Upload trainer_state.json
Browse files- trainer_state.json +324 -0
trainer_state.json
ADDED
@@ -0,0 +1,324 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.9465930018416207,
|
5 |
+
"global_step": 24000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.06,
|
12 |
+
"learning_rate": 1.9590750971966443e-05,
|
13 |
+
"loss": 2.4744,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.12,
|
18 |
+
"learning_rate": 1.9181501943932885e-05,
|
19 |
+
"loss": 1.9135,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.18,
|
24 |
+
"learning_rate": 1.8772252915899326e-05,
|
25 |
+
"loss": 1.796,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.25,
|
30 |
+
"learning_rate": 1.8363003887865768e-05,
|
31 |
+
"loss": 1.7038,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.31,
|
36 |
+
"learning_rate": 1.795375485983221e-05,
|
37 |
+
"loss": 1.6516,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.37,
|
42 |
+
"learning_rate": 1.754450583179865e-05,
|
43 |
+
"loss": 1.6246,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.43,
|
48 |
+
"learning_rate": 1.7135256803765093e-05,
|
49 |
+
"loss": 1.5577,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.49,
|
54 |
+
"learning_rate": 1.6726007775731534e-05,
|
55 |
+
"loss": 1.5208,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.55,
|
60 |
+
"learning_rate": 1.6316758747697976e-05,
|
61 |
+
"loss": 1.5118,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.61,
|
66 |
+
"learning_rate": 1.5907509719664418e-05,
|
67 |
+
"loss": 1.5089,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.68,
|
72 |
+
"learning_rate": 1.5498260691630856e-05,
|
73 |
+
"loss": 1.4391,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.74,
|
78 |
+
"learning_rate": 1.50890116635973e-05,
|
79 |
+
"loss": 1.4344,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.8,
|
84 |
+
"learning_rate": 1.467976263556374e-05,
|
85 |
+
"loss": 1.4238,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.86,
|
90 |
+
"learning_rate": 1.4270513607530182e-05,
|
91 |
+
"loss": 1.4384,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.92,
|
96 |
+
"learning_rate": 1.3861264579496626e-05,
|
97 |
+
"loss": 1.3665,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.98,
|
102 |
+
"learning_rate": 1.3452015551463067e-05,
|
103 |
+
"loss": 1.3999,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.0,
|
108 |
+
"eval_em": 0.5297689323663349,
|
109 |
+
"eval_f1": 0.6106438572272215,
|
110 |
+
"eval_loss": 1.3775073289871216,
|
111 |
+
"eval_runtime": 131.0166,
|
112 |
+
"eval_samples_per_second": 90.508,
|
113 |
+
"eval_steps_per_second": 5.663,
|
114 |
+
"step": 8145
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 1.04,
|
118 |
+
"learning_rate": 1.3042766523429509e-05,
|
119 |
+
"loss": 1.1856,
|
120 |
+
"step": 8500
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 1.1,
|
124 |
+
"learning_rate": 1.263351749539595e-05,
|
125 |
+
"loss": 1.0984,
|
126 |
+
"step": 9000
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.17,
|
130 |
+
"learning_rate": 1.222426846736239e-05,
|
131 |
+
"loss": 1.1179,
|
132 |
+
"step": 9500
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 1.23,
|
136 |
+
"learning_rate": 1.1815019439328832e-05,
|
137 |
+
"loss": 1.0886,
|
138 |
+
"step": 10000
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 1.29,
|
142 |
+
"learning_rate": 1.1405770411295274e-05,
|
143 |
+
"loss": 1.1104,
|
144 |
+
"step": 10500
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.35,
|
148 |
+
"learning_rate": 1.0996521383261715e-05,
|
149 |
+
"loss": 1.1339,
|
150 |
+
"step": 11000
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 1.41,
|
154 |
+
"learning_rate": 1.0587272355228157e-05,
|
155 |
+
"loss": 1.1361,
|
156 |
+
"step": 11500
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 1.47,
|
160 |
+
"learning_rate": 1.01780233271946e-05,
|
161 |
+
"loss": 1.1057,
|
162 |
+
"step": 12000
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 1.53,
|
166 |
+
"learning_rate": 9.76877429916104e-06,
|
167 |
+
"loss": 1.0967,
|
168 |
+
"step": 12500
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.6,
|
172 |
+
