ChasingMercer
commited on
Commit
•
2d1218f
1
Parent(s):
e9efa32
- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +464 -131
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 6.0,
|
3 |
+
"total_flos": 2.549919337377528e+18,
|
4 |
+
"train_loss": 0.2920494159759834,
|
5 |
+
"train_runtime": 2402.1234,
|
6 |
+
"train_samples_per_second": 13.71,
|
7 |
+
"train_steps_per_second": 0.427
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 6.0,
|
3 |
+
"total_flos": 2.549919337377528e+18,
|
4 |
+
"train_loss": 0.2920494159759834,
|
5 |
+
"train_runtime": 2402.1234,
|
6 |
+
"train_samples_per_second": 13.71,
|
7 |
+
"train_steps_per_second": 0.427
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,358 +1,691 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./weather-base/checkpoint-
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
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":
|
13 |
-
"loss": 2.
|
14 |
"step": 10
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.12,
|
18 |
-
"learning_rate":
|
19 |
-
"loss": 2.
|
20 |
"step": 20
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.17,
|
24 |
-
"learning_rate":
|
25 |
-
"loss":
|
26 |
"step": 30
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.23,
|
30 |
-
"learning_rate":
|
31 |
-
"loss": 1.
|
32 |
"step": 40
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.29,
|
36 |
-
"learning_rate":
|
37 |
-
"loss":
|
38 |
"step": 50
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.35,
|
42 |
-
"learning_rate":
|
43 |
-
"loss": 0.
|
44 |
"step": 60
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.41,
|
48 |
-
"learning_rate":
|
49 |
-
"loss": 0.
|
50 |
"step": 70
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.47,
|
54 |
-
"learning_rate":
|
55 |
-
"loss": 0.
|
56 |
"step": 80
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.52,
|
60 |
-
"learning_rate": 4.
|
61 |
-
"loss": 0.
|
62 |
"step": 90
|
63 |
},
|
64 |
{
|
65 |
"epoch": 0.58,
|
66 |
-
"learning_rate": 4.
|
67 |
-
"loss": 0.
|
68 |
"step": 100
|
69 |
},
|
70 |
{
|
71 |
"epoch": 0.64,
|
72 |
-
"learning_rate": 4.
|
73 |
-
"loss": 0.
|
74 |
"step": 110
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.7,
|
78 |
-
"learning_rate": 4.
|
79 |
-
"loss": 0.
|
80 |
"step": 120
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.76,
|
84 |
-
"learning_rate": 4.
|
85 |
-
"loss": 0.
|
86 |
"step": 130
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.82,
|
90 |
-
"learning_rate": 4.
|
91 |
-
"loss": 0.
|
92 |
"step": 140
|
93 |
},
|
94 |
{
|
95 |
"epoch": 0.87,
|
96 |
-
"learning_rate":
|
97 |
-
"loss": 0.
|
98 |
"step": 150
|
99 |
},
|
100 |
{
|
101 |
"epoch": 0.93,
|
102 |
-
"learning_rate":
|
103 |
-
"loss": 0.
|
104 |
"step": 160
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.99,
|
108 |
-
"learning_rate":
|
109 |
-
"loss": 0.
|
110 |
"step": 170
|
111 |
},
|
112 |
{
|
113 |
"epoch": 1.0,
|
114 |
-
"eval_accuracy": 0.
|
115 |
-
"eval_loss": 0.
|
116 |
-
"eval_runtime":
|
117 |
-
"eval_samples_per_second":
|
118 |
-
"eval_steps_per_second": 5.
|
119 |
"step": 171
|
120 |
},
|
121 |
{
|
122 |
"epoch": 1.05,
|
123 |
-
"learning_rate":
|
124 |
-
"loss": 0.
|
125 |
"step": 180
|
126 |
},
|
127 |
{
|
128 |
"epoch": 1.11,
|
129 |
-
"learning_rate":
|
130 |
-
"loss": 0.
|
131 |
"step": 190
|
132 |
},
|
133 |
{
|
134 |
"epoch": 1.17,
|
135 |
-
"learning_rate":
|
136 |
-
"loss": 0.
|
137 |
"step": 200
|
138 |
},
|
139 |
{
|
140 |
"epoch": 1.23,
|
141 |
-
"learning_rate":
|
142 |
-
"loss": 0.
