a75d1e25ca016c9037a136bdb2d7e4cb
3
+ size 4943274328
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3aee0f0b3e077f3e136ce0effdfef20752cdd9b27aa84d8eafe1e2a024f223b2
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:ba107b43cbf135d7adec23830c5439405fafa5da00d923edeb699b9125bd9c2a
3
+ size 14244
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31f20d3fd8f2ca072c88294e1b245d2fc0feec4c84032b4ae4d3711181320962
3
+ size 1064
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,383 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.03210704488765544,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0006421408977531089,
13
+ "grad_norm": 868.92626953125,
14
+ "learning_rate": 8.000000000000001e-07,
15
+ "loss": 4.2813,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.0012842817955062177,
20
+ "grad_norm": 603.6827392578125,
21
+ "learning_rate": 1.6000000000000001e-06,
22
+ "loss": 5.7894,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0019264226932593266,
27
+ "grad_norm": 461.2832946777344,
28
+ "learning_rate": 2.4000000000000003e-06,
29
+ "loss": 5.5449,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.0025685635910124354,
34
+ "grad_norm": 186.7425537109375,
35
+ "learning_rate": 3.2000000000000003e-06,
36
+ "loss": 4.8136,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.0032107044887655445,
41
+ "grad_norm": 147.78977966308594,
42
+ "learning_rate": 4.000000000000001e-06,
43
+ "loss": 3.5293,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.003852845386518653,
48
+ "grad_norm": 113.7726821899414,
49
+ "learning_rate": 4.800000000000001e-06,
50
+ "loss": 3.1336,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.004494986284271762,
55
+ "grad_norm": 82.38671112060547,
56
+ "learning_rate": 5.600000000000001e-06,
57
+ "loss": 2.1499,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.005137127182024871,
62
+ "grad_norm": 95.98859405517578,
63
+ "learning_rate": 6.4000000000000006e-06,
64
+ "loss": 1.7044,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.00577926807977798,
69
+ "grad_norm": 103.78849029541016,
70
+ "learning_rate": 7.2000000000000005e-06,
71
+ "loss": 1.6203,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.006421408977531089,
76
+ "grad_norm": 130.01173400878906,
77
+ "learning_rate": 8.000000000000001e-06,
78
+ "loss": 1.4757,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.007063549875284197,
83
+ "grad_norm": 61.1568717956543,
84
+ "learning_rate": 8.8e-06,
85
+ "loss": 2.2121,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.007705690773037306,
90
+ "grad_norm": 142.62966918945312,
91
+ "learning_rate": 9.600000000000001e-06,
92
+ "loss": 1.4566,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.008347831670790415,
97
+ "grad_norm": 69.8824691772461,
98
+ "learning_rate": 1.04e-05,
99
+ "loss": 1.1926,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.008989972568543524,
104
+ "grad_norm": 103.1029052734375,
105
+ "learning_rate": 1.1200000000000001e-05,
106
+ "loss": 1.2254,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.009632113466296633,
111
+ "grad_norm": 107.92010498046875,
112
+ "learning_rate": 1.2e-05,
113
+ "loss": 1.2146,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.010274254364049742,
118
+ "grad_norm": 108.57059478759766,
119
+ "learning_rate": 1.2800000000000001e-05,
120
+ "loss": 1.1926,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.01091639526180285,
125
+ "grad_norm": 98.63533020019531,
126
+ "learning_rate": 1.3600000000000002e-05,
127
+ "loss": 1.1101,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.01155853615955596,
132
+ "grad_norm": 171.6598358154297,
133
+ "learning_rate": 1.4400000000000001e-05,
134
+ "loss": 1.1139,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.012200677057309069,
139
+ "grad_norm": 68.86021423339844,
140
+ "learning_rate": 1.5200000000000002e-05,
141
+ "loss": 1.1172,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.012842817955062178,
146
+ "grad_norm": 207.02130126953125,
147
+ "learning_rate": 1.6000000000000003e-05,
148
+ "loss": 1.2177,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.013484958852815285,
153
+ "grad_norm": 76.22872924804688,
154
+ "learning_rate": 1.6800000000000002e-05,
155
+ "loss": 1.1397,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.014127099750568394,
160
+ "grad_norm": 48.84059524536133,
161
+ "learning_rate": 1.76e-05,
162
+ "loss": 1.1708,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.014769240648321504,
167
+ "grad_norm": 93.089599609375,
168
+ "learning_rate": 1.8400000000000003e-05,
169
+ "loss": 1.1412,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.015411381546074613,
174
+ "grad_norm": 80.01580047607422,
175
+ "learning_rate": 1.9200000000000003e-05,
176
+ "loss": 1.0826,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.01605352244382772,
181
+ "grad_norm": 80.10456085205078,
182
+ "learning_rate": 2e-05,
183
+ "loss": 1.1087,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.01669566334158083,
188
+ "grad_norm": 73.42576599121094,
189
