Mostafa8Mehrabi commited on
Commit
8f4638a
·
verified ·
1 Parent(s): b6915d2

Checkpoint at step 1000 (FP16)

Browse files
checkpoint-1000/config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "head_dim": 128,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 224,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 896,
12
+ "layer_types": [
13
+ "full_attention",
14
+ "full_attention",
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention"
21
+ ],
22
+ "max_position_embeddings": 32768,
23
+ "max_window_layers": 28,
24
+ "model_type": "qwen3",
25
+ "num_attention_heads": 8,
26
+ "num_hidden_layers": 8,
27
+ "num_key_value_heads": 8,
28
+ "rms_norm_eps": 1e-06,
29
+ "rope_scaling": null,
30
+ "rope_theta": 1000000.0,
31
+ "sliding_window": null,
32
+ "tie_word_embeddings": false,
33
+ "torch_dtype": "float16",
34
+ "transformers_version": "4.53.2",
35
+ "use_cache": true,
36
+ "use_sliding_window": false,
37
+ "vocab_size": 50257
38
+ }
checkpoint-1000/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.53.2"
4
+ }
checkpoint-1000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67eb6d181f582ee4d85fcb514ab6f3f159b3029479e5a1f80404ce8c11e6136d
3
+ size 138721680
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45ba53398aa2480a679fa2b7e587dcdfc0bb347eeb9e7a280f1989c460ea03d5
3
+ size 277500235
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
3
+ size 14645
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9aa7e5f7f3366b7db6f25a9e3fa739116674bd641ce589a5940ff73f382fec0c
3
+ size 1383
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1dc8e131ce3fea54b5bbb10599e4aa96737710b736d49fec16ee3b6ee02cacfb
3
+ size 1465
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|endoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<|endoftext|>",
25
+ "lstrip": false,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "50256": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ }
13
+ },
14
+ "bos_token": "<|endoftext|>",
15
+ "clean_up_tokenization_spaces": false,
16
+ "eos_token": "<|endoftext|>",
17
+ "errors": "replace",
18
+ "extra_special_tokens": {},
19
+ "model_max_length": 1024,
20
+ "pad_token": "<|endoftext|>",
21
+ "tokenizer_class": "GPT2Tokenizer",
22
+ "unk_token": "<|endoftext|>"
23
+ }
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,749 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1000,
3
+ "best_metric": 4.4797258377075195,
4
+ "best_model_checkpoint": "./qwen3-30m-tinystories-checkpoints/checkpoint-1000",
5
+ "epoch": 0.05850690381465013,
6
+ "eval_steps": 1000,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 5.850690381465013e-05,
14
+ "grad_norm": Infinity,
15
+ "learning_rate": 0.0,
16
+ "loss": 11.3909,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.0005850690381465013,
21
+ "grad_norm": 74.58138275146484,
22
+ "learning_rate": 6.825273010920438e-08,
23
+ "loss": 11.4747,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.0011701380762930026,
28
+ "grad_norm": 70.98603057861328,
29
+ "learning_rate": 1.6575663026521062e-07,
30
+ "loss": 11.4193,
31
+ "step": 20
32
+ },
33
+ {
34
+ "epoch": 0.0017552071144395038,
35
+ "grad_norm": 67.07854461669922,
36
+ "learning_rate": 2.632605304212169e-07,
37
+ "loss": 11.3702,
38
+ "step": 30
39
+ },
40
+ {
41
+ "epoch": 0.002340276152586005,
42
+ "grad_norm": 65.53945922851562,
43
+ "learning_rate": 3.6076443057722313e-07,
44
+ "loss": 11.1989,
45
+ "step": 40
46
+ },
47
+ {
48
+ "epoch": 0.0029253451907325064,
49
+ "grad_norm": 56.668155670166016,
50
+ "learning_rate": 4.582683307332293e-07,
51
+ "loss": 10.894,
52
+ "step": 50