|
143 |
"step": 210
|
144 |
},
|
145 |
{
|
146 |
"epoch": 1.29,
|
147 |
-
"learning_rate":
|
148 |
-
"loss": 0.
|
149 |
"step": 220
|
150 |
},
|
151 |
{
|
152 |
"epoch": 1.34,
|
153 |
-
"learning_rate":
|
154 |
-
"loss": 0.
|
155 |
"step": 230
|
156 |
},
|
157 |
{
|
158 |
"epoch": 1.4,
|
159 |
-
"learning_rate":
|
160 |
-
"loss": 0.
|
161 |
"step": 240
|
162 |
},
|
163 |
{
|
164 |
"epoch": 1.46,
|
165 |
-
"learning_rate":
|
166 |
-
"loss": 0.
|
167 |
"step": 250
|
168 |
},
|
169 |
{
|
170 |
"epoch": 1.52,
|
171 |
-
"learning_rate":
|
172 |
-
"loss": 0.
|
173 |
"step": 260
|
174 |
},
|
175 |
{
|
176 |
"epoch": 1.58,
|
177 |
-
"learning_rate":
|
178 |
-
"loss": 0.
|
179 |
"step": 270
|
180 |
},
|
181 |
{
|
182 |
"epoch": 1.63,
|
183 |
-
"learning_rate":
|
184 |
-
"loss": 0.
|
185 |
"step": 280
|
186 |
},
|
187 |
{
|
188 |
"epoch": 1.69,
|
189 |
-
"learning_rate":
|
190 |
-
"loss": 0.
|
191 |
"step": 290
|
192 |
},
|
193 |
{
|
194 |
"epoch": 1.75,
|
195 |
-
"learning_rate":
|
196 |
-
"loss": 0.
|
197 |
"step": 300
|
198 |
},
|
199 |
{
|
200 |
"epoch": 1.81,
|
201 |
-
"learning_rate":
|
202 |
-
"loss": 0.
|
203 |
"step": 310
|
204 |
},
|
205 |
{
|
206 |
"epoch": 1.87,
|
207 |
-
"learning_rate":
|
208 |
-
"loss": 0.
|
209 |
"step": 320
|
210 |
},
|
211 |
{
|
212 |
"epoch": 1.93,
|
213 |
-
"learning_rate":
|
214 |
-
"loss": 0.
|
215 |
"step": 330
|
216 |
},
|
217 |
{
|
218 |
"epoch": 1.98,
|
219 |
-
"learning_rate":
|
220 |
-
"loss": 0.
|
221 |
"step": 340
|
222 |
},
|
223 |
{
|
224 |
"epoch": 2.0,
|
225 |
-
"eval_accuracy": 0.
|
226 |
-
"eval_loss": 0.
|
227 |
-
"eval_runtime":
|
228 |
-
"eval_samples_per_second":
|
229 |
-
"eval_steps_per_second": 2.
|
230 |
"step": 342
|
231 |
},
|
232 |
{
|
233 |
"epoch": 2.05,
|
234 |
-
"learning_rate":
|
235 |
-
"loss": 0.
|
236 |
"step": 350
|
237 |
},
|
238 |
{
|
239 |
"epoch": 2.1,
|
240 |
-
"learning_rate":
|
241 |
-
"loss": 0.
|
242 |
"step": 360
|
243 |
},
|
244 |
{
|
245 |
"epoch": 2.16,
|
246 |
-
"learning_rate":
|
247 |
-
"loss": 0.
|
248 |
"step": 370
|
249 |
},
|
250 |
{
|
251 |
"epoch": 2.22,
|
252 |
-
"learning_rate":
|
253 |
-
"loss": 0.
|
254 |
"step": 380
|
255 |
},
|
256 |
{
|
257 |
"epoch": 2.28,
|
258 |
-
"learning_rate":
|
259 |
-
"loss": 0.
|
260 |
"step": 390
|
261 |
},
|
262 |
{
|
263 |
"epoch": 2.34,
|
264 |
-
"learning_rate":
|
265 |
-
"loss": 0.
|
266 |
"step": 400
|
267 |
},
|
268 |
{
|
269 |
"epoch": 2.4,
|
270 |
-
"learning_rate":
|
271 |
-
"loss": 0.
