sedrickkeh commited on
Commit
9b1ebdd
·
verified ·
1 Parent(s): a6a7ec8

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: top_14_ranking_stackexchange
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # top_14_ranking_stackexchange
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.8005
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: top_14_ranking_stackexchange
 
16
 
17
  # top_14_ranking_stackexchange
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/top_14_ranking_stackexchange dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.8005
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.997084548104956,
3
+ "eval_loss": 0.8005240559577942,
4
+ "eval_runtime": 555.4162,
5
+ "eval_samples_per_second": 24.96,
6
+ "eval_steps_per_second": 0.391,
7
+ "total_flos": 2582698052812800.0,
8
+ "train_loss": 0.7834810720498151,
9
+ "train_runtime": 91136.0835,
10
+ "train_samples_per_second": 8.67,
11
+ "train_steps_per_second": 0.017
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.997084548104956,
3
+ "eval_loss": 0.8005240559577942,
4
+ "eval_runtime": 555.4162,
5
+ "eval_samples_per_second": 24.96,
6
+ "eval_steps_per_second": 0.391
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.997084548104956,
3
+ "total_flos": 2582698052812800.0,
4
+ "train_loss": 0.7834810720498151,
5
+ "train_runtime": 91136.0835,
6
+ "train_samples_per_second": 8.67,
7
+ "train_steps_per_second": 0.017
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.997084548104956,
5
+ "eval_steps": 500,
6
+ "global_step": 1542,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.019436345966958212,
13
+ "grad_norm": 3.5477320746078096,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.0532,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.038872691933916424,
20
+ "grad_norm": 2.8209235996953512,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.9684,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.05830903790087463,
27
+ "grad_norm": 1.1257852692164845,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.9294,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.07774538386783285,
34
+ "grad_norm": 0.888481539231949,
35
+ "learning_rate": 5e-06,
36
+ "loss": 0.9201,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.09718172983479106,
41
+ "grad_norm": 0.6469537614046023,
42
+ "learning_rate": 5e-06,
43
+ "loss": 0.8971,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.11661807580174927,
48
+ "grad_norm": 0.592544408615531,
49
+ "learning_rate": 5e-06,
50
+ "loss": 0.8837,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.1360544217687075,
55
+ "grad_norm": 0.6433313667438364,
56
+ "learning_rate": 5e-06,
57
+ "loss": 0.8749,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.1554907677356657,
62
+ "grad_norm": 1.344429915972531,
63
+ "learning_rate": 5e-06,
64
+ "loss": 0.867,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.1749271137026239,
69
+ "grad_norm": 0.8707935340603853,
70
+ "learning_rate": 5e-06,
71
+ "loss": 0.8623,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.19436345966958213,
76
+ "grad_norm": 0.7828161719097383,
77
+ "learning_rate": 5e-06,
78
+ "loss": 0.8624,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.21379980563654033,
83
+ "grad_norm": 0.682126021781355,
84
+ "learning_rate": 5e-06,
85
+ "loss": 0.8611,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.23323615160349853,
90
+ "grad_norm": 0.8081328723422389,
91
+ "learning_rate": 5e-06,
92
+ "loss": 0.8551,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.25267249757045673,
