LH-Tech-AI commited on
Commit
6b6b373
·
verified ·
1 Parent(s): ddf03d3

Training in progress, step 6000, checkpoint

Browse files
checkpoint-6000/config.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "float32",
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2048,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention"
28
+ ],
29
+ "max_position_embeddings": 2048,
30
+ "max_window_layers": 28,
31
+ "model_type": "qwen3",
32
+ "num_attention_heads": 12,
33
+ "num_hidden_layers": 12,
34
+ "num_key_value_heads": 4,
35
+ "pad_token_id": 1,
36
+ "rms_norm_eps": 1e-06,
37
+ "rope_parameters": {
38
+ "rope_theta": 10000,
39
+ "rope_type": "default"
40
+ },
41
+ "sliding_window": null,
42
+ "tie_word_embeddings": true,
43
+ "transformers_version": "5.14.1",
44
+ "use_cache": false,
45
+ "use_sliding_window": false,
46
+ "vocab_size": 32768
47
+ }
checkpoint-6000/generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 2,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "pad_token_id": 1,
8
+ "transformers_version": "5.14.1",
9
+ "use_cache": true
10
+ }
checkpoint-6000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55bece4cc9901c2e9b0ad84f92c8f7bf0030e25b7f97299cd4ee044a40781067
3
+ size 402750928
checkpoint-6000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:811e9a7faaf4d9ecbc1bb5d52a4cc2934c4de8b311d66611b820988e48292656
3
+ size 805587275
checkpoint-6000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ffffd41cb5466a23271a8c6303a885c8e75cf7a03496c0a7cb7f448827155d8
3
+ size 14645
checkpoint-6000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16654f369df2cbedc75c822565d79ed045c0683027f19b82b57fe6b856571ed6
3
+ size 1465
checkpoint-6000/trainer_state.json ADDED
@@ -0,0 +1,890 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.05242879105215299,
6
+ "eval_steps": 3000,
7
+ "global_step": 6000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.00043690659210127495,
14
+ "grad_norm": 1.8826903104782104,
15
+ "learning_rate": 2.4500000000000003e-05,
16
+ "loss": 9.762939453125,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.0008738131842025499,
21
+ "grad_norm": 0.9995923042297363,
22
+ "learning_rate": 4.9500000000000004e-05,
23
+ "loss": 8.305597534179688,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.001310719776303825,
28
+ "grad_norm": 0.9061496257781982,
29
+ "learning_rate": 7.45e-05,
30
+ "loss": 7.173495483398438,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.0017476263684050998,
35
+ "grad_norm": 1.6024847030639648,
36
+ "learning_rate": 9.95e-05,
37
+ "loss": 6.624603271484375,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.002184532960506375,
42
+ "grad_norm": 1.2123334407806396,
43
+ "learning_rate": 0.0001245,
44
+ "loss": 6.243399658203125,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.00262143955260765,
49
+ "grad_norm": 1.0432322025299072,
50
+ "learning_rate": 0.0001495,
51
+ "loss": 5.9879296875,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.0030583461447089245,
56
+ "grad_norm": 0.9280146360397339,