|
272 |
"step": 410
|
273 |
},
|
274 |
{
|
275 |
"epoch": 2.45,
|
276 |
-
"learning_rate":
|
277 |
-
"loss": 0.
|
278 |
"step": 420
|
279 |
},
|
280 |
{
|
281 |
"epoch": 2.51,
|
282 |
-
"learning_rate":
|
283 |
-
"loss": 0.
|
284 |
"step": 430
|
285 |
},
|
286 |
{
|
287 |
"epoch": 2.57,
|
288 |
-
"learning_rate":
|
289 |
-
"loss": 0.
|
290 |
"step": 440
|
291 |
},
|
292 |
{
|
293 |
"epoch": 2.63,
|
294 |
-
"learning_rate":
|
295 |
-
"loss": 0.
|
296 |
"step": 450
|
297 |
},
|
298 |
{
|
299 |
"epoch": 2.69,
|
300 |
-
"learning_rate":
|
301 |
-
"loss": 0.
|
302 |
"step": 460
|
303 |
},
|
304 |
{
|
305 |
"epoch": 2.75,
|
306 |
-
"learning_rate":
|
307 |
-
"loss": 0.
|
308 |
"step": 470
|
309 |
},
|
310 |
{
|
311 |
"epoch": 2.8,
|
312 |
-
"learning_rate":
|
313 |
-
"loss": 0.
|
314 |
"step": 480
|
315 |
},
|
316 |
{
|
317 |
"epoch": 2.86,
|
318 |
-
"learning_rate": 2.
|
319 |
-
"loss": 0.
|
320 |
"step": 490
|
321 |
},
|
322 |
{
|
323 |
"epoch": 2.92,
|
324 |
-
"learning_rate":
|
325 |
-
"loss": 0.
|
326 |
"step": 500
|
327 |
},
|
328 |
{
|
329 |
"epoch": 2.98,
|
330 |
-
"learning_rate":
|
331 |
-
"loss": 0.
|
332 |
"step": 510
|
333 |
},
|
334 |
{
|
335 |
"epoch": 3.0,
|
336 |
-
"eval_accuracy": 0.
|
337 |
-
"eval_loss": 0.
|
338 |
-
"eval_runtime":
|
339 |
-
"eval_samples_per_second":
|
340 |
-
"eval_steps_per_second": 5.
|
341 |
"step": 513
|
342 |
},
|
343 |
{
|
344 |
-
"epoch": 3.
|
345 |
-
"
|
346 |
-
"
|
347 |
-
"
|
348 |
-
|
349 |
-
|
350 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
351 |
}
|
352 |
],
|
353 |
-
"max_steps":
|
354 |
-
"num_train_epochs":
|
355 |
-
"total_flos":
|
356 |
"trial_name": null,
|
357 |
"trial_params": null
|
358 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9358600583090378,
|
3 |
+
"best_model_checkpoint": "./weather-base/checkpoint-855",
|
4 |
+
"epoch": 5.995633187772926,
|
5 |
+
"global_step": 1026,
|
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": 4.854368932038835e-06,
|
13 |
+
"loss": 2.5891,
|
14 |
"step": 10
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.12,
|
18 |
+
"learning_rate": 9.70873786407767e-06,
|
19 |
+
"loss": 2.3939,
|
20 |
"step": 20
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.17,
|
24 |
+
"learning_rate": 1.4563106796116505e-05,
|
25 |
+
"loss": 2.0509,
|
26 |
"step": 30
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.23,
|
30 |
+
"learning_rate": 1.941747572815534e-05,
|
31 |
+
"loss": 1.7403,
|
32 |
"step": 40
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.29,
|
36 |
+
"learning_rate": 2.4271844660194176e-05,
|
37 |
+
"loss": 1.3148,
|
38 |
"step": 50
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.35,
|
42 |
+
"learning_rate": 2.912621359223301e-05,
|
43 |
+
"loss": 0.8922,
|
44 |
"step": 60
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.41,
|
48 |
+
"learning_rate": 3.398058252427185e-05,
|
49 |
+
"loss": 0.6958,
|
50 |
"step": 70
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.47,
|
54 |
+
"learning_rate": 3.883495145631068e-05,
|
55 |
+
"loss": 0.5293,
|
56 |
"step": 80
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.52,
|
60 |
+
"learning_rate": 4.368932038834951e-05,
|
61 |
+
"loss": 0.5498,
|
62 |
"step": 90
|
63 |
},
|
64 |
{