97
+ "grad_norm": 0.6691994077909735,
98
+ "learning_rate": 5e-06,
99
+ "loss": 0.8583,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.272108843537415,
104
+ "grad_norm": 0.586140300081327,
105
+ "learning_rate": 5e-06,
106
+ "loss": 0.8523,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2915451895043732,
111
+ "grad_norm": 0.8552001050583659,
112
+ "learning_rate": 5e-06,
113
+ "loss": 0.8497,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.3109815354713314,
118
+ "grad_norm": 0.501523965918074,
119
+ "learning_rate": 5e-06,
120
+ "loss": 0.8464,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3304178814382896,
125
+ "grad_norm": 0.6785445153206255,
126
+ "learning_rate": 5e-06,
127
+ "loss": 0.848,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.3498542274052478,
132
+ "grad_norm": 0.6494060293856031,
133
+ "learning_rate": 5e-06,
134
+ "loss": 0.8459,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.369290573372206,
139
+ "grad_norm": 0.6020737926390343,
140
+ "learning_rate": 5e-06,
141
+ "loss": 0.8484,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.38872691933916426,
146
+ "grad_norm": 0.48912875825316915,
147
+ "learning_rate": 5e-06,
148
+ "loss": 0.8449,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.40816326530612246,
153
+ "grad_norm": 0.8781137047011839,
154
+ "learning_rate": 5e-06,
155
+ "loss": 0.8395,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.42759961127308066,
160
+ "grad_norm": 0.5879468826056136,
161
+ "learning_rate": 5e-06,
162
+ "loss": 0.8387,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.44703595724003886,
167
+ "grad_norm": 0.6017675792916065,
168
+ "learning_rate": 5e-06,
169
+ "loss": 0.8363,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.46647230320699706,
174
+ "grad_norm": 0.616647981494789,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.8336,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.4859086491739553,
181
+ "grad_norm": 0.855268617382177,
182
+ "learning_rate": 5e-06,
183
+ "loss": 0.8336,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.5053449951409135,
188
+ "grad_norm": 0.780188445457583,
189
+ "learning_rate": 5e-06,
190
+ "loss": 0.831,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.5247813411078717,
195
+ "grad_norm": 0.5840816004625115,
196
+ "learning_rate": 5e-06,
197
+ "loss": 0.8319,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.54421768707483,
202
+ "grad_norm": 0.6535257947752856,
203
+ "learning_rate": 5e-06,
204
+ "loss": 0.833,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5636540330417882,
209
+ "grad_norm": 0.6710102563759031,
210
+ "learning_rate": 5e-06,
211
+ "loss": 0.8354,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.5830903790087464,
216
+ "grad_norm": 0.5947252085113186,
217
+ "learning_rate": 5e-06,
218
+ "loss": 0.8315,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.6025267249757046,
223
+ "grad_norm": 0.5902161171422673,
224
+ "learning_rate": 5e-06,
225
+ "loss": 0.83,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.6219630709426628,
230
+ "grad_norm": 0.5757165816873938,
231
+ "learning_rate": 5e-06,
232
+ "loss": 0.8299,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.641399416909621,
237
+ "grad_norm": 0.6751753868245474,
238
+ "learning_rate": 5e-06,
239
+ "loss": 0.827,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6608357628765792,
244
+ "grad_norm": 0.535560817394151,
245
+ "learning_rate": 5e-06,
246