57
+ "learning_rate": 0.00017449999999999999,
58
+ "loss": 5.776607666015625,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.0034952527368101996,
63
+ "grad_norm": 1.0289900302886963,
64
+ "learning_rate": 0.00019950000000000002,
65
+ "loss": 5.5920703125,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.003932159328911475,
70
+ "grad_norm": 1.0593407154083252,
71
+ "learning_rate": 0.0002245,
72
+ "loss": 5.437271728515625,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.00436906592101275,
77
+ "grad_norm": 1.0122662782669067,
78
+ "learning_rate": 0.0002495,
79
+ "loss": 5.287777709960937,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.004805972513114025,
84
+ "grad_norm": 0.7662743926048279,
85
+ "learning_rate": 0.0002745,
86
+ "loss": 5.155453491210937,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.0052428791052153,
91
+ "grad_norm": 0.7941539287567139,
92
+ "learning_rate": 0.0002995,
93
+ "loss": 5.030631713867187,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.005679785697316574,
98
+ "grad_norm": 0.6195730566978455,
99
+ "learning_rate": 0.00032450000000000003,
100
+ "loss": 4.909389953613282,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.006116692289417849,
105
+ "grad_norm": 0.6587674617767334,
106
+ "learning_rate": 0.0003495,
107
+ "loss": 4.786273498535156,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.006553598881519124,
112
+ "grad_norm": 0.7233771681785583,
113
+ "learning_rate": 0.0003745,
114
+ "loss": 4.6648342895507815,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.006990505473620399,
119
+ "grad_norm": 0.5489601492881775,
120
+ "learning_rate": 0.0003995,
121
+ "loss": 4.555048522949218,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.007427412065721674,
126
+ "grad_norm": 0.6154086589813232,
127
+ "learning_rate": 0.0004245,
128
+ "loss": 4.461051330566407,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.00786431865782295,
133
+ "grad_norm": 0.41333407163619995,
134
+ "learning_rate": 0.00044950000000000003,
135
+ "loss": 4.374936828613281,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.008301225249924224,
140
+ "grad_norm": 0.5036063194274902,
141
+ "learning_rate": 0.0004745,
142
+ "loss": 4.3033209228515625,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.0087381318420255,
147
+ "grad_norm": 0.3687891364097595,
148
+ "learning_rate": 0.0004995,
149
+ "loss": 4.233715209960938,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.009175038434126774,
154
+ "grad_norm": 0.5433846116065979,
155
+ "learning_rate": 0.0005245,
156
+ "loss": 4.186253967285157,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.00961194502622805,
161
+ "grad_norm": 0.439351350069046,
162
+ "learning_rate": 0.0005495,
163
+ "loss": 4.139383850097656,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.010048851618329325,
168
+ "grad_norm": 0.36215880513191223,
169
+ "learning_rate": 0.0005745,
170
+ "loss": 4.08899658203125,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.0104857582104306,
175