|
65 |
"epoch": 0.58,
|
66 |
+
"learning_rate": 4.854368932038835e-05,
|
67 |
+
"loss": 0.4659,
|
68 |
"step": 100
|
69 |
},
|
70 |
{
|
71 |
"epoch": 0.64,
|
72 |
+
"learning_rate": 4.962080173347779e-05,
|
73 |
+
"loss": 0.4436,
|
74 |
"step": 110
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.7,
|
78 |
+
"learning_rate": 4.907908992416035e-05,
|
79 |
+
"loss": 0.4404,
|
80 |
"step": 120
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.76,
|
84 |
+
"learning_rate": 4.853737811484291e-05,
|
85 |
+
"loss": 0.4192,
|
86 |
"step": 130
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.82,
|
90 |
+
"learning_rate": 4.799566630552546e-05,
|
91 |
+
"loss": 0.3667,
|
92 |
"step": 140
|
93 |
},
|
94 |
{
|
95 |
"epoch": 0.87,
|
96 |
+
"learning_rate": 4.745395449620802e-05,
|
97 |
+
"loss": 0.3844,
|
98 |
"step": 150
|
99 |
},
|
100 |
{
|
101 |
"epoch": 0.93,
|
102 |
+
"learning_rate": 4.691224268689058e-05,
|
103 |
+
"loss": 0.4201,
|
104 |
"step": 160
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.99,
|
108 |
+
"learning_rate": 4.6370530877573135e-05,
|
109 |
+
"loss": 0.3368,
|
110 |
"step": 170
|
111 |
},
|
112 |
{
|
113 |
"epoch": 1.0,
|
114 |
+
"eval_accuracy": 0.9008746355685131,
|
115 |
+
"eval_loss": 0.2779894173145294,
|
116 |
+
"eval_runtime": 14.9875,
|
117 |
+
"eval_samples_per_second": 45.771,
|
118 |
+
"eval_steps_per_second": 5.738,
|
119 |
"step": 171
|
120 |
},
|
121 |
{
|
122 |
"epoch": 1.05,
|
123 |
+
"learning_rate": 4.582881906825569e-05,
|
124 |
+
"loss": 0.3262,
|
125 |
"step": 180
|
126 |
},
|
127 |
{
|
128 |
"epoch": 1.11,
|
129 |
+
"learning_rate": 4.528710725893825e-05,
|
130 |
+
"loss": 0.2425,
|
131 |
"step": 190
|
132 |
},
|
133 |
{
|
134 |
"epoch": 1.17,
|
135 |
+
"learning_rate": 4.47453954496208e-05,
|
136 |
+
"loss": 0.2162,
|
137 |
"step": 200
|
138 |
},
|
139 |
{
|
140 |
"epoch": 1.23,
|
141 |
+
"learning_rate": 4.420368364030336e-05,
|
142 |
+
"loss": 0.3276,
|
143 |
"step": 210
|
144 |
},
|
145 |
{
|
146 |
"epoch": 1.29,
|
147 |
+
"learning_rate": 4.366197183098591e-05,
|
148 |
+
"loss": 0.3236,
|
149 |
"step": 220
|
150 |
},
|
151 |
{
|
152 |
"epoch": 1.34,
|
153 |
+
"learning_rate": 4.312026002166848e-05,
|
154 |
+
"loss": 0.2479,
|
155 |
"step": 230
|
156 |
},
|
157 |
{
|
158 |
"epoch": 1.4,
|
159 |
+
"learning_rate": 4.2578548212351035e-05,
|
160 |
+
"loss": 0.3096,
|
161 |
"step": 240
|
162 |
},
|
163 |
{
|
164 |
"epoch": 1.46,
|
165 |
+
"learning_rate": 4.2036836403033586e-05,
|
166 |
+
"loss": 0.2766,
|
167 |
"step": 250
|
168 |
},
|
169 |
{
|
170 |
"epoch": 1.52,
|
171 |
+
"learning_rate": 4.1495124593716144e-05,
|
172 |
+
"loss": 0.301,
|
173 |
"step": 260
|
174 |
},
|
175 |
{
|
176 |
"epoch": 1.58,
|
177 |
+
"learning_rate": 4.09534127843987e-05,
|
178 |
+
"loss": 0.2831,
|
179 |
"step": 270
|
180 |
},
|
181 |
{
|
182 |
"epoch": 1.63,
|
183 |
+
"learning_rate": 4.0411700975081254e-05,
|
184 |
+
"loss": 0.2805,
|
185 |
"step": 280
|
186 |
},
|
187 |
{
|
188 |
"epoch": 1.69,
|
189 |
+
"learning_rate": 3.986998916576382e-05,
|
190 |
+
"loss": 0.2897,
|
191 |
"step": 290
|
192 |
},
|
193 |
{
|
194 |
"epoch": 1.75,
|
195 |
+
"learning_rate": 3.932827735644638e-05,
|
196 |
+
"loss": 0.2581,
|
197 |
"step": 300
|