+ "loss": 0.8264,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.6802721088435374,
251
+ "grad_norm": 0.6183377886108462,
252
+ "learning_rate": 5e-06,
253
+ "loss": 0.8243,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.6997084548104956,
258
+ "grad_norm": 0.6200501825686097,
259
+ "learning_rate": 5e-06,
260
+ "loss": 0.828,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.7191448007774538,
265
+ "grad_norm": 0.5622254912052161,
266
+ "learning_rate": 5e-06,
267
+ "loss": 0.8287,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.738581146744412,
272
+ "grad_norm": 0.6720527159909909,
273
+ "learning_rate": 5e-06,
274
+ "loss": 0.8234,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.7580174927113703,
279
+ "grad_norm": 0.5315560166276624,
280
+ "learning_rate": 5e-06,
281
+ "loss": 0.824,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.7774538386783285,
286
+ "grad_norm": 0.6413527148042328,
287
+ "learning_rate": 5e-06,
288
+ "loss": 0.8194,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.7968901846452867,
293
+ "grad_norm": 0.6402327795437167,
294
+ "learning_rate": 5e-06,
295
+ "loss": 0.8243,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.8163265306122449,
300
+ "grad_norm": 0.6290935177044384,
301
+ "learning_rate": 5e-06,
302
+ "loss": 0.8201,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.8357628765792031,
307
+ "grad_norm": 0.6137598310285064,
308
+ "learning_rate": 5e-06,
309
+ "loss": 0.8187,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.8551992225461613,
314
+ "grad_norm": 0.6137894354862566,
315
+ "learning_rate": 5e-06,
316
+ "loss": 0.8217,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.8746355685131195,
321
+ "grad_norm": 0.7376542092364302,
322
+ "learning_rate": 5e-06,
323
+ "loss": 0.8191,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.8940719144800777,
328
+ "grad_norm": 0.5443940007812901,
329
+ "learning_rate": 5e-06,
330
+ "loss": 0.8179,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.9135082604470359,
335
+ "grad_norm": 0.6778023088897194,
336
+ "learning_rate": 5e-06,
337
+ "loss": 0.8158,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.9329446064139941,
342
+ "grad_norm": 0.6040677193471313,
343
+ "learning_rate": 5e-06,
344
+ "loss": 0.8176,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.9523809523809523,
349
+ "grad_norm": 0.5741967517130403,
350
+ "learning_rate": 5e-06,
351
+ "loss": 0.8215,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.9718172983479106,
356
+ "grad_norm": 0.5184977635424605,
357
+ "learning_rate": 5e-06,
358
+ "loss": 0.8218,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.9912536443148688,
363
+ "grad_norm": 0.5484601778792662,
364
+ "learning_rate": 5e-06,
365
+ "loss": 0.8158,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.9990281827016521,
370
+ "eval_loss": 0.8152287006378174,
371
+ "eval_runtime": 547.9504,
372
+ "eval_samples_per_second": 25.3,
373
+ "eval_steps_per_second": 0.396,
374
+ "step": 514
375
+ },
376
+ {
377
+ "epoch": 1.010689990281827,
378
+ "grad_norm": 0.6821565020818955,
379
+ "learning_rate": 5e-06,
380
+ "loss": 0.8355,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 1.0301263362487851,
385
+ "grad_norm": 0.6277981964271326,
386
+ "learning_rate": 5e-06,
387
+ "loss": 0.7749,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 1.0495626822157433,
392
+ "grad_norm": 0.6114756472079457,
393
+ "learning_rate": 5e-06,
394
+ "loss": 0.7751,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 1.0689990281827018,
399