53
+ },
54
+ {
55
+ "epoch": 0.0035104142288790077,
56
+ "grad_norm": 47.96501159667969,
57
+ "learning_rate": 5.557722308892356e-07,
58
+ "loss": 10.6112,
59
+ "step": 60
60
+ },
61
+ {
62
+ "epoch": 0.004095483267025509,
63
+ "grad_norm": 36.578582763671875,
64
+ "learning_rate": 6.532761310452418e-07,
65
+ "loss": 10.2903,
66
+ "step": 70
67
+ },
68
+ {
69
+ "epoch": 0.00468055230517201,
70
+ "grad_norm": 30.915164947509766,
71
+ "learning_rate": 7.507800312012481e-07,
72
+ "loss": 9.9734,
73
+ "step": 80
74
+ },
75
+ {
76
+ "epoch": 0.0052656213433185115,
77
+ "grad_norm": 24.249662399291992,
78
+ "learning_rate": 8.482839313572544e-07,
79
+ "loss": 9.7175,
80
+ "step": 90
81
+ },
82
+ {
83
+ "epoch": 0.005850690381465013,
84
+ "grad_norm": 17.660844802856445,
85
+ "learning_rate": 9.457878315132606e-07,
86
+ "loss": 9.4751,
87
+ "step": 100
88
+ },
89
+ {
90
+ "epoch": 0.006435759419611514,
91
+ "grad_norm": 14.554573059082031,
92
+ "learning_rate": 1.043291731669267e-06,
93
+ "loss": 9.2603,
94
+ "step": 110
95
+ },
96
+ {
97
+ "epoch": 0.007020828457758015,
98
+ "grad_norm": 12.098219871520996,
99
+ "learning_rate": 1.140795631825273e-06,
100
+ "loss": 9.0319,
101
+ "step": 120
102
+ },
103
+ {
104
+ "epoch": 0.007605897495904517,
105
+ "grad_norm": 10.675163269042969,
106
+ "learning_rate": 1.2382995319812794e-06,
107
+ "loss": 8.8657,
108
+ "step": 130
109
+ },
110
+ {
111
+ "epoch": 0.008190966534051018,
112
+ "grad_norm": 9.221511840820312,
113
+ "learning_rate": 1.3358034321372856e-06,
114
+ "loss": 8.6798,
115
+ "step": 140
116
+ },
117
+ {
118
+ "epoch": 0.00877603557219752,
119
+ "grad_norm": 8.275217056274414,
120
+ "learning_rate": 1.4333073322932917e-06,
121
+ "loss": 8.5072,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 0.00936110461034402,
126
+ "grad_norm": 7.335099220275879,
127
+ "learning_rate": 1.5308112324492981e-06,
128
+ "loss": 8.3525,
129
+ "step": 160
130
+ },
131
+ {
132
+ "epoch": 0.009946173648490523,
133
+ "grad_norm": 6.6783833503723145,
134
+ "learning_rate": 1.6283151326053041e-06,
135
+ "loss": 8.2143,
136
+ "step": 170
137
+ },
138
+ {
139
+ "epoch": 0.010531242686637023,
140
+ "grad_norm": 5.890490531921387,
141
+ "learning_rate": 1.7258190327613103e-06,
142
+ "loss": 8.1013,
143
+ "step": 180
144
+ },
145
+ {
146
+ "epoch": 0.011116311724783525,
147
+ "grad_norm": 5.453526496887207,
148
+ "learning_rate": 1.823322932917317e-06,
149
+ "loss": 7.9714,
150
+ "step": 190
151
+ },
152
+ {
153
+ "epoch": 0.011701380762930026,
154
+ "grad_norm": 4.6703081130981445,
155
+ "learning_rate": 1.920826833073323e-06,
156
+ "loss": 7.8631,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 0.012286449801076528,
161
+ "grad_norm": 4.414166450500488,
162
+ "learning_rate": 2.0183307332293293e-06,
163
+ "loss": 7.7655,
164
+ "step": 210
165
+ },
166
+ {
167
+ "epoch": 0.012871518839223028,
168
+ "grad_norm": 4.081826686859131,
169
+ "learning_rate": 2.1158346333853355e-06,
170
+ "loss": 7.6717,
171
+ "step": 220
172
+ },
173
+ {
174
+ "epoch": 0.01345658787736953,
175
+ "grad_norm": 4.213140487670898,
176
+ "learning_rate": 2.2133385335413417e-06,
177
+ "loss": 7.5924,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 0.01404165691551603,
182
+ "grad_norm": 3.859170436859131,
183
+ "learning_rate": 2.310842433697348e-06,
184
+ "loss": 7.511,
185
+ "step": 240
186
+ },
187
+ {
188
+ "epoch": 0.014626725953662533,
189
+ "grad_norm": 3.9012818336486816,
190
+ "learning_rate": 2.408346333853354e-06,
191
+ "loss": 7.4278,
192
+ "step": 250
193
+ },
194
+ {
195
+ "epoch": 0.015211794991809033,
196
