Delete trainer_state.json
Browse files- trainer_state.json +0 -390
trainer_state.json
DELETED
@@ -1,390 +0,0 @@
|
|
1 |
-
{
|
2 |
-
"best_metric": null,
|
3 |
-
"best_model_checkpoint": null,
|
4 |
-
"epoch": 2.0,
|
5 |
-
"eval_steps": 10,
|
6 |
-
"global_step": 54,
|
7 |
-
"is_hyper_param_search": false,
|
8 |
-
"is_local_process_zero": true,
|
9 |
-
"is_world_process_zero": true,
|
10 |
-
"log_history": [
|
11 |
-
{
|
12 |
-
"epoch": 0.04,
|
13 |
-
"learning_rate": 0.0,
|
14 |
-
"loss": 0.9219,
|
15 |
-
"step": 1
|
16 |
-
},
|
17 |
-
{
|
18 |
-
"epoch": 0.07,
|
19 |
-
"learning_rate": 0.0001292029674220179,
|
20 |
-
"loss": 0.8431,
|
21 |
-
"step": 2
|
22 |
-
},
|
23 |
-
{
|
24 |
-
"epoch": 0.11,
|
25 |
-
"learning_rate": 0.00020478185834579558,
|
26 |
-
"loss": 0.8259,
|
27 |
-
"step": 3
|
28 |
-
},
|
29 |
-
{
|
30 |
-
"epoch": 0.15,
|
31 |
-
"learning_rate": 0.0002584059348440358,
|
32 |
-
"loss": 0.8513,
|
33 |
-
"step": 4
|
34 |
-
},
|
35 |
-
{
|
36 |
-
"epoch": 0.19,
|
37 |
-
"learning_rate": 0.0003,
|
38 |
-
"loss": 0.8788,
|
39 |
-
"step": 5
|
40 |
-
},
|
41 |
-
{
|
42 |
-
"epoch": 0.22,
|
43 |
-
"learning_rate": 0.0003,
|
44 |
-
"loss": 0.8157,
|
45 |
-
"step": 6
|
46 |
-
},
|
47 |
-
{
|
48 |
-
"epoch": 0.26,
|
49 |
-
"learning_rate": 0.0003,
|
50 |
-
"loss": 0.8119,
|
51 |
-
"step": 7
|
52 |
-
},
|
53 |
-
{
|
54 |
-
"epoch": 0.3,
|
55 |
-
"learning_rate": 0.0003,
|
56 |
-
"loss": 0.8002,
|
57 |
-
"step": 8
|
58 |
-
},
|
59 |
-
{
|
60 |
-
"epoch": 0.33,
|
61 |
-
"learning_rate": 0.0003,
|
62 |
-
"loss": 0.743,
|
63 |
-
"step": 9
|
64 |
-
},
|
65 |
-
{
|
66 |
-
"epoch": 0.37,
|
67 |
-
"learning_rate": 0.0003,
|
68 |
-
"loss": 0.793,
|
69 |
-
"step": 10
|
70 |
-
},
|
71 |
-
{
|
72 |
-
"epoch": 0.37,
|
73 |
-
"eval_accuracy": 0.22686079972512133,
|
74 |
-
"eval_loss": 0.7526410222053528,
|
75 |
-
"eval_runtime": 229.5348,
|
76 |
-
"eval_samples_per_second": 0.157,
|
77 |
-
"eval_steps_per_second": 0.009,
|
78 |
-
"step": 10
|
79 |
-
},
|
80 |
-
{
|
81 |
-
"epoch": 0.41,
|
82 |
-
"learning_rate": 0.0003,
|
83 |
-
"loss": 0.7927,
|
84 |
-
"step": 11
|
85 |
-
},
|
86 |
-
{
|
87 |
-
"epoch": 0.44,
|
88 |
-
"learning_rate": 0.0003,
|
89 |
-
"loss": 0.7881,
|
90 |
-
"step": 12
|
91 |
-
},
|
92 |
-
{
|
93 |
-
"epoch": 0.48,
|
94 |
-
"learning_rate": 0.0003,
|
95 |
-