+ "grad_norm": 0.8259615966427586,
400
+ "learning_rate": 5e-06,
401
+ "loss": 0.7756,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 1.08843537414966,
406
+ "grad_norm": 0.6113352198439804,
407
+ "learning_rate": 5e-06,
408
+ "loss": 0.7789,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 1.1078717201166182,
413
+ "grad_norm": 0.5269512429419262,
414
+ "learning_rate": 5e-06,
415
+ "loss": 0.7784,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 1.1273080660835764,
420
+ "grad_norm": 0.5792710933468033,
421
+ "learning_rate": 5e-06,
422
+ "loss": 0.7757,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 1.1467444120505346,
427
+ "grad_norm": 0.5467198449481481,
428
+ "learning_rate": 5e-06,
429
+ "loss": 0.7757,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 1.1661807580174928,
434
+ "grad_norm": 0.6190447420364188,
435
+ "learning_rate": 5e-06,
436
+ "loss": 0.7754,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 1.185617103984451,
441
+ "grad_norm": 0.7074708178383962,
442
+ "learning_rate": 5e-06,
443
+ "loss": 0.7738,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 1.2050534499514092,
448
+ "grad_norm": 0.5708793884696434,
449
+ "learning_rate": 5e-06,
450
+ "loss": 0.7675,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 1.2244897959183674,
455
+ "grad_norm": 0.5278424041049065,
456
+ "learning_rate": 5e-06,
457
+ "loss": 0.7724,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 1.2439261418853256,
462
+ "grad_norm": 0.48269223376284837,
463
+ "learning_rate": 5e-06,
464
+ "loss": 0.7678,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 1.2633624878522838,
469
+ "grad_norm": 0.6628438951702088,
470
+ "learning_rate": 5e-06,
471
+ "loss": 0.772,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 1.282798833819242,
476
+ "grad_norm": 0.49956285734450795,
477
+ "learning_rate": 5e-06,
478
+ "loss": 0.78,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 1.3022351797862002,
483
+ "grad_norm": 0.4482989241465936,
484
+ "learning_rate": 5e-06,
485
+ "loss": 0.7712,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 1.3216715257531584,
490
+ "grad_norm": 0.515252608881534,
491
+ "learning_rate": 5e-06,
492
+ "loss": 0.7712,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 1.3411078717201166,
497
+ "grad_norm": 0.5392787594093453,
498
+ "learning_rate": 5e-06,
499
+ "loss": 0.7733,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 1.3605442176870748,
504
+ "grad_norm": 0.5239288581769422,
505
+ "learning_rate": 5e-06,
506
+ "loss": 0.7744,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 1.379980563654033,
511
+ "grad_norm": 0.5368087860350439,
512
+ "learning_rate": 5e-06,
513
+ "loss": 0.7721,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 1.3994169096209912,
518
+ "grad_norm": 0.5331498843832938,
519
+ "learning_rate": 5e-06,
520
+ "loss": 0.7702,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 1.4188532555879494,
525
+ "grad_norm": 0.5840718481917428,
526
+ "learning_rate": 5e-06,
527
+ "loss": 0.7755,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 1.4382896015549078,
532
+ "grad_norm": 0.537679455083028,
533
+ "learning_rate": 5e-06,
534
+ "loss": 0.7719,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 1.4577259475218658,
539
+ "grad_norm": 0.7948828701245976,
540
+ "learning_rate": 5e-06,
541
+ "loss": 0.7717,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 1.4771622934888242,
546
+ "grad_norm": 0.5813227807421696,
547
+ "learning_rate": 5e-06,
548