+ "grad_norm": 0.3970701992511749,
176
+ "learning_rate": 0.0005995000000000001,
177
+ "loss": 4.061238403320313,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.010922664802531873,
182
+ "grad_norm": 0.3410918712615967,
183
+ "learning_rate": 0.0006245000000000001,
184
+ "loss": 4.026336669921875,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 0.011359571394633148,
189
+ "grad_norm": 0.3047930598258972,
190
+ "learning_rate": 0.0006495,
191
+ "loss": 3.9928851318359375,
192
+ "step": 1300
193
+ },
194
+ {
195
+ "epoch": 0.011796477986734423,
196
+ "grad_norm": 0.2819700241088867,
197
+ "learning_rate": 0.0006745,
198
+ "loss": 3.9584146118164063,
199
+ "step": 1350
200
+ },
201
+ {
202
+ "epoch": 0.012233384578835698,
203
+ "grad_norm": 0.29095014929771423,
204
+ "learning_rate": 0.0006995,
205
+ "loss": 3.9291363525390626,
206
+ "step": 1400
207
+ },
208
+ {
209
+ "epoch": 0.012670291170936973,
210
+ "grad_norm": 0.2987957298755646,
211
+ "learning_rate": 0.0007245000000000001,
212
+ "loss": 3.9090225219726564,
213
+ "step": 1450
214
+ },
215
+ {
216
+ "epoch": 0.013107197763038248,
217
+ "grad_norm": 0.28419050574302673,
218
+ "learning_rate": 0.0007495000000000001,
219
+ "loss": 3.890744934082031,
220
+ "step": 1500
221
+ },
222
+ {
223
+ "epoch": 0.013544104355139523,
224
+ "grad_norm": 0.3039053976535797,
225
+ "learning_rate": 0.0007745,
226
+ "loss": 3.861073303222656,
227
+ "step": 1550
228
+ },
229
+ {
230
+ "epoch": 0.013981010947240798,
231
+ "grad_norm": 0.25676754117012024,
232
+ "learning_rate": 0.0007995,
233
+ "loss": 3.8428253173828124,
234
+ "step": 1600
235
+ },
236
+ {
237
+ "epoch": 0.014417917539342073,
238
+ "grad_norm": 0.2573896646499634,
239
+ "learning_rate": 0.0008245,
240
+ "loss": 3.8260653686523436,
241
+ "step": 1650
242
+ },
243
+ {
244
+ "epoch": 0.014854824131443348,
245
+ "grad_norm": 0.272127240896225,
246
+ "learning_rate": 0.0008495000000000001,
247
+ "loss": 3.8006341552734373,
248
+ "step": 1700
249
+ },
250
+ {
251
+ "epoch": 0.015291730723544623,
252
+ "grad_norm": 0.23205944895744324,
253
+ "learning_rate": 0.0008745000000000001,
254
+ "loss": 3.7862649536132813,
255
+ "step": 1750
256
+ },
257
+ {
258
+ "epoch": 0.0157286373156459,
259
+ "grad_norm": 0.22847425937652588,
260
+ "learning_rate": 0.0008995,
261
+ "loss": 3.771934814453125,
262
+ "step": 1800
263
+ },
264
+ {
265
+ "epoch": 0.016165543907747174,
266
+ "grad_norm": 0.2651786208152771,
267
+ "learning_rate": 0.0009245,
268
+ "loss": 3.76009521484375,
269
+ "step": 1850
270
+ },
271
+ {
272
+ "epoch": 0.01660245049984845,
273
+ "grad_norm": 0.24903950095176697,
274
+ "learning_rate": 0.0009495,
275
+ "loss": 3.7454238891601563,
276
+ "step": 1900
277
+ },
278
+ {
279
+ "epoch": 0.017039357091949724,
280
+ "grad_norm": 0.24524764716625214,
281
+ "learning_rate": 0.0009745000000000001,
282
+ "loss": 3.735928955078125,
283
+ "step": 1950
284
+ },
285
+ {
286
+ "epoch": 0.017476263684051,
287
+ "grad_norm": 0.22325558960437775,
288
+ "learning_rate": 0.0009995000000000002,
289