198 |
},
|
199 |
{
|
200 |
"epoch": 1.81,
|
201 |
+
"learning_rate": 3.878656554712893e-05,
|
202 |
+
"loss": 0.2201,
|
203 |
"step": 310
|
204 |
},
|
205 |
{
|
206 |
"epoch": 1.87,
|
207 |
+
"learning_rate": 3.8244853737811486e-05,
|
208 |
+
"loss": 0.3577,
|
209 |
"step": 320
|
210 |
},
|
211 |
{
|
212 |
"epoch": 1.93,
|
213 |
+
"learning_rate": 3.7703141928494044e-05,
|
214 |
+
"loss": 0.2503,
|
215 |
"step": 330
|
216 |
},
|
217 |
{
|
218 |
"epoch": 1.98,
|
219 |
+
"learning_rate": 3.71614301191766e-05,
|
220 |
+
"loss": 0.2129,
|
221 |
"step": 340
|
222 |
},
|
223 |
{
|
224 |
"epoch": 2.0,
|
225 |
+
"eval_accuracy": 0.9300291545189504,
|
226 |
+
"eval_loss": 0.23329117894172668,
|
227 |
+
"eval_runtime": 42.4804,
|
228 |
+
"eval_samples_per_second": 16.149,
|
229 |
+
"eval_steps_per_second": 2.024,
|
230 |
"step": 342
|
231 |
},
|
232 |
{
|
233 |
"epoch": 2.05,
|
234 |
+
"learning_rate": 3.661971830985916e-05,
|
235 |
+
"loss": 0.1552,
|
236 |
"step": 350
|
237 |
},
|
238 |
{
|
239 |
"epoch": 2.1,
|
240 |
+
"learning_rate": 3.607800650054171e-05,
|
241 |
+
"loss": 0.2058,
|
242 |
"step": 360
|
243 |
},
|
244 |
{
|
245 |
"epoch": 2.16,
|
246 |
+
"learning_rate": 3.553629469122427e-05,
|
247 |
+
"loss": 0.1833,
|
248 |
"step": 370
|
249 |
},
|
250 |
{
|
251 |
"epoch": 2.22,
|
252 |
+
"learning_rate": 3.499458288190683e-05,
|
253 |
+
"loss": 0.1666,
|
254 |
"step": 380
|
255 |
},
|
256 |
{
|
257 |
"epoch": 2.28,
|
258 |
+
"learning_rate": 3.445287107258938e-05,
|
259 |
+
"loss": 0.1654,
|
260 |
"step": 390
|
261 |
},
|
262 |
{
|
263 |
"epoch": 2.34,
|
264 |
+
"learning_rate": 3.3911159263271944e-05,
|
265 |
+
"loss": 0.229,
|
266 |
"step": 400
|
267 |
},
|
268 |
{
|
269 |
"epoch": 2.4,
|
270 |
+
"learning_rate": 3.33694474539545e-05,
|
271 |
+
"loss": 0.1942,
|
272 |
"step": 410
|
273 |
},
|
274 |
{
|
275 |
"epoch": 2.45,
|
276 |
+
"learning_rate": 3.2827735644637054e-05,
|
277 |
+
"loss": 0.213,
|
278 |
"step": 420
|
279 |
},
|
280 |
{
|
281 |
"epoch": 2.51,
|
282 |
+
"learning_rate": 3.228602383531961e-05,
|
283 |
+
"loss": 0.1931,
|
284 |
"step": 430
|
285 |
},
|
286 |
{
|
287 |
"epoch": 2.57,
|
288 |
+
"learning_rate": 3.174431202600217e-05,
|
289 |
+
"loss": 0.1458,
|
290 |
"step": 440
|
291 |
},
|
292 |
{
|
293 |
"epoch": 2.63,
|
294 |
+
"learning_rate": 3.120260021668472e-05,
|
295 |
+
"loss": 0.1693,
|
296 |
"step": 450
|
297 |
},
|
298 |
{
|
299 |
"epoch": 2.69,
|
300 |
+
"learning_rate": 3.0660888407367286e-05,
|
301 |
+
"loss": 0.2115,
|
302 |
"step": 460
|
303 |
},
|
304 |
{
|
305 |
"epoch": 2.75,
|
306 |
+
"learning_rate": 3.0119176598049837e-05,
|
307 |
+
"loss": 0.1264,
|
308 |
"step": 470
|
309 |
},
|
310 |
{
|
311 |
"epoch": 2.8,
|
312 |
+
"learning_rate": 2.9577464788732395e-05,
|
313 |
+
"loss": 0.2278,
|
314 |
"step": 480
|
315 |
},
|
316 |
{
|
317 |
"epoch": 2.86,
|
318 |
+
"learning_rate": 2.9035752979414954e-05,
|
319 |
+
"loss": 0.1669,
|
320 |
"step": 490
|
321 |
},
|
322 |
{
|
323 |
"epoch": 2.92,
|
324 |
+
"learning_rate": 2.8494041170097508e-05,
|
325 |
+
"loss": 0.2084,
|
326 |
"step": 500
|
327 |
},
|
328 |
{
|
329 |
"epoch": 2.98,
|
330 |
+
"learning_rate": 2.7952329360780066e-05,
|
331 |
+
"loss": 0.1827,
|
332 |
"step": 510
|
333 |
},