"loss": 0.7832,
|
96 |
-
"step": 13
|
97 |
-
},
|
98 |
-
{
|
99 |
-
"epoch": 0.52,
|
100 |
-
"learning_rate": 0.0003,
|
101 |
-
"loss": 0.7478,
|
102 |
-
"step": 14
|
103 |
-
},
|
104 |
-
{
|
105 |
-
"epoch": 0.56,
|
106 |
-
"learning_rate": 0.0003,
|
107 |
-
"loss": 0.7615,
|
108 |
-
"step": 15
|
109 |
-
},
|
110 |
-
{
|
111 |
-
"epoch": 0.59,
|
112 |
-
"learning_rate": 0.0003,
|
113 |
-
"loss": 0.741,
|
114 |
-
"step": 16
|
115 |
-
},
|
116 |
-
{
|
117 |
-
"epoch": 0.63,
|
118 |
-
"learning_rate": 0.0003,
|
119 |
-
"loss": 0.7564,
|
120 |
-
"step": 17
|
121 |
-
},
|
122 |
-
{
|
123 |
-
"epoch": 0.67,
|
124 |
-
"learning_rate": 0.0003,
|
125 |
-
"loss": 0.7315,
|
126 |
-
"step": 18
|
127 |
-
},
|
128 |
-
{
|
129 |
-
"epoch": 0.7,
|
130 |
-
"learning_rate": 0.0003,
|
131 |
-
"loss": 0.7531,
|
132 |
-
"step": 19
|
133 |
-
},
|
134 |
-
{
|
135 |
-
"epoch": 0.74,
|
136 |
-
"learning_rate": 0.0003,
|
137 |
-
"loss": 0.7549,
|
138 |
-
"step": 20
|
139 |
-
},
|
140 |
-
{
|
141 |
-
"epoch": 0.74,
|
142 |
-
"eval_accuracy": 0.22750504660052398,
|
143 |
-
"eval_loss": 0.7324042320251465,
|
144 |
-
"eval_runtime": 224.0786,
|
145 |
-
"eval_samples_per_second": 0.161,
|
146 |
-
"eval_steps_per_second": 0.009,
|
147 |
-
"step": 20
|
148 |
-
},
|
149 |
-
{
|
150 |
-
"epoch": 0.78,
|
151 |
-
"learning_rate": 0.0003,
|
152 |
-
"loss": 0.7742,
|
153 |
-
"step": 21
|
154 |
-
},
|
155 |
-
{
|
156 |
-
"epoch": 0.81,
|
157 |
-
"learning_rate": 0.0003,
|
158 |
-
"loss": 0.7763,
|
159 |
-
"step": 22
|
160 |
-
},
|
161 |
-
{
|
162 |
-
"epoch": 0.85,
|
163 |
-
"learning_rate": 0.0003,
|
164 |
-
"loss": 0.6916,
|
165 |
-
"step": 23
|
166 |
-
},
|
167 |
-
{
|
168 |
-
"epoch": 0.89,
|
169 |
-
"learning_rate": 0.0003,
|
170 |
-
"loss": 0.6979,
|
171 |
-
"step": 24
|
172 |
-
},
|
173 |
-
{
|
174 |
-
"epoch": 0.93,
|
175 |
-
"learning_rate": 0.0003,
|
176 |
-
"loss": 0.7256,
|
177 |
-
"step": 25
|
178 |
-
},
|
179 |
-
{
|
180 |
-
"epoch": 0.96,
|
181 |
-
"learning_rate": 0.0003,
|
182 |
-
"loss": 0.7626,
|
183 |
-
"step": 26
|
184 |
-
},
|
185 |
-
{
|
186 |
-
"epoch": 1.0,
|
187 |
-
"learning_rate": 0.0003,
|
188 |
-
"loss": 0.7329,
|
189 |
-
"step": 27
|
190 |
-
},
|
191 |
-
{
|
192 |
-
"epoch": 1.04,
|
193 |
-
"learning_rate": 0.0003,
|
194 |
-
"loss": 0.6927,
|
195 |
-
"step": 28
|
196 |
-
},
|
197 |
-
{
|