+ "loss": 3.7209417724609377,
290
+ "step": 2000
291
+ },
292
+ {
293
+ "epoch": 0.017913170276152274,
294
+ "grad_norm": 0.23860357701778412,
295
+ "learning_rate": 0.001,
296
+ "loss": 3.7122769165039062,
297
+ "step": 2050
298
+ },
299
+ {
300
+ "epoch": 0.01835007686825355,
301
+ "grad_norm": 0.2162218689918518,
302
+ "learning_rate": 0.001,
303
+ "loss": 3.690401611328125,
304
+ "step": 2100
305
+ },
306
+ {
307
+ "epoch": 0.018786983460354824,
308
+ "grad_norm": 0.2171812504529953,
309
+ "learning_rate": 0.001,
310
+ "loss": 3.6825027465820312,
311
+ "step": 2150
312
+ },
313
+ {
314
+ "epoch": 0.0192238900524561,
315
+ "grad_norm": 0.22677843272686005,
316
+ "learning_rate": 0.001,
317
+ "loss": 3.6679034423828125,
318
+ "step": 2200
319
+ },
320
+ {
321
+ "epoch": 0.019660796644557374,
322
+ "grad_norm": 0.23934350907802582,
323
+ "learning_rate": 0.001,
324
+ "loss": 3.6494000244140623,
325
+ "step": 2250
326
+ },
327
+ {
328
+ "epoch": 0.02009770323665865,
329
+ "grad_norm": 0.20819289982318878,
330
+ "learning_rate": 0.001,
331
+ "loss": 3.640907897949219,
332
+ "step": 2300
333
+ },
334
+ {
335
+ "epoch": 0.020534609828759924,
336
+ "grad_norm": 0.22890810668468475,
337
+ "learning_rate": 0.001,
338
+ "loss": 3.6309326171875,
339
+ "step": 2350
340
+ },
341
+ {
342
+ "epoch": 0.0209715164208612,
343
+ "grad_norm": 0.2085026353597641,
344
+ "learning_rate": 0.001,
345
+ "loss": 3.6261746215820314,
346
+ "step": 2400
347
+ },
348
+ {
349
+ "epoch": 0.021408423012962474,
350
+ "grad_norm": 0.22136099636554718,
351
+ "learning_rate": 0.001,
352
+ "loss": 3.614878845214844,
353
+ "step": 2450
354
+ },
355
+ {
356
+ "epoch": 0.021845329605063746,
357
+ "grad_norm": 0.21210630238056183,
358
+ "learning_rate": 0.001,
359
+ "loss": 3.599817810058594,
360
+ "step": 2500
361
+ },
362
+ {
363
+ "epoch": 0.02228223619716502,
364
+ "grad_norm": 0.24638274312019348,
365
+ "learning_rate": 0.001,
366
+ "loss": 3.591536865234375,
367
+ "step": 2550
368
+ },
369
+ {
370
+ "epoch": 0.022719142789266296,
371
+ "grad_norm": 0.20501184463500977,
372
+ "learning_rate": 0.001,
373
+ "loss": 3.58053955078125,
374
+ "step": 2600
375
+ },
376
+ {
377
+ "epoch": 0.02315604938136757,
378
+ "grad_norm": 0.2101815640926361,
379
+ "learning_rate": 0.001,
380
+ "loss": 3.5824459838867186,
381
+ "step": 2650
382
+ },
383
+ {
384
+ "epoch": 0.023592955973468846,
385
+ "grad_norm": 0.22906431555747986,
386
+ "learning_rate": 0.001,
387
+ "loss": 3.564821472167969,
388
+ "step": 2700
389
+ },
390
+ {
391
+ "epoch": 0.02402986256557012,
392
+ "grad_norm": 0.20708167552947998,
393
+ "learning_rate": 0.001,
394
+ "loss": 3.5566305541992187,
395
+ "step": 2750
396
+ },
397
+ {
398
+ "epoch": 0.024466769157671396,
399
+ "grad_norm": 0.21956811845302582,
400
+ "learning_rate": 0.001,
401
+ "loss": 3.5507748413085936,
402
+ "step": 2800
403
+ },
404
+ {
405
+ "epoch": 0.02490367574977267,
406
+ "grad_norm": 0.24124488234519958,
407
+ "learning_rate": 0.001,
408
+ "loss": 3.5426095581054686,
409
+ "step": 2850
410
+ },
411