+ "loss": 0.7763,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 1.4965986394557822,
553
+ "grad_norm": 0.6049608273143411,
554
+ "learning_rate": 5e-06,
555
+ "loss": 0.7719,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 1.5160349854227406,
560
+ "grad_norm": 0.6057676712274179,
561
+ "learning_rate": 5e-06,
562
+ "loss": 0.7742,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 1.5354713313896986,
567
+ "grad_norm": 0.625042201984692,
568
+ "learning_rate": 5e-06,
569
+ "loss": 0.7767,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 1.554907677356657,
574
+ "grad_norm": 0.5502470811006085,
575
+ "learning_rate": 5e-06,
576
+ "loss": 0.769,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 1.574344023323615,
581
+ "grad_norm": 0.5857965121292225,
582
+ "learning_rate": 5e-06,
583
+ "loss": 0.7709,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 1.5937803692905734,
588
+ "grad_norm": 0.613513782295781,
589
+ "learning_rate": 5e-06,
590
+ "loss": 0.7755,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 1.6132167152575316,
595
+ "grad_norm": 0.583240417199926,
596
+ "learning_rate": 5e-06,
597
+ "loss": 0.7738,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 1.6326530612244898,
602
+ "grad_norm": 0.5231701163074167,
603
+ "learning_rate": 5e-06,
604
+ "loss": 0.7731,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 1.652089407191448,
609
+ "grad_norm": 0.4914609295920534,
610
+ "learning_rate": 5e-06,
611
+ "loss": 0.7714,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 1.6715257531584062,
616
+ "grad_norm": 0.5250445862469549,
617
+ "learning_rate": 5e-06,
618
+ "loss": 0.7747,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 1.6909620991253644,
623
+ "grad_norm": 0.5415316148110199,
624
+ "learning_rate": 5e-06,
625
+ "loss": 0.7685,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 1.7103984450923226,
630
+ "grad_norm": 0.5426871063126633,
631
+ "learning_rate": 5e-06,
632
+ "loss": 0.7687,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 1.7298347910592808,
637
+ "grad_norm": 0.5573890854917875,
638
+ "learning_rate": 5e-06,
639
+ "loss": 0.7732,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 1.749271137026239,
644
+ "grad_norm": 0.5966240890058521,
645
+ "learning_rate": 5e-06,
646
+ "loss": 0.7695,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 1.7687074829931972,
651
+ "grad_norm": 0.48852508189672406,
652
+ "learning_rate": 5e-06,
653
+ "loss": 0.7716,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 1.7881438289601554,
658
+ "grad_norm": 0.510423049422432,
659
+ "learning_rate": 5e-06,
660
+ "loss": 0.7695,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 1.8075801749271136,
665
+ "grad_norm": 0.5128362713697912,
666
+ "learning_rate": 5e-06,
667
+ "loss": 0.768,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 1.8270165208940718,
672
+ "grad_norm": 0.5100892708497722,
673
+ "learning_rate": 5e-06,
674
+ "loss": 0.7721,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 1.8464528668610303,
679
+ "grad_norm": 0.5050841552954286,
680
+ "learning_rate": 5e-06,
681
+ "loss": 0.7703,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 1.8658892128279883,
686
+ "grad_norm": 0.5100122636839687,
687
+ "learning_rate": 5e-06,
688
+ "loss": 0.7675,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 1.8853255587949467,
693
+ "grad_norm": 0.5858103448907649,
694
+ "learning_rate": 5e-06,
695
+ "loss": 0.7719,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 1.9047619047619047,
700
+ "grad_norm": 0.7186164442084766,