+ "grad_norm": 3.8075311183929443,
197
+ "learning_rate": 2.5058502340093603e-06,
198
+ "loss": 7.3492,
199
+ "step": 260
200
+ },
201
+ {
202
+ "epoch": 0.015796864029955535,
203
+ "grad_norm": 3.5343337059020996,
204
+ "learning_rate": 2.603354134165367e-06,
205
+ "loss": 7.2719,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 0.016381933068102036,
210
+ "grad_norm": 3.5580379962921143,
211
+ "learning_rate": 2.700858034321373e-06,
212
+ "loss": 7.2373,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 0.016967002106248536,
217
+ "grad_norm": 3.4421544075012207,
218
+ "learning_rate": 2.7983619344773792e-06,
219
+ "loss": 7.1697,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 0.01755207114439504,
224
+ "grad_norm": 3.7167348861694336,
225
+ "learning_rate": 2.8958658346333854e-06,
226
+ "loss": 7.0948,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 0.01813714018254154,
231
+ "grad_norm": 4.353510856628418,
232
+ "learning_rate": 2.993369734789392e-06,
233
+ "loss": 7.0373,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 0.01872220922068804,
238
+ "grad_norm": 4.191625118255615,
239
+ "learning_rate": 3.0908736349453982e-06,
240
+ "loss": 6.9602,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 0.01930727825883454,
245
+ "grad_norm": 4.205361843109131,
246
+ "learning_rate": 3.188377535101404e-06,
247
+ "loss": 6.903,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 0.019892347296981045,
252
+ "grad_norm": 4.100222110748291,
253
+ "learning_rate": 3.28588143525741e-06,
254
+ "loss": 6.8439,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 0.020477416335127546,
259
+ "grad_norm": 4.314079284667969,
260
+ "learning_rate": 3.3833853354134164e-06,
261
+ "loss": 6.7787,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 0.021062485373274046,
266
+ "grad_norm": 6.652833938598633,
267
+ "learning_rate": 3.4808892355694226e-06,
268
+ "loss": 6.7226,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 0.021647554411420546,
273
+ "grad_norm": 4.056787014007568,
274
+ "learning_rate": 3.5783931357254296e-06,
275
+ "loss": 6.664,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 0.02223262344956705,
280
+ "grad_norm": 4.815715312957764,
281
+ "learning_rate": 3.675897035881436e-06,
282
+ "loss": 6.6125,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 0.02281769248771355,
287
+ "grad_norm": 4.876301288604736,
288
+ "learning_rate": 3.773400936037442e-06,
289
+ "loss": 6.5621,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 0.02340276152586005,
294
+ "grad_norm": 4.692805290222168,
295
+ "learning_rate": 3.870904836193448e-06,
296
+ "loss": 6.5075,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 0.02398783056400655,
301
+ "grad_norm": 4.293951034545898,
302
+ "learning_rate": 3.968408736349454e-06,
303
+ "loss": 6.4574,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 0.024572899602153055,
308
+ "grad_norm": 4.072322845458984,
309
+ "learning_rate": 4.06591263650546e-06,
310
+ "loss": 6.3977,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 0.025157968640299556,
315
+ "grad_norm": 5.009401321411133,
316
+ "learning_rate": 4.163416536661466e-06,
317
+ "loss": 6.3332,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 0.025743037678446056,
322
+ "grad_norm": 4.7960286140441895,
323
+ "learning_rate": 4.2609204368174725e-06,
324
+ "loss": 6.2778,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 0.026328106716592557,
329
+ "grad_norm": 3.423039197921753,
330
+ "learning_rate": 4.358424336973479e-06,
331
+ "loss": 6.2403,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 0.02691317575473906,
