Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +313 -0
trainer_state.json
ADDED
@@ -0,0 +1,313 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 40.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 1.0,
|
13 |
+
"grad_norm": 3.7084410190582275,
|
14 |
+
"learning_rate": 9.984893239410646e-05,
|
15 |
+
"loss": 2.8145,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 2.0,
|
20 |
+
"grad_norm": 2.862682342529297,
|
21 |
+
"learning_rate": 9.939054496480722e-05,
|
22 |
+
"loss": 2.0914,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 3.0,
|
27 |
+
"grad_norm": 3.9071481227874756,
|
28 |
+
"learning_rate": 9.862764840177711e-05,
|
29 |
+
"loss": 1.931,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 4.0,
|
34 |
+
"grad_norm": 3.2915053367614746,
|
35 |
+
"learning_rate": 9.756494621601604e-05,
|
36 |
+
"loss": 1.7925,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 5.0,
|
41 |
+
"grad_norm": 4.6224589347839355,
|
42 |
+
"learning_rate": 9.620899031988359e-05,
|
43 |
+
"loss": 1.6613,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 6.0,
|
48 |
+
"grad_norm": 6.952088832855225,
|
49 |
+
"learning_rate": 9.45681406323805e-05,
|
50 |
+
"loss": 1.5279,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 7.0,
|
55 |
+
"grad_norm": 10.756989479064941,
|
56 |
+
"learning_rate": 9.265251353746812e-05,
|
57 |
+
"loss": 1.3985,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 8.0,
|
62 |
+
"grad_norm": 7.803109645843506,
|
63 |
+
"learning_rate": 9.047391951319722e-05,
|
64 |
+
"loss": 1.2695,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 9.0,
|
69 |
+
"grad_norm": 7.662687301635742,
|
70 |
+
"learning_rate": 8.804579031618346e-05,
|
71 |
+
"loss": 1.1485,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 10.0,
|
76 |
+
"grad_norm": 12.890942573547363,
|
77 |
+
"learning_rate": 8.54108314553238e-05,
|
78 |
+
"loss": 1.0421,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 11.0,
|
83 |
+
"grad_norm": 11.642329216003418,
|
84 |
+
"learning_rate": 8.253208447561882e-05,
|
85 |
+
"loss": 0.9218,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 12.0,
|
90 |
+
"grad_norm": 10.444331169128418,
|
91 |
+
"learning_rate": 7.945276637710582e-05,
|
92 |
+
"loss": 0.8349,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 13.0,
|
97 |
+
"grad_norm": 12.028226852416992,
|
98 |
+
"learning_rate": 7.619186217983924e-05,
|
99 |
+
"loss": 0.7391,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 14.0,
|
104 |
+
"grad_norm": 13.66382122039795,
|
105 |
+
"learning_rate": 7.2769476442556e-05,
|
106 |
+
"loss": 0.6468,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 15.0,
|
111 |
+
"grad_norm": 20.563343048095703,
|
112 |
+
"learning_rate": 6.920670931138513e-05,
|
113 |
+
"loss": 0.5789,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 16.0,
|
118 |
+
"grad_norm": Infinity,
|
119 |
+
"learning_rate": 6.556285043239972e-05,
|
120 |
+
"loss": 0.5189,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 17.0,
|
125 |
+
"grad_norm": 14.84958267211914,
|
126 |
+
"learning_rate": 6.178679031907088e-05,
|
127 |
+
"loss": 0.4609,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 18.0,
|
132 |
+
"grad_norm": 10.033011436462402,
|
133 |
+
"learning_rate": 5.7938060723919484e-05,
|
134 |
+
"loss": 0.412,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 19.0,
|
139 |
+
"grad_norm": 10.275761604309082,
|
140 |
+
"learning_rate": 5.404039034473205e-05,
|
141 |
+
"loss": 0.365,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 20.0,
|
146 |
+
"grad_norm": 21.667903900146484,
|
147 |
+
"learning_rate": 5.011780961550321e-05,
|
148 |
+
"loss": 0.3297,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 21.0,
|
153 |
+
"grad_norm": 15.773643493652344,
|
154 |
+
"learning_rate": 4.619450255081913e-05,
|
155 |
+
"loss": 0.2924,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 22.0,
|
160 |
+
"grad_norm": 8.906744956970215,
|
161 |
+