701
+ "learning_rate": 5e-06,
702
+ "loss": 0.7704,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 1.924198250728863,
707
+ "grad_norm": 0.49187325414902705,
708
+ "learning_rate": 5e-06,
709
+ "loss": 0.7744,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 1.943634596695821,
714
+ "grad_norm": 0.48132179189009133,
715
+ "learning_rate": 5e-06,
716
+ "loss": 0.7666,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 1.9630709426627795,
721
+ "grad_norm": 0.6042950330363083,
722
+ "learning_rate": 5e-06,
723
+ "loss": 0.7682,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 1.9825072886297375,
728
+ "grad_norm": 0.5107563561120851,
729
+ "learning_rate": 5e-06,
730
+ "loss": 0.7688,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 2.0,
735
+ "eval_loss": 0.8017936944961548,
736
+ "eval_runtime": 549.3546,
737
+ "eval_samples_per_second": 25.235,
738
+ "eval_steps_per_second": 0.395,
739
+ "step": 1029
740
+ },
741
+ {
742
+ "epoch": 2.001943634596696,
743
+ "grad_norm": 0.8520236880105687,
744
+ "learning_rate": 5e-06,
745
+ "loss": 0.8065,
746
+ "step": 1030
747
+ },
748
+ {
749
+ "epoch": 2.021379980563654,
750
+ "grad_norm": 0.6759369335284802,
751
+ "learning_rate": 5e-06,
752
+ "loss": 0.7247,
753
+ "step": 1040
754
+ },
755
+ {
756
+ "epoch": 2.0408163265306123,
757
+ "grad_norm": 0.6810404313004184,
758
+ "learning_rate": 5e-06,
759
+ "loss": 0.7257,
760
+ "step": 1050
761
+ },
762
+ {
763
+ "epoch": 2.0602526724975703,
764
+ "grad_norm": 0.5976865962159595,
765
+ "learning_rate": 5e-06,
766
+ "loss": 0.7266,
767
+ "step": 1060
768
+ },
769
+ {
770
+ "epoch": 2.0796890184645287,
771
+ "grad_norm": 0.5170299538250477,
772
+ "learning_rate": 5e-06,
773
+ "loss": 0.7262,
774
+ "step": 1070
775
+ },
776
+ {
777
+ "epoch": 2.0991253644314867,
778
+ "grad_norm": 0.636746039057963,
779
+ "learning_rate": 5e-06,
780
+ "loss": 0.7271,
781
+ "step": 1080
782
+ },
783
+ {
784
+ "epoch": 2.118561710398445,
785
+ "grad_norm": 0.7347207445428233,
786
+ "learning_rate": 5e-06,
787
+ "loss": 0.7221,
788
+ "step": 1090
789
+ },
790
+ {
791
+ "epoch": 2.1379980563654035,
792
+ "grad_norm": 0.6669755018146,
793
+ "learning_rate": 5e-06,
794
+ "loss": 0.7277,
795
+ "step": 1100
796
+ },
797
+ {
798
+ "epoch": 2.1574344023323615,
799
+ "grad_norm": 0.5653238227925762,
800
+ "learning_rate": 5e-06,
801
+ "loss": 0.726,
802
+ "step": 1110
803
+ },
804
+ {
805
+ "epoch": 2.17687074829932,
806
+ "grad_norm": 0.5720425000083328,
807
+ "learning_rate": 5e-06,
808
+ "loss": 0.7271,
809
+ "step": 1120
810
+ },
811
+ {
812
+ "epoch": 2.196307094266278,
813
+ "grad_norm": 0.5494809428119856,
814
+ "learning_rate": 5e-06,
815
+ "loss": 0.727,
816
+ "step": 1130
817
+ },
818
+ {
819
+ "epoch": 2.2157434402332363,
820
+ "grad_norm": 0.6708852933316355,
821
+ "learning_rate": 5e-06,
822
+ "loss": 0.7286,
823
+ "step": 1140
824
+ },
825
+ {
826
+ "epoch": 2.2351797862001943,
827
+ "grad_norm": 0.5649904350477953,
828
+ "learning_rate": 5e-06,
829
+ "loss": 0.7253,
830
+ "step": 1150
831
+ },
832
+ {
833
+ "epoch": 2.2546161321671527,
834
+ "grad_norm": 0.6681152322447659,
835
+ "learning_rate": 5e-06,
836
+ "loss": 0.7227,
837
+ "step": 1160
838
+ },
839
+ {
840
+ "epoch": 2.2740524781341107,
841
+ "grad_norm": 0.7452957502891413,
842
+ "learning_rate": 5e-06,
843
+ "loss": 0.7271,
844
+ "step": 1170
845
+ },
846
+ {
847
+ "epoch": 2.293488824101069,
848
+ "grad_norm": 0.730891816162587,
849
+ "learning_rate": 5e-06,
850
+ "loss": 0.7297,
851
+ "step": 1180