|
334 |
{
|
335 |
"epoch": 3.0,
|
336 |
+
"eval_accuracy": 0.9212827988338192,
|
337 |
+
"eval_loss": 0.24402616918087006,
|
338 |
+
"eval_runtime": 14.6992,
|
339 |
+
"eval_samples_per_second": 46.669,
|
340 |
+
"eval_steps_per_second": 5.851,
|
341 |
"step": 513
|
342 |
},
|
343 |
{
|
344 |
+
"epoch": 3.04,
|
345 |
+
"learning_rate": 2.7410617551462624e-05,
|
346 |
+
"loss": 0.1953,
|
347 |
+
"step": 520
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 3.1,
|
351 |
+
"learning_rate": 2.686890574214518e-05,
|
352 |
+
"loss": 0.1461,
|
353 |
+
"step": 530
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 3.16,
|
357 |
+
"learning_rate": 2.6327193932827737e-05,
|
358 |
+
"loss": 0.2002,
|
359 |
+
"step": 540
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 3.22,
|
363 |
+
"learning_rate": 2.5785482123510295e-05,
|
364 |
+
"loss": 0.1134,
|
365 |
+
"step": 550
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 3.27,
|
369 |
+
"learning_rate": 2.524377031419285e-05,
|
370 |
+
"loss": 0.1767,
|
371 |
+
"step": 560
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.33,
|
375 |
+
"learning_rate": 2.4702058504875408e-05,
|
376 |
+
"loss": 0.1726,
|
377 |
+
"step": 570
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.39,
|
381 |
+
"learning_rate": 2.4160346695557963e-05,
|
382 |
+
"loss": 0.1053,
|
383 |
+
"step": 580
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.45,
|
387 |
+
"learning_rate": 2.361863488624052e-05,
|
388 |
+
"loss": 0.1547,
|
389 |
+
"step": 590
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.51,
|
393 |
+
"learning_rate": 2.307692307692308e-05,
|
394 |
+
"loss": 0.1597,
|
395 |
+
"step": 600
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.56,
|
399 |
+
"learning_rate": 2.2535211267605634e-05,
|
400 |
+
"loss": 0.1319,
|
401 |
+
"step": 610
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.62,
|
405 |
+
"learning_rate": 2.1993499458288192e-05,
|
406 |
+
"loss": 0.1278,
|
407 |
+
"step": 620
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 3.68,
|
411 |
+
"learning_rate": 2.1451787648970747e-05,
|
412 |
+
"loss": 0.1113,
|
413 |
+
"step": 630
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 3.74,
|
417 |
+
"learning_rate": 2.0910075839653308e-05,
|
418 |
+
"loss": 0.1664,
|
419 |
+
"step": 640
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 3.8,
|
423 |
+
"learning_rate": 2.0368364030335863e-05,
|
424 |
+
"loss": 0.1301,
|
425 |
+
"step": 650
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 3.86,
|
429 |
+
"learning_rate": 1.9826652221018418e-05,
|
430 |
+
"loss": 0.0976,
|
431 |
+
"step": 660
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 3.91,
|
435 |
+
"learning_rate": 1.9284940411700976e-05,
|
436 |
+
"loss": 0.1927,
|
437 |
+
"step": 670
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 3.97,
|
441 |
+
"learning_rate": 1.8743228602383534e-05,
|
442 |
+
"loss": 0.1475,
|
443 |
+
"step": 680
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 4.0,
|
447 |
+
"eval_accuracy": 0.9314868804664723,
|
448 |
+
"eval_loss": 0.2305707186460495,
|
449 |
+
"eval_runtime": 40.2886,
|
450 |
+
"eval_samples_per_second": 17.027,
|
451 |
+
"eval_steps_per_second": 2.135,
|
452 |
+
"step": 684