336
+ "grad_norm": 4.546107292175293,
337
+ "learning_rate": 4.455928237129486e-06,
338
+ "loss": 6.1845,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 0.02749824479288556,
343
+ "grad_norm": 5.357204914093018,
344
+ "learning_rate": 4.553432137285492e-06,
345
+ "loss": 6.1027,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 0.02808331383103206,
350
+ "grad_norm": 4.221200942993164,
351
+ "learning_rate": 4.650936037441498e-06,
352
+ "loss": 6.0522,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 0.02866838286917856,
357
+ "grad_norm": 3.5187366008758545,
358
+ "learning_rate": 4.748439937597504e-06,
359
+ "loss": 6.0063,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 0.029253451907325066,
364
+ "grad_norm": 4.514864444732666,
365
+ "learning_rate": 4.8459438377535105e-06,
366
+ "loss": 5.9795,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.029838520945471566,
371
+ "grad_norm": 3.3947811126708984,
372
+ "learning_rate": 4.943447737909517e-06,
373
+ "loss": 5.9023,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.030423589983618066,
378
+ "grad_norm": 5.161691665649414,
379
+ "learning_rate": 5.040951638065523e-06,
380
+ "loss": 5.8573,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.031008659021764567,
385
+ "grad_norm": 5.400602340698242,
386
+ "learning_rate": 5.138455538221529e-06,
387
+ "loss": 5.8028,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.03159372805991107,
392
+ "grad_norm": 4.410223007202148,
393
+ "learning_rate": 5.235959438377535e-06,
394
+ "loss": 5.7999,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.03217879709805757,
399
+ "grad_norm": 3.708657741546631,
400
+ "learning_rate": 5.3334633385335414e-06,
401
+ "loss": 5.7253,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.03276386613620407,
406
+ "grad_norm": 5.148003101348877,
407
+ "learning_rate": 5.430967238689548e-06,
408
+ "loss": 5.6967,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.03334893517435057,
413
+ "grad_norm": 5.288032054901123,
414
+ "learning_rate": 5.528471138845554e-06,
415
+ "loss": 5.6411,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.03393400421249707,
420
+ "grad_norm": 3.524038314819336,
421
+ "learning_rate": 5.62597503900156e-06,
422
+ "loss": 5.6076,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.03451907325064357,
427
+ "grad_norm": 3.5057692527770996,
428
+ "learning_rate": 5.723478939157566e-06,
429
+ "loss": 5.5643,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.03510414228879008,
434
+ "grad_norm": 4.593623161315918,
435
+ "learning_rate": 5.820982839313572e-06,
436
+ "loss": 5.5258,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.03568921132693658,
441
+ "grad_norm": 3.617720127105713,
442
+ "learning_rate": 5.918486739469579e-06,
443
+ "loss": 5.4787,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.03627428036508308,
448
+ "grad_norm": 3.6132466793060303,
449
+ "learning_rate": 6.015990639625586e-06,
450
+ "loss": 5.4579,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.03685934940322958,
455
+ "grad_norm": 4.541604518890381,
456
+ "learning_rate": 6.113494539781592e-06,
457
+ "loss": 5.4188,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.03744441844137608,
462
+ "grad_norm": 4.4366254806518555,
463
+ "learning_rate": 6.210998439937598e-06,
464
+ "loss": 5.3733,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.03802948747952258,
469
+ "grad_norm": 4.232390403747559,
470
+ "learning_rate": 6.308502340093604e-06,
471
+ "loss": 5.3406,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.03861455651766908,
476