852
+ },
853
+ {
854
+ "epoch": 2.312925170068027,
855
+ "grad_norm": 0.5422066002537126,
856
+ "learning_rate": 5e-06,
857
+ "loss": 0.7291,
858
+ "step": 1190
859
+ },
860
+ {
861
+ "epoch": 2.3323615160349855,
862
+ "grad_norm": 0.5368279161848004,
863
+ "learning_rate": 5e-06,
864
+ "loss": 0.7291,
865
+ "step": 1200
866
+ },
867
+ {
868
+ "epoch": 2.3517978620019435,
869
+ "grad_norm": 0.5831776325357405,
870
+ "learning_rate": 5e-06,
871
+ "loss": 0.7299,
872
+ "step": 1210
873
+ },
874
+ {
875
+ "epoch": 2.371234207968902,
876
+ "grad_norm": 0.4988880812457934,
877
+ "learning_rate": 5e-06,
878
+ "loss": 0.7286,
879
+ "step": 1220
880
+ },
881
+ {
882
+ "epoch": 2.39067055393586,
883
+ "grad_norm": 0.6745820300056689,
884
+ "learning_rate": 5e-06,
885
+ "loss": 0.7246,
886
+ "step": 1230
887
+ },
888
+ {
889
+ "epoch": 2.4101068999028183,
890
+ "grad_norm": 0.6502142374990822,
891
+ "learning_rate": 5e-06,
892
+ "loss": 0.7274,
893
+ "step": 1240
894
+ },
895
+ {
896
+ "epoch": 2.4295432458697763,
897
+ "grad_norm": 0.6686318012888572,
898
+ "learning_rate": 5e-06,
899
+ "loss": 0.7273,
900
+ "step": 1250
901
+ },
902
+ {
903
+ "epoch": 2.4489795918367347,
904
+ "grad_norm": 0.5959741098147713,
905
+ "learning_rate": 5e-06,
906
+ "loss": 0.7296,
907
+ "step": 1260
908
+ },
909
+ {
910
+ "epoch": 2.4684159378036927,
911
+ "grad_norm": 0.6933996442905096,
912
+ "learning_rate": 5e-06,
913
+ "loss": 0.7265,
914
+ "step": 1270
915
+ },
916
+ {
917
+ "epoch": 2.487852283770651,
918
+ "grad_norm": 0.5220203781381132,
919
+ "learning_rate": 5e-06,
920
+ "loss": 0.7279,
921
+ "step": 1280
922
+ },
923
+ {
924
+ "epoch": 2.5072886297376096,
925
+ "grad_norm": 0.5382425321528858,
926
+ "learning_rate": 5e-06,
927
+ "loss": 0.7305,
928
+ "step": 1290
929
+ },
930
+ {
931
+ "epoch": 2.5267249757045676,
932
+ "grad_norm": 0.5181218910911854,
933
+ "learning_rate": 5e-06,
934
+ "loss": 0.7249,
935
+ "step": 1300
936
+ },
937
+ {
938
+ "epoch": 2.5461613216715255,
939
+ "grad_norm": 0.6478067615615305,
940
+ "learning_rate": 5e-06,
941
+ "loss": 0.7319,
942
+ "step": 1310
943
+ },
944
+ {
945
+ "epoch": 2.565597667638484,
946
+ "grad_norm": 0.5078942293566884,
947
+ "learning_rate": 5e-06,
948
+ "loss": 0.7288,
949
+ "step": 1320
950
+ },
951
+ {
952
+ "epoch": 2.5850340136054424,
953
+ "grad_norm": 0.6268137880948265,
954
+ "learning_rate": 5e-06,
955
+ "loss": 0.7299,
956
+ "step": 1330
957
+ },
958
+ {
959
+ "epoch": 2.6044703595724004,
960
+ "grad_norm": 0.7996921164519973,
961
+ "learning_rate": 5e-06,
962
+ "loss": 0.7332,
963
+ "step": 1340
964
+ },
965
+ {
966
+ "epoch": 2.6239067055393583,
967
+ "grad_norm": 0.6835765316105765,
968
+ "learning_rate": 5e-06,
969
+ "loss": 0.7332,
970
+ "step": 1350
971
+ },
972
+ {
973
+ "epoch": 2.6433430515063168,
974
+ "grad_norm": 0.6208677881375628,
975
+ "learning_rate": 5e-06,
976
+ "loss": 0.7255,
977
+ "step": 1360
978
+ },
979
+ {
980
+ "epoch": 2.662779397473275,
981
+ "grad_norm": 0.5685215567462071,
982
+ "learning_rate": 5e-06,
983
+ "loss": 0.7311,
984
+ "step": 1370
985
+ },
986
+ {
987
+ "epoch": 2.682215743440233,
988
+ "grad_norm": 0.502556515949076,
989
+ "learning_rate": 5e-06,
990
+ "loss": 0.7288,
991
+ "step": 1380
992
+ },
993
+ {
994
+ "epoch": 2.7016520894071916,
995
+ "grad_norm": 0.5557319329244653,
996
+ "learning_rate": 5e-06,
997
+ "loss": 0.7252,
998
+ "step": 1390
999
+ },
1000
+ {
1001
+ "epoch": 2.7210884353741496,
1002
+ "grad_norm": 0.5681474343567127,
1003
+ "learning_rate": 5e-06,
1004