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 4.03,
|
456 |
+
"learning_rate": 1.820151679306609e-05,
|
457 |
+
"loss": 0.1206,
|
458 |
+
"step": 690
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 4.09,
|
462 |
+
"learning_rate": 1.7659804983748647e-05,
|
463 |
+
"loss": 0.1543,
|
464 |
+
"step": 700
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 4.15,
|
468 |
+
"learning_rate": 1.7118093174431205e-05,
|
469 |
+
"loss": 0.0912,
|
470 |
+
"step": 710
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 4.21,
|
474 |
+
"learning_rate": 1.657638136511376e-05,
|
475 |
+
"loss": 0.1354,
|
476 |
+
"step": 720
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 4.27,
|
480 |
+
"learning_rate": 1.6034669555796317e-05,
|
481 |
+
"loss": 0.0883,
|
482 |
+
"step": 730
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 4.33,
|
486 |
+
"learning_rate": 1.5492957746478872e-05,
|
487 |
+
"loss": 0.1252,
|
488 |
+
"step": 740
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 4.38,
|
492 |
+
"learning_rate": 1.4951245937161432e-05,
|
493 |
+
"loss": 0.0996,
|
494 |
+
"step": 750
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 4.44,
|
498 |
+
"learning_rate": 1.4409534127843988e-05,
|
499 |
+
"loss": 0.0987,
|
500 |
+
"step": 760
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 4.5,
|
504 |
+
"learning_rate": 1.3867822318526543e-05,
|
505 |
+
"loss": 0.0881,
|
506 |
+
"step": 770
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.56,
|
510 |
+
"learning_rate": 1.3326110509209103e-05,
|
511 |
+
"loss": 0.1232,
|
512 |
+
"step": 780
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 4.62,
|
516 |
+
"learning_rate": 1.278439869989166e-05,
|
517 |
+
"loss": 0.1109,
|
518 |
+
"step": 790
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 4.68,
|
522 |
+
"learning_rate": 1.2242686890574214e-05,
|
523 |
+
"loss": 0.1524,
|
524 |
+
"step": 800
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 4.73,
|
528 |
+
"learning_rate": 1.1700975081256772e-05,
|
529 |
+
"loss": 0.0863,
|
530 |
+
"step": 810
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 4.79,
|
534 |
+
"learning_rate": 1.1159263271939329e-05,
|
535 |
+
"loss": 0.1184,
|
536 |
+
"step": 820
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 4.85,
|
540 |
+
"learning_rate": 1.0617551462621887e-05,
|
541 |
+
"loss": 0.1177,
|
542 |
+
"step": 830
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 4.91,
|
546 |
+
"learning_rate": 1.0075839653304441e-05,
|
547 |
+
"loss": 0.1187,
|
548 |
+
"step": 840
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 4.97,
|
552 |
+
"learning_rate": 9.534127843987e-06,
|
553 |
+
"loss": 0.1284,
|
554 |
+
"step": 850
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 5.0,
|
558 |
+
"eval_accuracy": 0.9358600583090378,
|
559 |
+
"eval_loss": 0.21915613114833832,
|
560 |
+
"eval_runtime": 14.6548,
|
561 |
+
"eval_samples_per_second": 46.81,
|
562 |
+
"eval_steps_per_second": 5.868,
|
563 |
+
"step": 855
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 5.03,
|
567 |
+
"learning_rate": 8.992416034669558e-06,
|
568 |
+
"loss": 0.1107,
|
569 |
+
"step": 860
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 5.09,
|
573 |
+