"learning_rate": 9.359525271127482e-06,
|
173 |
+
"loss": 1.1119,
|
174 |
+
"step": 13000
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 1.66,
|
178 |
+
"learning_rate": 8.950276243093923e-06,
|
179 |
+
"loss": 1.0718,
|
180 |
+
"step": 13500
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 1.72,
|
184 |
+
"learning_rate": 8.541027215060365e-06,
|
185 |
+
"loss": 1.093,
|
186 |
+
"step": 14000
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.78,
|
190 |
+
"learning_rate": 8.131778187026806e-06,
|
191 |
+
"loss": 1.0782,
|
192 |
+
"step": 14500
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 1.84,
|
196 |
+
"learning_rate": 7.722529158993248e-06,
|
197 |
+
"loss": 1.0655,
|
198 |
+
"step": 15000
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 1.9,
|
202 |
+
"learning_rate": 7.313280130959689e-06,
|
203 |
+
"loss": 1.0953,
|
204 |
+
"step": 15500
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 1.96,
|
208 |
+
"learning_rate": 6.904031102926131e-06,
|
209 |
+
"loss": 1.0765,
|
210 |
+
"step": 16000
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 2.0,
|
214 |
+
"eval_em": 0.5461291954798448,
|
215 |
+
"eval_f1": 0.6263308103966033,
|
216 |
+
"eval_loss": 1.385632038116455,
|
217 |
+
"eval_runtime": 134.345,
|
218 |
+
"eval_samples_per_second": 88.265,
|
219 |
+
"eval_steps_per_second": 5.523,
|
220 |
+
"step": 16290
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 2.03,
|
224 |
+
"learning_rate": 6.494782074892573e-06,
|
225 |
+
"loss": 0.9839,
|
226 |
+
"step": 16500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 2.09,
|
230 |
+
"learning_rate": 6.0855330468590145e-06,
|
231 |
+
"loss": 0.8102,
|
232 |
+
"step": 17000
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 2.15,
|
236 |
+
"learning_rate": 5.676284018825455e-06,
|
237 |
+
"loss": 0.8344,
|
238 |
+
"step": 17500
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 2.21,
|
242 |
+
"learning_rate": 5.267034990791897e-06,
|
243 |
+
"loss": 0.8254,
|
244 |
+
"step": 18000
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 2.27,
|
248 |
+
"learning_rate": 4.857785962758339e-06,
|
249 |
+
"loss": 0.8348,
|
250 |
+
"step": 18500
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 2.33,
|
254 |
+
"learning_rate": 4.44853693472478e-06,
|
255 |
+
"loss": 0.8377,
|
256 |
+
"step": 19000
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 2.39,
|
260 |
+
"learning_rate": 4.0392879066912225e-06,
|
261 |
+
"loss": 0.8376,
|
262 |
+
"step": 19500
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 2.46,
|
266 |
+
"learning_rate": 3.6300388786576637e-06,
|
267 |
+
"loss": 0.8231,
|
268 |
+
"step": 20000
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 2.52,
|
272 |
+
"learning_rate": 3.220789850624105e-06,
|
273 |
+
"loss": 0.8427,
|
274 |
+
"step": 20500
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 2.58,
|
278 |
+
"learning_rate": 2.8115408225905465e-06,
|
279 |
+
"loss": 0.8219,
|
280 |
+
"step": 21000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 2.64,
|
284 |
+
"learning_rate": 2.402291794556988e-06,
|
285 |
+
"loss": 0.8272,
|
286 |
+
"step": 21500
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 2.7,
|
290 |
+
"learning_rate": 1.9930427665234297e-06,
|
291 |
+
"loss": 0.8353,
|
292 |
+
"step": 22000
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 2.76,
|
296 |
+
"learning_rate": 1.5837937384898713e-06,
|
297 |
+
"loss": 0.8305,
|
298 |
+
"step": 22500
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 2.82,
|
302 |
+
"learning_rate": 1.1745447104563129e-06,
|
303 |
+
"loss": 0.8231,
|
304 |
+
"step": 23000
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 2.89,
|
308 |
+
"learning_rate": 7.652956824227544e-07,
|
309 |
+
"loss": 0.8398,
|
310 |
+
"step": 23500
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 2.95,
|
314 |
+
"learning_rate": 3.5604665438919586e-07,
|
315 |
+
"loss": 0.8243,
|
316 |
+
"step": 24000
|
317 |
+
}
|
318 |
+
],
|
319 |
+
"max_steps": 24435,
|
320 |
+
"num_train_epochs": 3,
|
321 |
+
"total_flos": 7.525072232402227e+16,
|
322 |
+
"trial_name": null,
|
323 |
+
"trial_params": null
|
324 |
+
}
|