Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +329 -0
trainer_state.json
ADDED
@@ -0,0 +1,329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9993906154783668,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 410,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02437538086532602,
|
13 |
+
"grad_norm": 0.381931334733963,
|
14 |
+
"learning_rate": 1.2195121951219513e-05,
|
15 |
+
"loss": 3.6679,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.04875076173065204,
|
20 |
+
"grad_norm": 0.19698743522167206,
|
21 |
+
"learning_rate": 2.4390243902439026e-05,
|
22 |
+
"loss": 3.5098,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.07312614259597806,
|
27 |
+
"grad_norm": 0.26034015417099,
|
28 |
+
"learning_rate": 3.6585365853658535e-05,
|
29 |
+
"loss": 3.3353,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.09750152346130408,
|
34 |
+
"grad_norm": 0.29549556970596313,
|
35 |
+
"learning_rate": 4.878048780487805e-05,
|
36 |
+
"loss": 3.2342,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.1218769043266301,
|
41 |
+
"grad_norm": 0.2754429578781128,
|
42 |
+
"learning_rate": 4.992664502959351e-05,
|
43 |
+
"loss": 3.1812,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.14625228519195613,
|
48 |
+
"grad_norm": 0.3207014799118042,
|
49 |
+
"learning_rate": 4.967362490933723e-05,
|
50 |
+
"loss": 3.1661,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.17062766605728213,
|
55 |
+
"grad_norm": 0.30315226316452026,
|
56 |
+
"learning_rate": 4.924186648858207e-05,
|
57 |
+
"loss": 3.0649,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.19500304692260817,
|
62 |
+
"grad_norm": 0.26564204692840576,
|
63 |
+
"learning_rate": 4.863449747015384e-05,
|
64 |
+
"loss": 3.0024,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.21937842778793418,
|
69 |
+
"grad_norm": 0.3084321618080139,
|
70 |
+
"learning_rate": 4.7855917698280054e-05,
|
71 |
+
"loss": 2.9751,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.2437538086532602,
|
76 |
+
"grad_norm": 0.2992245554924011,
|
77 |
+
"learning_rate": 4.691176728566159e-05,
|
78 |
+
"loss": 3.0007,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.2681291895185862,
|
83 |
+
"grad_norm": 0.3074418306350708,
|
84 |
+
"learning_rate": 4.580888575591068e-05,
|
85 |
+
"loss": 2.9229,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.29250457038391225,
|
90 |
+
"grad_norm": 0.2845458984375,
|
91 |
+
"learning_rate": 4.455526249733178e-05,
|
92 |
+
"loss": 2.8955,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.3168799512492383,
|
97 |
+
"grad_norm": 0.3487965166568756,
|
98 |
+
"learning_rate": 4.3159978886963226e-05,
|
99 |
+
"loss": 2.8837,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.34125533211456427,
|
104 |
+
"grad_norm": 0.307179719209671,
|
105 |
+
"learning_rate": 4.163314250413913e-05,
|
106 |
+
"loss": 2.8862,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.3656307129798903,
|
111 |
+
"grad_norm": 0.3133622407913208,
|
112 |
+
"learning_rate": 3.9985813910135304e-05,
|
113 |
+
"loss": 2.9099,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.39000609384521634,
|
118 |
+
"grad_norm": 0.3793308138847351,
|
119 |
+
"learning_rate": 3.8229926524315016e-05,
|
120 |
+
"loss": 2.8219,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.4143814747105424,
|
125 |
+
"grad_norm": 0.3438339829444885,
|
126 |
+
"learning_rate": 3.6378200177200224e-05,
|
127 |
+
"loss": 2.7811,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.43875685557586835,
|
132 |
+
"grad_norm": 0.33851078152656555,
|
133 |
+
"learning_rate": 3.444404896669865e-05,
|
134 |
+
"loss": 2.8869,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.4631322364411944,
|
139 |
+
"grad_norm": 0.3436271846294403,
|
140 |
+
"learning_rate": 3.2441484084985865e-05,
|
141 |
+
"loss": 2.8078,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.4875076173065204,
|
146 |
+
"grad_norm": 0.29718300700187683,
|
147 |
+
"learning_rate": 3.0385012319974537e-05,
|
148 |
+
"loss": 2.8488,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.5118829981718465,
|
153 |
+
"grad_norm": 0.3266434073448181,
|
154 |
+
"learning_rate": 2.8289530966636625e-05,
|
155 |
+
"loss": 2.7989,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.5362583790371724,
|
160 |
+
"grad_norm": 0.3141980767250061,
|
161 |
+
"learning_rate": 2.617021990945197e-05,
|
162 |
+
"loss": 2.7921,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.5606337599024985,
|
167 |
+