+ "learning_rate": 1.9999025240093045e-05,
190
+ "loss": 1.1167,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.01733780423933394,
195
+ "grad_norm": 56.622398376464844,
196
+ "learning_rate": 1.9996101150403543e-05,
197
+ "loss": 1.1225,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.017979945137087047,
202
+ "grad_norm": 57.88623809814453,
203
+ "learning_rate": 1.9991228300988586e-05,
204
+ "loss": 1.0063,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.018622086034840158,
209
+ "grad_norm": 65.95738983154297,
210
+ "learning_rate": 1.9984407641819812e-05,
211
+ "loss": 1.0361,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.019264226932593265,
216
+ "grad_norm": 64.07682800292969,
217
+ "learning_rate": 1.9975640502598243e-05,
218
+ "loss": 1.2384,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.019906367830346376,
223
+ "grad_norm": 51.745906829833984,
224
+ "learning_rate": 1.9964928592495046e-05,
225
+ "loss": 0.9249,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.020548508728099484,
230
+ "grad_norm": 39.79248046875,
231
+ "learning_rate": 1.9952273999818312e-05,
232
+ "loss": 1.0824,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.02119064962585259,
237
+ "grad_norm": 58.02740478515625,
238
+ "learning_rate": 1.9937679191605964e-05,
239
+ "loss": 0.9394,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.0218327905236057,
244
+ "grad_norm": 54.71721649169922,
245
+ "learning_rate": 1.9921147013144782e-05,
246
+ "loss": 0.7972,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.02247493142135881,
251
+ "grad_norm": 42.93053436279297,
252
+ "learning_rate": 1.9902680687415704e-05,
253
+ "loss": 0.9798,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.02311707231911192,
258
+ "grad_norm": 49.934303283691406,
259
+ "learning_rate": 1.988228381446553e-05,
260
+ "loss": 0.8748,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.023759213216865027,
265
+ "grad_norm": 59.80929183959961,
266
+ "learning_rate": 1.985996037070505e-05,
267
+ "loss": 0.9775,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.024401354114618138,
272
+ "grad_norm": 30.937265396118164,
273
+ "learning_rate": 1.983571470813386e-05,
274
+ "loss": 0.9484,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.025043495012371245,
279
+ "grad_norm": 42.207393646240234,
280
+ "learning_rate": 1.9809551553491918e-05,
281
+ "loss": 0.749,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.025685635910124356,
286
+ "grad_norm": 49.683250427246094,
287
+ "learning_rate": 1.9781476007338058e-05,
288
+ "loss": 0.7524,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.026327776807877463,
293
+ "grad_norm": 61.64522171020508,
294
+ "learning_rate": 1.9751493543055634e-05,
295
+ "loss": 0.7272,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.02696991770563057,
300
+ "grad_norm": 56.983917236328125,
301
+ "learning_rate": 1.9719610005785466e-05,
302
+ "loss": 0.8223,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.02761205860338368,
307
+ "grad_norm": 31.799327850341797,
308
+ "learning_rate": 1.9685831611286312e-05,
309
+ "loss": 0.7497,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.02825419950113679,
314
+ "grad_norm": 45.978851318359375,
315
+ "learning_rate": 1.9650164944723116e-05,
316
+ "loss": 0.7766,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.0288963403988899,
321
+ "grad_norm": 45.03113555908203,
322
+ "learning_rate": 1.961261695938319e-05,
323
+ "loss": 0.7855,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.029538481296643007,
328
+ "grad_norm": 28.83675193786621,
329
+ "learning_rate": 1.9573194975320672e-05,
330
+ "loss": 0.7067,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.030180622194396118,
335
+ "grad_norm": 39.01563262939453,
336
+ "learning_rate": 1.9531906677929472e-05,
337
+ "loss": 0.774,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.030822763092149225,
342
+ "grad_norm": 30.023176193237305,
343
+ "learning_rate": 1.9488760116444966e-05,
344
+ "loss": 0.7703,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.031464903989902336,
349
+ "grad_norm": 41.739227294921875,
350
+ "learning_rate": 1.944376370237481e-05,
351
+ "loss": 0.7466,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.03210704488765544,
356
+ "grad_norm": 19.699201583862305,
357
+ "learning_rate": 1.9396926207859085e-05,
358
+ "loss": 0.7279,
359
+ "step": 500
360
+ }
361
+ ],
362
+ "logging_steps": 10,
363
+ "max_steps": 2500,
364
+ "num_input_tokens_seen": 0,
365
+ "num_train_epochs": 1,
366
+ "save_steps": 500,
367
+ "stateful_callbacks": {
368
+ "TrainerControl": {
369
+ "args": {
370
+ "should_epoch_stop": false,
371
+ "should_evaluate": false,
372
+ "should_log": false,
373
+ "should_save": true,
374
+ "should_training_stop": false
375
+ },
376
+ "attributes": {}
377
+ }
378
+ },
379
+ "total_flos": 1.91328310788096e+17,
380
+ "train_batch_size": 1,
381
+ "trial_name": null,
382
+ "trial_params": null
383
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c4b49420d674cf9733abd42c9ce9cdc85a680255b470678b0c84c01862d660b
3
+ size 5432