+ {
412
+ "epoch": 0.025340582341873946,
413
+ "grad_norm": 0.20921742916107178,
414
+ "learning_rate": 0.001,
415
+ "loss": 3.531602783203125,
416
+ "step": 2900
417
+ },
418
+ {
419
+ "epoch": 0.02577748893397522,
420
+ "grad_norm": 0.19601280987262726,
421
+ "learning_rate": 0.001,
422
+ "loss": 3.5225894165039064,
423
+ "step": 2950
424
+ },
425
+ {
426
+ "epoch": 0.026214395526076496,
427
+ "grad_norm": 0.20243392884731293,
428
+ "learning_rate": 0.001,
429
+ "loss": 3.5183840942382814,
430
+ "step": 3000
431
+ },
432
+ {
433
+ "epoch": 0.026214395526076496,
434
+ "eval_loss": 3.518667221069336,
435
+ "eval_runtime": 5.8945,
436
+ "eval_samples_per_second": 165.579,
437
+ "eval_steps_per_second": 10.349,
438
+ "step": 3000
439
+ },
440
+ {
441
+ "epoch": 0.02665130211817777,
442
+ "grad_norm": 0.1983480155467987,
443
+ "learning_rate": 0.001,
444
+ "loss": 3.5175689697265624,
445
+ "step": 3050
446
+ },
447
+ {
448
+ "epoch": 0.027088208710279046,
449
+ "grad_norm": 0.20962044596672058,
450
+ "learning_rate": 0.001,
451
+ "loss": 3.501134033203125,
452
+ "step": 3100
453
+ },
454
+ {
455
+ "epoch": 0.02752511530238032,
456
+ "grad_norm": 0.20161673426628113,
457
+ "learning_rate": 0.001,
458
+ "loss": 3.508039855957031,
459
+ "step": 3150
460
+ },
461
+ {
462
+ "epoch": 0.027962021894481597,
463
+ "grad_norm": 0.21198803186416626,
464
+ "learning_rate": 0.001,
465
+ "loss": 3.493387451171875,
466
+ "step": 3200
467
+ },
468
+ {
469
+ "epoch": 0.02839892848658287,
470
+ "grad_norm": 0.21940743923187256,
471
+ "learning_rate": 0.001,
472
+ "loss": 3.4941644287109375,
473
+ "step": 3250
474
+ },
475
+ {
476
+ "epoch": 0.028835835078684147,
477
+ "grad_norm": 0.23343515396118164,
478
+ "learning_rate": 0.001,
479
+ "loss": 3.4835015869140626,
480
+ "step": 3300
481
+ },
482
+ {
483
+ "epoch": 0.02927274167078542,
484
+ "grad_norm": 0.19761498272418976,
485
+ "learning_rate": 0.001,
486
+ "loss": 3.482951965332031,
487
+ "step": 3350
488
+ },
489
+ {
490
+ "epoch": 0.029709648262886697,
491
+ "grad_norm": 0.22693106532096863,
492
+ "learning_rate": 0.001,
493
+ "loss": 3.475736389160156,
494
+ "step": 3400
495
+ },
496
+ {
497
+ "epoch": 0.030146554854987972,
498
+ "grad_norm": 0.19969327747821808,
499
+ "learning_rate": 0.001,
500
+ "loss": 3.4689141845703126,
501
+ "step": 3450
502
+ },
503
+ {
504
+ "epoch": 0.030583461447089247,
505
+ "grad_norm": 0.23987655341625214,
506
+ "learning_rate": 0.001,
507
+ "loss": 3.464185791015625,
508
+ "step": 3500
509
+ },
510
+ {
511
+ "epoch": 0.031020368039190522,
512
+ "grad_norm": 0.20894227921962738,
513
+ "learning_rate": 0.001,
514
+ "loss": 3.4571575927734375,
515
+ "step": 3550
516
+ },
517
+ {
518
+ "epoch": 0.0314572746312918,
519
+ "grad_norm": 0.22110916674137115,
520
+ "learning_rate": 0.001,
521
+ "loss": 3.4559121704101563,
522
+ "step": 3600
523
+ },
524
+ {
525
+ "epoch": 0.03189418122339307,
526
+ "grad_norm": 0.21295017004013062,
527
+ "learning_rate": 0.001,
528
+ "loss": 3.451357421875,
529
+ "step": 3650
530
+ },
531
+ {
532