"learning_rate": 8.450704225352112e-06,
|
574 |
+
"loss": 0.1019,
|
575 |
+
"step": 870
|
576 |
+
},
|
577 |
+
{
|
578 |
+
"epoch": 5.15,
|
579 |
+
"learning_rate": 7.90899241603467e-06,
|
580 |
+
"loss": 0.0951,
|
581 |
+
"step": 880
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 5.2,
|
585 |
+
"learning_rate": 7.367280606717227e-06,
|
586 |
+
"loss": 0.0644,
|
587 |
+
"step": 890
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 5.26,
|
591 |
+
"learning_rate": 6.825568797399784e-06,
|
592 |
+
"loss": 0.1038,
|
593 |
+
"step": 900
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 5.32,
|
597 |
+
"learning_rate": 6.28385698808234e-06,
|
598 |
+
"loss": 0.0945,
|
599 |
+
"step": 910
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 5.38,
|
603 |
+
"learning_rate": 5.742145178764898e-06,
|
604 |
+
"loss": 0.0943,
|
605 |
+
"step": 920
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 5.44,
|
609 |
+
"learning_rate": 5.200433369447454e-06,
|
610 |
+
"loss": 0.0832,
|
611 |
+
"step": 930
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 5.49,
|
615 |
+
"learning_rate": 4.658721560130011e-06,
|
616 |
+
"loss": 0.0872,
|
617 |
+
"step": 940
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 5.55,
|
621 |
+
"learning_rate": 4.117009750812568e-06,
|
622 |
+
"loss": 0.0865,
|
623 |
+
"step": 950
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 5.61,
|
627 |
+
"learning_rate": 3.5752979414951246e-06,
|
628 |
+
"loss": 0.1009,
|
629 |
+
"step": 960
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 5.67,
|
633 |
+
"learning_rate": 3.033586132177682e-06,
|
634 |
+
"loss": 0.0913,
|
635 |
+
"step": 970
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 5.73,
|
639 |
+
"learning_rate": 2.4918743228602387e-06,
|
640 |
+
"loss": 0.0767,
|
641 |
+
"step": 980
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 5.79,
|
645 |
+
"learning_rate": 1.950162513542795e-06,
|
646 |
+
"loss": 0.129,
|
647 |
+
"step": 990
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 5.84,
|
651 |
+
"learning_rate": 1.4084507042253521e-06,
|
652 |
+
"loss": 0.049,
|
653 |
+
"step": 1000
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 5.9,
|
657 |
+
"learning_rate": 8.667388949079089e-07,
|
658 |
+
"loss": 0.062,
|
659 |
+
"step": 1010
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 5.96,
|
663 |
+
"learning_rate": 3.250270855904659e-07,
|
664 |
+
"loss": 0.0526,
|
665 |
+
"step": 1020
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 6.0,
|
669 |
+
"eval_accuracy": 0.9358600583090378,
|
670 |
+
"eval_loss": 0.2184152901172638,
|
671 |
+
"eval_runtime": 39.5726,
|
672 |
+
"eval_samples_per_second": 17.335,
|
673 |
+
"eval_steps_per_second": 2.173,
|
674 |
+
"step": 1026
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 6.0,
|
678 |
+
"step": 1026,
|
679 |
+
"total_flos": 2.549919337377528e+18,
|
680 |
+
"train_loss": 0.2920494159759834,
|
681 |
+
"train_runtime": 2402.1234,
|
682 |
+
"train_samples_per_second": 13.71,
|
683 |
+
"train_steps_per_second": 0.427
|
684 |
}
|
685 |
],
|
686 |
+
"max_steps": 1026,
|
687 |
+
"num_train_epochs": 6,
|
688 |
+
"total_flos": 2.549919337377528e+18,
|
689 |
"trial_name": null,
|
690 |
"trial_params": null
|
691 |
}
|