"grad_norm": 0.3401101231575012,
|
168 |
+
"learning_rate": 2.4042431657749117e-05,
|
169 |
+
"loss": 2.8359,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.5850091407678245,
|
174 |
+
"grad_norm": 0.31733494997024536,
|
175 |
+
"learning_rate": 2.1921580130533827e-05,
|
176 |
+
"loss": 2.7748,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.6093845216331505,
|
181 |
+
"grad_norm": 0.3072984218597412,
|
182 |
+
"learning_rate": 1.9823028996459486e-05,
|
183 |
+
"loss": 2.7994,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.6337599024984766,
|
188 |
+
"grad_norm": 0.35069364309310913,
|
189 |
+
"learning_rate": 1.7761980377816287e-05,
|
190 |
+
"loss": 2.7459,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.6581352833638026,
|
195 |
+
"grad_norm": 0.3242040276527405,
|
196 |
+
"learning_rate": 1.5753364724779092e-05,
|
197 |
+
"loss": 2.838,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.6825106642291285,
|
202 |
+
"grad_norm": 0.3111414611339569,
|
203 |
+
"learning_rate": 1.381173265767623e-05,
|
204 |
+
"loss": 2.7298,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.7068860450944546,
|
209 |
+
"grad_norm": 0.3313029408454895,
|
210 |
+
"learning_rate": 1.1951149560785167e-05,
|
211 |
+
"loss": 2.7464,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.7312614259597806,
|
216 |
+
"grad_norm": 0.31917551159858704,
|
217 |
+
"learning_rate": 1.0185093691228534e-05,
|
218 |
+
"loss": 2.7118,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.7556368068251066,
|
223 |
+
"grad_norm": 0.3485548496246338,
|
224 |
+
"learning_rate": 8.526358541080173e-06,
|
225 |
+
"loss": 2.6857,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.7800121876904327,
|
230 |
+
"grad_norm": 0.31290727853775024,
|
231 |
+
"learning_rate": 6.986960159980327e-06,
|
232 |
+
"loss": 2.71,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.8043875685557587,
|
237 |
+
"grad_norm": 0.2982420325279236,
|
238 |
+
"learning_rate": 5.578050109624511e-06,
|
239 |
+
"loss": 2.8041,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.8287629494210847,
|
244 |
+
"grad_norm": 0.29485708475112915,
|
245 |
+
"learning_rate": 4.309834680692832e-06,
|
246 |
+
"loss": 2.6818,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.8531383302864107,
|
251 |
+
"grad_norm": 0.29043450951576233,
|
252 |
+
"learning_rate": 3.1915009574206262e-06,
|
253 |
+
"loss": 2.6666,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.8775137111517367,
|
258 |
+
"grad_norm": 0.32468873262405396,
|
259 |
+
"learning_rate": 2.231150265406512e-06,
|
260 |
+
"loss": 2.7277,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.9018890920170628,
|
265 |
+
"grad_norm": 0.3383389413356781,
|
266 |
+
"learning_rate": 1.435739484768603e-06,
|
267 |
+
"loss": 2.7256,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.9262644728823888,
|
272 |
+
"grad_norm": 0.31536731123924255,
|
273 |
+
"learning_rate": 8.110306537826601e-07,
|
274 |
+
"loss": 2.7143,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.9506398537477148,
|
279 |
+
"grad_norm": 0.30096814036369324,
|
280 |
+
"learning_rate": 3.6154922807863643e-07,
|
281 |
+
"loss": 2.7688,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.9750152346130408,
|
286 |
+
"grad_norm": 0.2999325096607208,
|
287 |
+
"learning_rate": 9.055129777021665e-08,
|
288 |
+
"loss": 2.7355,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.9993906154783668,
|
293 |
+
"grad_norm": 0.31551599502563477,
|
294 |
+
"learning_rate": 0.0,
|
295 |
+
"loss": 2.751,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.9993906154783668,
|
300 |
+
"step": 410,
|
301 |
+
"total_flos": 6.066570733918618e+17,
|
302 |
+
"train_loss": 2.8908857531663847,
|
303 |
+
"train_runtime": 5592.6214,
|
304 |
+
"train_samples_per_second": 2.346,
|
305 |
+
"train_steps_per_second": 0.073
|
306 |
+
}
|
307 |
+
],
|
308 |
+
"logging_steps": 10,
|
309 |
+
"max_steps": 410,
|
310 |
+
"num_input_tokens_seen": 0,
|
311 |
+
"num_train_epochs": 1,
|
312 |
+
"save_steps": 1000,
|
313 |
+
"stateful_callbacks": {
|
314 |
+
"TrainerControl": {
|
315 |
+
"args": {
|
316 |
+
"should_epoch_stop": false,
|
317 |
+
"should_evaluate": false,
|
318 |
+
"should_log": false,
|
319 |
+
"should_save": true,
|
320 |
+
"should_training_stop": true
|
321 |
+
},
|
322 |
+
"attributes": {}
|
323 |
+
}
|
324 |
+
},
|
325 |
+
"total_flos": 6.066570733918618e+17,
|
326 |
+
"train_batch_size": 8,
|
327 |
+
"trial_name": null,
|
328 |
+
"trial_params": null
|
329 |
+
}
|