+ "epoch": 0.03233108781549435,
533
+ "grad_norm": 0.19029361009597778,
534
+ "learning_rate": 0.001,
535
+ "loss": 3.4517440795898438,
536
+ "step": 3700
537
+ },
538
+ {
539
+ "epoch": 0.03276799440759562,
540
+ "grad_norm": 0.2307061105966568,
541
+ "learning_rate": 0.001,
542
+ "loss": 3.4454116821289062,
543
+ "step": 3750
544
+ },
545
+ {
546
+ "epoch": 0.0332049009996969,
547
+ "grad_norm": 0.2172439694404602,
548
+ "learning_rate": 0.001,
549
+ "loss": 3.4348458862304687,
550
+ "step": 3800
551
+ },
552
+ {
553
+ "epoch": 0.03364180759179817,
554
+ "grad_norm": 0.19329442083835602,
555
+ "learning_rate": 0.001,
556
+ "loss": 3.437733154296875,
557
+ "step": 3850
558
+ },
559
+ {
560
+ "epoch": 0.03407871418389945,
561
+ "grad_norm": 0.2241591513156891,
562
+ "learning_rate": 0.001,
563
+ "loss": 3.426455383300781,
564
+ "step": 3900
565
+ },
566
+ {
567
+ "epoch": 0.03451562077600072,
568
+ "grad_norm": 0.2319568693637848,
569
+ "learning_rate": 0.001,
570
+ "loss": 3.433238830566406,
571
+ "step": 3950
572
+ },
573
+ {
574
+ "epoch": 0.034952527368102,
575
+ "grad_norm": 0.1924877166748047,
576
+ "learning_rate": 0.001,
577
+ "loss": 3.4249606323242188,
578
+ "step": 4000
579
+ },
580
+ {
581
+ "epoch": 0.03538943396020327,
582
+ "grad_norm": 0.2120707631111145,
583
+ "learning_rate": 0.001,
584
+ "loss": 3.421075439453125,
585
+ "step": 4050
586
+ },
587
+ {
588
+ "epoch": 0.03582634055230455,
589
+ "grad_norm": 0.19918771088123322,
590
+ "learning_rate": 0.001,
591
+ "loss": 3.4203054809570315,
592
+ "step": 4100
593
+ },
594
+ {
595
+ "epoch": 0.03626324714440582,
596
+ "grad_norm": 0.2003307044506073,
597
+ "learning_rate": 0.001,
598
+ "loss": 3.411707763671875,
599
+ "step": 4150
600
+ },
601
+ {
602
+ "epoch": 0.0367001537365071,
603
+ "grad_norm": 0.2270653247833252,
604
+ "learning_rate": 0.001,
605
+ "loss": 3.4078097534179688,
606
+ "step": 4200
607
+ },
608
+ {
609
+ "epoch": 0.03713706032860837,
610
+ "grad_norm": 0.21652576327323914,
611
+ "learning_rate": 0.001,
612
+ "loss": 3.4062127685546875,
613
+ "step": 4250
614
+ },
615
+ {
616
+ "epoch": 0.03757396692070965,
617
+ "grad_norm": 0.23578432202339172,
618
+ "learning_rate": 0.001,
619
+ "loss": 3.400606689453125,
620
+ "step": 4300
621
+ },
622
+ {
623
+ "epoch": 0.03801087351281092,
624
+ "grad_norm": 0.20475627481937408,
625
+ "learning_rate": 0.001,
626
+ "loss": 3.395752258300781,
627
+ "step": 4350
628
+ },
629
+ {
630
+ "epoch": 0.0384477801049122,
631
+ "grad_norm": 0.2542702555656433,
632
+ "learning_rate": 0.001,
633
+ "loss": 3.396254577636719,
634
+ "step": 4400
635
+ },
636
+ {
637
+ "epoch": 0.03888468669701347,
638
+ "grad_norm": 0.18298493325710297,
639
+ "learning_rate": 0.001,
640
+ "loss": 3.393648986816406,
641
+ "step": 4450
642
+ },
643
+ {
644
+ "epoch": 0.03932159328911475,
645
+ "grad_norm": 0.22572273015975952,
646
+ "learning_rate": 0.001,
647
+ "loss": 3.3865216064453123,
648
+ "step": 4500
649
+ },
650
+ {
651
+ "epoch": 0.03975849988121602,
652
+ "grad_norm": 0.19484151899814606,
653