"learning_rate": 4.229465764336555e-05,
|
162 |
+
"loss": 0.2637,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 23.0,
|
167 |
+
"grad_norm": 7.449186325073242,
|
168 |
+
"learning_rate": 3.848052866617049e-05,
|
169 |
+
"loss": 0.2479,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 24.0,
|
174 |
+
"grad_norm": 15.881937980651855,
|
175 |
+
"learning_rate": 3.4736008447197823e-05,
|
176 |
+
"loss": 0.2266,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 25.0,
|
181 |
+
"grad_norm": 9.639005661010742,
|
182 |
+
"learning_rate": 3.104737877380828e-05,
|
183 |
+
"loss": 0.1991,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 26.0,
|
188 |
+
"grad_norm": 94.2000961303711,
|
189 |
+
"learning_rate": 2.7475598312664285e-05,
|
190 |
+
"loss": 0.1865,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 27.0,
|
195 |
+
"grad_norm": 6.111320972442627,
|
196 |
+
"learning_rate": 2.404268827804637e-05,
|
197 |
+
"loss": 0.1715,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 28.0,
|
202 |
+
"grad_norm": 9.6605806350708,
|
203 |
+
"learning_rate": 2.0769813701876333e-05,
|
204 |
+
"loss": 0.1607,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 29.0,
|
209 |
+
"grad_norm": 4.928094863891602,
|
210 |
+
"learning_rate": 1.7677152944257513e-05,
|
211 |
+
"loss": 0.1465,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 30.0,
|
216 |
+
"grad_norm": 6.157331943511963,
|
217 |
+
"learning_rate": 1.4783773287174684e-05,
|
218 |
+
"loss": 0.1411,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 31.0,
|
223 |
+
"grad_norm": 9.299696922302246,
|
224 |
+
"learning_rate": 1.2107513378360164e-05,
|
225 |
+
"loss": 0.1338,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 32.0,
|
230 |
+
"grad_norm": 3.8831334114074707,
|
231 |
+
"learning_rate": 9.664873250098511e-06,
|
232 |
+
"loss": 0.1273,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 33.0,
|
237 |
+
"grad_norm": 5.329362869262695,
|
238 |
+
"learning_rate": 7.470912591040696e-06,
|
239 |
+
"loss": 0.1222,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 34.0,
|
244 |
+
"grad_norm": 6.742835521697998,
|
245 |
+
"learning_rate": 5.539157898215786e-06,
|
246 |
+
"loss": 0.1175,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 35.0,
|
251 |
+
"grad_norm": 3.8493688106536865,
|
252 |
+
"learning_rate": 3.881519081678658e-06,
|
253 |
+
"loss": 0.1162,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 36.0,
|
258 |
+
"grad_norm": 6.8995041847229,
|
259 |
+
"learning_rate": 2.5082160359541142e-06,
|
260 |
+
"loss": 0.1136,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 37.0,
|
265 |
+
"grad_norm": 2.0875461101531982,
|
266 |
+
"learning_rate": 1.4277156309886575e-06,
|
267 |
+
"loss": 0.1131,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 38.0,
|
272 |
+
"grad_norm": 4.801537036895752,
|
273 |
+
"learning_rate": 6.466795110817214e-07,
|
274 |
+
"loss": 0.1097,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 39.0,
|
279 |
+
"grad_norm": 3.8173491954803467,
|
280 |
+
"learning_rate": 1.6992302363341704e-07,
|
281 |
+
"loss": 0.1082,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 40.0,
|
286 |
+
"grad_norm": 9.46168041229248,
|
287 |
+
"learning_rate": 3.8553092647219956e-10,
|
288 |
+
"loss": 0.1102,
|
289 |
+
"step": 4000
|
290 |
+
}
|
291 |
+
],
|
292 |
+
"logging_steps": 500,
|
293 |
+
"max_steps": 4000,
|
294 |
+
"num_input_tokens_seen": 0,
|
295 |
+
"num_train_epochs": 40,
|
296 |
+
"save_steps": 500,
|
297 |
+
"stateful_callbacks": {
|
298 |
+
"TrainerControl": {
|
299 |
+
"args": {
|
300 |
+
"should_epoch_stop": false,
|
301 |
+
"should_evaluate": false,
|
302 |
+
"should_log": false,
|
303 |
+
"should_save": true,
|
304 |
+
"should_training_stop": true
|
305 |
+
},
|
306 |
+
"attributes": {}
|
307 |
+
}
|
308 |
+
},
|
309 |
+
"total_flos": 615486789058560.0,
|
310 |
+
"train_batch_size": 1,
|
311 |
+
"trial_name": null,
|
312 |
+
"trial_params": null
|
313 |
+
}
|