+ "loss": 0.7324,
1005
+ "step": 1400
1006
+ },
1007
+ {
1008
+ "epoch": 2.740524781341108,
1009
+ "grad_norm": 0.5718856742663174,
1010
+ "learning_rate": 5e-06,
1011
+ "loss": 0.7314,
1012
+ "step": 1410
1013
+ },
1014
+ {
1015
+ "epoch": 2.759961127308066,
1016
+ "grad_norm": 0.5207890225289823,
1017
+ "learning_rate": 5e-06,
1018
+ "loss": 0.7327,
1019
+ "step": 1420
1020
+ },
1021
+ {
1022
+ "epoch": 2.7793974732750244,
1023
+ "grad_norm": 0.5610775317663003,
1024
+ "learning_rate": 5e-06,
1025
+ "loss": 0.7325,
1026
+ "step": 1430
1027
+ },
1028
+ {
1029
+ "epoch": 2.7988338192419824,
1030
+ "grad_norm": 0.49435740403271333,
1031
+ "learning_rate": 5e-06,
1032
+ "loss": 0.7306,
1033
+ "step": 1440
1034
+ },
1035
+ {
1036
+ "epoch": 2.818270165208941,
1037
+ "grad_norm": 0.5548340643212172,
1038
+ "learning_rate": 5e-06,
1039
+ "loss": 0.7248,
1040
+ "step": 1450
1041
+ },
1042
+ {
1043
+ "epoch": 2.837706511175899,
1044
+ "grad_norm": 0.8162943266425523,
1045
+ "learning_rate": 5e-06,
1046
+ "loss": 0.728,
1047
+ "step": 1460
1048
+ },
1049
+ {
1050
+ "epoch": 2.857142857142857,
1051
+ "grad_norm": 0.5605722072050826,
1052
+ "learning_rate": 5e-06,
1053
+ "loss": 0.7321,
1054
+ "step": 1470
1055
+ },
1056
+ {
1057
+ "epoch": 2.8765792031098156,
1058
+ "grad_norm": 0.5811312094740239,
1059
+ "learning_rate": 5e-06,
1060
+ "loss": 0.7312,
1061
+ "step": 1480
1062
+ },
1063
+ {
1064
+ "epoch": 2.8960155490767736,
1065
+ "grad_norm": 0.581575144969956,
1066
+ "learning_rate": 5e-06,
1067
+ "loss": 0.7321,
1068
+ "step": 1490
1069
+ },
1070
+ {
1071
+ "epoch": 2.9154518950437316,
1072
+ "grad_norm": 0.5099095695816087,
1073
+ "learning_rate": 5e-06,
1074
+ "loss": 0.7334,
1075
+ "step": 1500
1076
+ },
1077
+ {
1078
+ "epoch": 2.93488824101069,
1079
+ "grad_norm": 0.5572721217938963,
1080
+ "learning_rate": 5e-06,
1081
+ "loss": 0.7308,
1082
+ "step": 1510
1083
+ },
1084
+ {
1085
+ "epoch": 2.9543245869776484,
1086
+ "grad_norm": 0.522930114991094,
1087
+ "learning_rate": 5e-06,
1088
+ "loss": 0.7295,
1089
+ "step": 1520
1090
+ },
1091
+ {
1092
+ "epoch": 2.9737609329446064,
1093
+ "grad_norm": 0.6436319992243297,
1094
+ "learning_rate": 5e-06,
1095
+ "loss": 0.7331,
1096
+ "step": 1530
1097
+ },
1098
+ {
1099
+ "epoch": 2.9931972789115644,
1100
+ "grad_norm": 0.6235339876579238,
1101
+ "learning_rate": 5e-06,
1102
+ "loss": 0.7263,
1103
+ "step": 1540
1104
+ },
1105
+ {
1106
+ "epoch": 2.997084548104956,
1107
+ "eval_loss": 0.8005240559577942,
1108
+ "eval_runtime": 550.1116,
1109
+ "eval_samples_per_second": 25.2,
1110
+ "eval_steps_per_second": 0.394,
1111
+ "step": 1542
1112
+ },
1113
+ {
1114
+ "epoch": 2.997084548104956,
1115
+ "step": 1542,
1116
+ "total_flos": 2582698052812800.0,
1117
+ "train_loss": 0.7834810720498151,
1118
+ "train_runtime": 91136.0835,
1119
+ "train_samples_per_second": 8.67,
1120
+ "train_steps_per_second": 0.017
1121
+ }
1122
+ ],
1123
+ "logging_steps": 10,
1124
+ "max_steps": 1542,
1125
+ "num_input_tokens_seen": 0,
1126
+ "num_train_epochs": 3,
1127
+ "save_steps": 500,
1128
+ "stateful_callbacks": {
1129
+ "TrainerControl": {
1130
+ "args": {
1131
+ "should_epoch_stop": false,
1132
+ "should_evaluate": false,
1133
+ "should_log": false,
1134
+ "should_save": true,
1135
+ "should_training_stop": true
1136
+ },
1137
+ "attributes": {}
1138
+ }
1139
+ },
1140
+ "total_flos": 2582698052812800.0,
1141
+ "train_batch_size": 8,
1142
+ "trial_name": null,
1143
+ "trial_params": null
1144
+ }
training_eval_loss.png ADDED
training_loss.png ADDED