+ "grad_norm": 3.4591336250305176,
477
+ "learning_rate": 6.4060062402496095e-06,
478
+ "loss": 5.2955,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.03919962555581558,
483
+ "grad_norm": 5.510738849639893,
484
+ "learning_rate": 6.5035101404056166e-06,
485
+ "loss": 5.2574,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.03978469459396209,
490
+ "grad_norm": 5.914499282836914,
491
+ "learning_rate": 6.601014040561624e-06,
492
+ "loss": 5.2101,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.04036976363210859,
497
+ "grad_norm": 4.961345672607422,
498
+ "learning_rate": 6.698517940717629e-06,
499
+ "loss": 5.2078,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.04095483267025509,
504
+ "grad_norm": 4.458784580230713,
505
+ "learning_rate": 6.796021840873636e-06,
506
+ "loss": 5.1342,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.04153990170840159,
511
+ "grad_norm": 4.344489574432373,
512
+ "learning_rate": 6.893525741029641e-06,
513
+ "loss": 5.1115,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.04212497074654809,
518
+ "grad_norm": 3.6247169971466064,
519
+ "learning_rate": 6.991029641185648e-06,
520
+ "loss": 5.1027,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.04271003978469459,
525
+ "grad_norm": 4.234716415405273,
526
+ "learning_rate": 7.088533541341654e-06,
527
+ "loss": 5.0847,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.04329510882284109,
532
+ "grad_norm": 3.6908812522888184,
533
+ "learning_rate": 7.18603744149766e-06,
534
+ "loss": 5.0225,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.04388017786098759,
539
+ "grad_norm": 4.514861583709717,
540
+ "learning_rate": 7.283541341653667e-06,
541
+ "loss": 5.0108,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.0444652468991341,
546
+ "grad_norm": 4.391042709350586,
547
+ "learning_rate": 7.381045241809672e-06,
548
+ "loss": 4.9743,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.0450503159372806,
553
+ "grad_norm": 4.511202335357666,
554
+ "learning_rate": 7.478549141965679e-06,
555
+ "loss": 4.964,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.0456353849754271,
560
+ "grad_norm": 4.672698497772217,
561
+ "learning_rate": 7.576053042121685e-06,
562
+ "loss": 4.9322,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.0462204540135736,
567
+ "grad_norm": 5.7832207679748535,
568
+ "learning_rate": 7.673556942277693e-06,
569
+ "loss": 4.9004,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.0468055230517201,
574
+ "grad_norm": 5.457101345062256,
575
+ "learning_rate": 7.771060842433697e-06,
576
+ "loss": 4.8727,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 0.0473905920898666,
581
+ "grad_norm": 5.284400463104248,
582
+ "learning_rate": 7.868564742589705e-06,
583
+ "loss": 4.8428,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 0.0479756611280131,
588
+ "grad_norm": 5.2293829917907715,
589
+ "learning_rate": 7.96606864274571e-06,
590
+ "loss": 4.8219,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 0.04856073016615961,
595
+ "grad_norm": 4.629702091217041,
596
+ "learning_rate": 8.063572542901716e-06,
597
+ "loss": 4.7985,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 0.04914579920430611,
602
+ "grad_norm": 4.912006378173828,
603
+ "learning_rate": 8.161076443057723e-06,
604
+ "loss": 4.8071,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 0.04973086824245261,
609
+ "grad_norm": 4.5750346183776855,
610
+ "learning_rate": 8.258580343213728e-06,
611
+ "loss": 4.7561,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 0.05031593728059911,
616
+ "grad_norm": 4.587753772735596,
617