198 |
-
"epoch": 1.07,
|
199 |
-
"learning_rate": 0.0003,
|
200 |
-
"loss": 0.6781,
|
201 |
-
"step": 29
|
202 |
-
},
|
203 |
-
{
|
204 |
-
"epoch": 1.11,
|
205 |
-
"learning_rate": 0.0003,
|
206 |
-
"loss": 0.672,
|
207 |
-
"step": 30
|
208 |
-
},
|
209 |
-
{
|
210 |
-
"epoch": 1.11,
|
211 |
-
"eval_accuracy": 0.2278271700382253,
|
212 |
-
"eval_loss": 0.7340984344482422,
|
213 |
-
"eval_runtime": 224.2193,
|
214 |
-
"eval_samples_per_second": 0.161,
|
215 |
-
"eval_steps_per_second": 0.009,
|
216 |
-
"step": 30
|
217 |
-
},
|
218 |
-
{
|
219 |
-
"epoch": 1.15,
|
220 |
-
"learning_rate": 0.0003,
|
221 |
-
"loss": 0.6875,
|
222 |
-
"step": 31
|
223 |
-
},
|
224 |
-
{
|
225 |
-
"epoch": 1.19,
|
226 |
-
"learning_rate": 0.0003,
|
227 |
-
"loss": 0.7242,
|
228 |
-
"step": 32
|
229 |
-
},
|
230 |
-
{
|
231 |
-
"epoch": 1.22,
|
232 |
-
"learning_rate": 0.0003,
|
233 |
-
"loss": 0.7121,
|
234 |
-
"step": 33
|
235 |
-
},
|
236 |
-
{
|
237 |
-
"epoch": 1.26,
|
238 |
-
"learning_rate": 0.0003,
|
239 |
-
"loss": 0.6614,
|
240 |
-
"step": 34
|
241 |
-
},
|
242 |
-
{
|
243 |
-
"epoch": 1.3,
|
244 |
-
"learning_rate": 0.0003,
|
245 |
-
"loss": 0.6649,
|
246 |
-
"step": 35
|
247 |
-
},
|
248 |
-
{
|
249 |
-
"epoch": 1.33,
|
250 |
-
"learning_rate": 0.0003,
|
251 |
-
"loss": 0.6056,
|
252 |
-
"step": 36
|
253 |
-
},
|
254 |
-
{
|
255 |
-
"epoch": 1.37,
|
256 |
-
"learning_rate": 0.0003,
|
257 |
-
"loss": 0.656,
|
258 |
-
"step": 37
|
259 |
-
},
|
260 |
-
{
|
261 |
-
"epoch": 1.41,
|
262 |
-
"learning_rate": 0.0003,
|
263 |
-
"loss": 0.6624,
|
264 |
-
"step": 38
|
265 |
-
},
|
266 |
-
{
|
267 |
-
"epoch": 1.44,
|
268 |
-
"learning_rate": 0.0003,
|
269 |
-
"loss": 0.6428,
|
270 |
-
"step": 39
|
271 |
-
},
|
272 |
-
{
|
273 |
-
"epoch": 1.48,
|
274 |
-
"learning_rate": 0.0003,
|
275 |
-
"loss": 0.6425,
|
276 |
-
"step": 40
|
277 |
-
},
|
278 |
-
{
|
279 |
-
"epoch": 1.48,
|
280 |
-
"eval_accuracy": 0.22737619722544344,
|
281 |
-
"eval_loss": 0.7448343634605408,
|
282 |
-
"eval_runtime": 223.9913,
|
283 |
-
"eval_samples_per_second": 0.161,
|
284 |
-
"eval_steps_per_second": 0.009,
|
285 |
-
"step": 40
|
286 |
-
},
|
287 |
-
{
|
288 |
-
"epoch": 1.52,
|
289 |
-
"learning_rate": 0.0003,
|
290 |
-
"loss": 0.6096,
|
291 |
-
"step": 41
|
292 |
-
},
|
293 |
-
{
|
294 |
-
"epoch": 1.56,
|
295 |
-