+ "learning_rate": 0.001,
654
+ "loss": 3.3849478149414063,
655
+ "step": 4550
656
+ },
657
+ {
658
+ "epoch": 0.0401954064733173,
659
+ "grad_norm": 0.19386635720729828,
660
+ "learning_rate": 0.001,
661
+ "loss": 3.38037109375,
662
+ "step": 4600
663
+ },
664
+ {
665
+ "epoch": 0.04063231306541857,
666
+ "grad_norm": 0.20124493539333344,
667
+ "learning_rate": 0.001,
668
+ "loss": 3.382728271484375,
669
+ "step": 4650
670
+ },
671
+ {
672
+ "epoch": 0.04106921965751985,
673
+ "grad_norm": 0.19963102042675018,
674
+ "learning_rate": 0.001,
675
+ "loss": 3.381324462890625,
676
+ "step": 4700
677
+ },
678
+ {
679
+ "epoch": 0.04150612624962112,
680
+ "grad_norm": 0.20611299574375153,
681
+ "learning_rate": 0.001,
682
+ "loss": 3.3788223266601562,
683
+ "step": 4750
684
+ },
685
+ {
686
+ "epoch": 0.0419430328417224,
687
+ "grad_norm": 0.19405238330364227,
688
+ "learning_rate": 0.001,
689
+ "loss": 3.3741259765625,
690
+ "step": 4800
691
+ },
692
+ {
693
+ "epoch": 0.04237993943382367,
694
+ "grad_norm": 0.20914608240127563,
695
+ "learning_rate": 0.001,
696
+ "loss": 3.370547180175781,
697
+ "step": 4850
698
+ },
699
+ {
700
+ "epoch": 0.04281684602592495,
701
+ "grad_norm": 0.19159191846847534,
702
+ "learning_rate": 0.001,
703
+ "loss": 3.3625115966796875,
704
+ "step": 4900
705
+ },
706
+ {
707
+ "epoch": 0.04325375261802622,
708
+ "grad_norm": 0.19163894653320312,
709
+ "learning_rate": 0.001,
710
+ "loss": 3.360504455566406,
711
+ "step": 4950
712
+ },
713
+ {
714
+ "epoch": 0.04369065921012749,
715
+ "grad_norm": 0.19763918220996857,
716
+ "learning_rate": 0.001,
717
+ "loss": 3.3601028442382814,
718
+ "step": 5000
719
+ },
720
+ {
721
+ "epoch": 0.04412756580222877,
722
+ "grad_norm": 0.18974512815475464,
723
+ "learning_rate": 0.001,
724
+ "loss": 3.3579150390625,
725
+ "step": 5050
726
+ },
727
+ {
728
+ "epoch": 0.04456447239433004,
729
+ "grad_norm": 0.21315094828605652,
730
+ "learning_rate": 0.001,
731
+ "loss": 3.355892333984375,
732
+ "step": 5100
733
+ },
734
+ {
735
+ "epoch": 0.04500137898643132,
736
+ "grad_norm": 0.1904531568288803,
737
+ "learning_rate": 0.001,
738
+ "loss": 3.360745849609375,
739
+ "step": 5150
740
+ },
741
+ {
742
+ "epoch": 0.04543828557853259,
743
+ "grad_norm": 0.19907866418361664,
744
+ "learning_rate": 0.001,
745
+ "loss": 3.357358703613281,
746
+ "step": 5200
747
+ },
748
+ {
749
+ "epoch": 0.04587519217063387,
750
+ "grad_norm": 0.22703590989112854,
751
+ "learning_rate": 0.001,
752
+ "loss": 3.356294860839844,
753
+ "step": 5250
754
+ },
755
+ {
756
+ "epoch": 0.04631209876273514,
757
+ "grad_norm": 0.1892520785331726,
758
+ "learning_rate": 0.001,
759
+ "loss": 3.3490374755859373,
760
+ "step": 5300
761
+ },
762
+ {
763
+ "epoch": 0.04674900535483642,
764
+ "grad_norm": 0.18953296542167664,
765
+ "learning_rate": 0.001,
766
+ "loss": 3.3442022705078127,
767
+ "step": 5350
768
+ },
769
+ {
770
+ "epoch": 0.04718591194693769,
771
+ "grad_norm": 0.1933782547712326,
772
+ "learning_rate": 0.001,
773
+ "loss": 3.3514590454101563,
774
+ "step": 5400