+ "learning_rate": 8.356084243369736e-06,
618
+ "loss": 4.7323,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 0.05090100631874561,
623
+ "grad_norm": 4.676549434661865,
624
+ "learning_rate": 8.45358814352574e-06,
625
+ "loss": 4.6787,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 0.05148607535689211,
630
+ "grad_norm": 4.039685249328613,
631
+ "learning_rate": 8.551092043681748e-06,
632
+ "loss": 4.6808,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 0.05207114439503861,
637
+ "grad_norm": 4.448736667633057,
638
+ "learning_rate": 8.648595943837753e-06,
639
+ "loss": 4.6849,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 0.05265621343318511,
644
+ "grad_norm": 4.613405227661133,
645
+ "learning_rate": 8.74609984399376e-06,
646
+ "loss": 4.6309,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 0.05324128247133162,
651
+ "grad_norm": 4.352312088012695,
652
+ "learning_rate": 8.843603744149765e-06,
653
+ "loss": 4.6045,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 0.05382635150947812,
658
+ "grad_norm": 7.403280258178711,
659
+ "learning_rate": 8.941107644305773e-06,
660
+ "loss": 4.6065,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 0.05441142054762462,
665
+ "grad_norm": 6.342307090759277,
666
+ "learning_rate": 9.03861154446178e-06,
667
+ "loss": 4.5865,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 0.05499648958577112,
672
+ "grad_norm": 4.280608177185059,
673
+ "learning_rate": 9.136115444617785e-06,
674
+ "loss": 4.5471,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 0.05558155862391762,
679
+ "grad_norm": 5.916503429412842,
680
+ "learning_rate": 9.233619344773792e-06,
681
+ "loss": 4.5704,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 0.05616662766206412,
686
+ "grad_norm": 5.4643940925598145,
687
+ "learning_rate": 9.331123244929798e-06,
688
+ "loss": 4.5332,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 0.05675169670021062,
693
+ "grad_norm": 6.90775728225708,
694
+ "learning_rate": 9.428627145085804e-06,
695
+ "loss": 4.5466,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 0.05733676573835712,
700
+ "grad_norm": 4.458855152130127,
701
+ "learning_rate": 9.52613104524181e-06,
702
+ "loss": 4.4947,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 0.05792183477650363,
707
+ "grad_norm": 4.308935642242432,
708
+ "learning_rate": 9.623634945397816e-06,
709
+ "loss": 4.4815,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 0.05850690381465013,
714
+ "grad_norm": 4.152543544769287,
715
+ "learning_rate": 9.721138845553823e-06,
716
+ "loss": 4.4704,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 0.05850690381465013,
721
+ "eval_loss": 4.4797258377075195,
722
+ "eval_runtime": 32.8353,
723
+ "eval_samples_per_second": 687.889,
724
+ "eval_steps_per_second": 5.391,
725
+ "step": 1000
726
+ }
727
+ ],
728
+ "logging_steps": 10,
729
+ "max_steps": 51276,
730
+ "num_input_tokens_seen": 0,
731
+ "num_train_epochs": 3,
732
+ "save_steps": 1000,
733
+ "stateful_callbacks": {
734
+ "TrainerControl": {
735
+ "args": {
736
+ "should_epoch_stop": false,
737
+ "should_evaluate": false,
738
+ "should_log": false,
739
+ "should_save": true,
740
+ "should_training_stop": false
741
+ },
742
+ "attributes": {}
743
+ }
744
+ },
745
+ "total_flos": 9201055137398784.0,
746
+ "train_batch_size": 128,
747
+ "trial_name": null,
748
+ "trial_params": null
749
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ab409182da5adbcb18002f22d82b6427b8d967f738680c7220c631231212807
3
+ size 5841
checkpoint-1000/vocab.json ADDED
The diff for this file is too large to render. See raw diff