"learning_rate": 0.0003,
|
296 |
-
"loss": 0.6172,
|
297 |
-
"step": 42
|
298 |
-
},
|
299 |
-
{
|
300 |
-
"epoch": 1.59,
|
301 |
-
"learning_rate": 0.0003,
|
302 |
-
"loss": 0.5967,
|
303 |
-
"step": 43
|
304 |
-
},
|
305 |
-
{
|
306 |
-
"epoch": 1.63,
|
307 |
-
"learning_rate": 0.0003,
|
308 |
-
"loss": 0.5958,
|
309 |
-
"step": 44
|
310 |
-
},
|
311 |
-
{
|
312 |
-
"epoch": 1.67,
|
313 |
-
"learning_rate": 0.0003,
|
314 |
-
"loss": 0.5921,
|
315 |
-
"step": 45
|
316 |
-
},
|
317 |
-
{
|
318 |
-
"epoch": 1.7,
|
319 |
-
"learning_rate": 0.0003,
|
320 |
-
"loss": 0.6026,
|
321 |
-
"step": 46
|
322 |
-
},
|
323 |
-
{
|
324 |
-
"epoch": 1.74,
|
325 |
-
"learning_rate": 0.0003,
|
326 |
-
"loss": 0.5941,
|
327 |
-
"step": 47
|
328 |
-
},
|
329 |
-
{
|
330 |
-
"epoch": 1.78,
|
331 |
-
"learning_rate": 0.0003,
|
332 |
-
"loss": 0.612,
|
333 |
-
"step": 48
|
334 |
-
},
|
335 |
-
{
|
336 |
-
"epoch": 1.81,
|
337 |
-
"learning_rate": 0.0003,
|
338 |
-
"loss": 0.6084,
|
339 |
-
"step": 49
|
340 |
-
},
|
341 |
-
{
|
342 |
-
"epoch": 1.85,
|
343 |
-
"learning_rate": 0.0003,
|
344 |
-
"loss": 0.5446,
|
345 |
-
"step": 50
|
346 |
-
},
|
347 |
-
{
|
348 |
-
"epoch": 1.85,
|
349 |
-
"eval_accuracy": 0.22646351415195637,
|
350 |
-
"eval_loss": 0.7739428281784058,
|
351 |
-
"eval_runtime": 223.8688,
|
352 |
-
"eval_samples_per_second": 0.161,
|
353 |
-
"eval_steps_per_second": 0.009,
|
354 |
-
"step": 50
|
355 |
-
},
|
356 |
-
{
|
357 |
-
"epoch": 1.89,
|
358 |
-
"learning_rate": 0.0003,
|
359 |
-
"loss": 0.5375,
|
360 |
-
"step": 51
|
361 |
-
},
|
362 |
-
{
|
363 |
-
"epoch": 1.93,
|
364 |
-
"learning_rate": 0.0003,
|
365 |
-
"loss": 0.5548,
|
366 |
-
"step": 52
|
367 |
-
},
|
368 |
-
{
|
369 |
-
"epoch": 1.96,
|
370 |
-
"learning_rate": 0.0003,
|
371 |
-
"loss": 0.6092,
|
372 |
-
"step": 53
|
373 |
-
},
|
374 |
-
{
|
375 |
-
"epoch": 2.0,
|
376 |
-
"learning_rate": 0.0003,
|
377 |
-
"loss": 0.4286,
|
378 |
-
"step": 54
|
379 |
-
}
|
380 |
-
],
|
381 |
-
"logging_steps": 1.0,
|
382 |
-
"max_steps": 135,
|
383 |
-
"num_input_tokens_seen": 0,
|
384 |
-
"num_train_epochs": 5,
|
385 |
-
"save_steps": 500,
|
386 |
-
"total_flos": 68912311697408.0,
|
387 |
-
"train_batch_size": 2,
|
388 |
-
"trial_name": null,
|
389 |
-
"trial_params": null
|
390 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|