775
+ },
776
+ {
777
+ "epoch": 0.04762281853903897,
778
+ "grad_norm": 0.22573590278625488,
779
+ "learning_rate": 0.001,
780
+ "loss": 3.3468975830078125,
781
+ "step": 5450
782
+ },
783
+ {
784
+ "epoch": 0.04805972513114024,
785
+ "grad_norm": 0.2123604267835617,
786
+ "learning_rate": 0.001,
787
+ "loss": 3.3379788208007812,
788
+ "step": 5500
789
+ },
790
+ {
791
+ "epoch": 0.04849663172324152,
792
+ "grad_norm": 0.18217475712299347,
793
+ "learning_rate": 0.001,
794
+ "loss": 3.344074401855469,
795
+ "step": 5550
796
+ },
797
+ {
798
+ "epoch": 0.04893353831534279,
799
+ "grad_norm": 0.2128428965806961,
800
+ "learning_rate": 0.001,
801
+ "loss": 3.3349212646484374,
802
+ "step": 5600
803
+ },
804
+ {
805
+ "epoch": 0.04937044490744407,
806
+ "grad_norm": 0.19785532355308533,
807
+ "learning_rate": 0.001,
808
+ "loss": 3.327950744628906,
809
+ "step": 5650
810
+ },
811
+ {
812
+ "epoch": 0.04980735149954534,
813
+ "grad_norm": 0.18241338431835175,
814
+ "learning_rate": 0.001,
815
+ "loss": 3.334012451171875,
816
+ "step": 5700
817
+ },
818
+ {
819
+ "epoch": 0.05024425809164662,
820
+ "grad_norm": 0.2020372748374939,
821
+ "learning_rate": 0.001,
822
+ "loss": 3.3329043579101563,
823
+ "step": 5750
824
+ },
825
+ {
826
+ "epoch": 0.05068116468374789,
827
+ "grad_norm": 0.2071504294872284,
828
+ "learning_rate": 0.001,
829
+ "loss": 3.33412353515625,
830
+ "step": 5800
831
+ },
832
+ {
833
+ "epoch": 0.05111807127584917,
834
+ "grad_norm": 0.20365303754806519,
835
+ "learning_rate": 0.001,
836
+ "loss": 3.322493591308594,
837
+ "step": 5850
838
+ },
839
+ {
840
+ "epoch": 0.05155497786795044,
841
+ "grad_norm": 0.21742033958435059,
842
+ "learning_rate": 0.001,
843
+ "loss": 3.329978332519531,
844
+ "step": 5900
845
+ },
846
+ {
847
+ "epoch": 0.05199188446005172,
848
+ "grad_norm": 0.18907567858695984,
849
+ "learning_rate": 0.001,
850
+ "loss": 3.328219299316406,
851
+ "step": 5950
852
+ },
853
+ {
854
+ "epoch": 0.05242879105215299,
855
+ "grad_norm": 0.18905521929264069,
856
+ "learning_rate": 0.001,
857
+ "loss": 3.323075866699219,
858
+ "step": 6000
859
+ },
860
+ {
861
+ "epoch": 0.05242879105215299,
862
+ "eval_loss": 3.3261899948120117,
863
+ "eval_runtime": 6.1096,
864
+ "eval_samples_per_second": 159.748,
865
+ "eval_steps_per_second": 9.984,
866
+ "step": 6000
867
+ }
868
+ ],
869
+ "logging_steps": 50,
870
+ "max_steps": 114440,
871
+ "num_input_tokens_seen": 0,
872
+ "num_train_epochs": 1,
873
+ "save_steps": 3000,
874
+ "stateful_callbacks": {
875
+ "TrainerControl": {
876
+ "args": {
877
+ "should_epoch_stop": false,
878
+ "should_evaluate": false,
879
+ "should_log": false,
880
+ "should_save": true,
881
+ "should_training_stop": false
882
+ },
883
+ "attributes": {}
884
+ }
885
+ },
886
+ "total_flos": 7.12679224246272e+17,
887
+ "train_batch_size": 16,
888
+ "trial_name": null,
889
+ "trial_params": null
890
+ }
checkpoint-6000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b89574d4fb532cbff16922a1304201180637df9c21820d655203050012091374
3
+ size 5329