Jeremiah Zhou
commited on
Commit
•
af9690f
1
Parent(s):
12ede64
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- runs/Jun14_18-03-03_pikachu/events.out.tfevents.1655205385.pikachu.4152742.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +547 -0
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.0,
|
3 |
+
"eval_accuracy": 0.9245835621453414,
|
4 |
+
"eval_loss": 0.2992143929004669,
|
5 |
+
"eval_runtime": 8.9543,
|
6 |
+
"eval_samples": 5463,
|
7 |
+
"eval_samples_per_second": 610.1,
|
8 |
+
"eval_steps_per_second": 76.277,
|
9 |
+
"train_loss": 0.20157804863496975,
|
10 |
+
"train_runtime": 4031.9268,
|
11 |
+
"train_samples": 104743,
|
12 |
+
"train_samples_per_second": 259.784,
|
13 |
+
"train_steps_per_second": 16.238
|
14 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.0,
|
3 |
+
"eval_accuracy": 0.9245835621453414,
|
4 |
+
"eval_loss": 0.2992143929004669,
|
5 |
+
"eval_runtime": 8.9543,
|
6 |
+
"eval_samples": 5463,
|
7 |
+
"eval_samples_per_second": 610.1,
|
8 |
+
"eval_steps_per_second": 76.277
|
9 |
+
}
|
runs/Jun14_18-03-03_pikachu/events.out.tfevents.1655205385.pikachu.4152742.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:956c0ab127f8c00393177c68ae1a9cd78b08b8e9648f7155fc84ea474917cf13
|
3 |
+
size 369
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.0,
|
3 |
+
"train_loss": 0.20157804863496975,
|
4 |
+
"train_runtime": 4031.9268,
|
5 |
+
"train_samples": 104743,
|
6 |
+
"train_samples_per_second": 259.784,
|
7 |
+
"train_steps_per_second": 16.238
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,547 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.9245835621453414,
|
3 |
+
"best_model_checkpoint": "./fine-tune/roberta-base/qnli/checkpoint-19641",
|
4 |
+
"epoch": 6.0,
|
5 |
+
"global_step": 39282,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.08,
|
12 |
+
"learning_rate": 2.545176889793841e-06,
|
13 |
+
"loss": 0.6928,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.15,
|
18 |
+
"learning_rate": 5.090353779587682e-06,
|
19 |
+
"loss": 0.4818,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.23,
|
24 |
+
"learning_rate": 7.635530669381522e-06,
|
25 |
+
"loss": 0.3934,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.31,
|
30 |
+
"learning_rate": 1.0180707559175364e-05,
|
31 |
+
"loss": 0.379,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.38,
|
36 |
+
"learning_rate": 1.2725884448969203e-05,
|
37 |
+
"loss": 0.3509,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.46,
|
42 |
+
"learning_rate": 1.5271061338763045e-05,
|
43 |
+
"loss": 0.3652,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.53,
|
48 |
+
"learning_rate": 1.7816238228556887e-05,
|
49 |
+
"loss": 0.3552,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.61,
|
54 |
+
"learning_rate": 1.997692595180449e-05,
|
55 |
+
"loss": 0.3218,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.69,
|
60 |
+
"learning_rate": 1.9814432654652997e-05,
|
61 |
+
"loss": 0.3598,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.76,
|
66 |
+
"learning_rate": 1.9651939357501505e-05,
|
67 |
+
"loss": 0.3159,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.84,
|
72 |
+
"learning_rate": 1.9489446060350013e-05,
|
73 |
+
"loss": 0.3182,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.92,
|
78 |
+
"learning_rate": 1.932695276319852e-05,
|
79 |
+
"loss": 0.3031,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.99,
|
84 |
+
"learning_rate": 1.916445946604703e-05,
|
85 |
+
"loss": 0.2986,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.0,
|
90 |
+
"eval_accuracy": 0.9170785282811642,
|
91 |
+
"eval_loss": 0.22146184742450714,
|
92 |
+
"eval_runtime": 9.2911,
|
93 |
+
"eval_samples_per_second": 587.983,
|
94 |
+
"eval_steps_per_second": 73.511,
|
95 |
+
"step": 6547
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 1.07,
|
99 |
+
"learning_rate": 1.9001966168895533e-05,
|
100 |
+
"loss": 0.2681,
|
101 |
+
"step": 7000
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 1.15,
|
105 |
+
"learning_rate": 1.883947287174404e-05,
|
106 |
+
"loss": 0.2681,
|
107 |
+
"step": 7500
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 1.22,
|
111 |
+
"learning_rate": 1.867697957459255e-05,
|
112 |
+
"loss": 0.2643,
|
113 |
+
"step": 8000
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 1.3,
|
117 |
+
"learning_rate": 1.8514486277441056e-05,
|
118 |
+
"loss": 0.253,
|
119 |
+
"step": 8500
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 1.37,
|
123 |
+
"learning_rate": 1.8351992980289564e-05,
|
124 |
+
"loss": 0.2503,
|
125 |
+
"step": 9000
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 1.45,
|
129 |
+
"learning_rate": 1.8189499683138072e-05,
|
130 |
+
"loss": 0.2597,
|
131 |
+
"step": 9500
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 1.53,
|
135 |
+
"learning_rate": 1.802700638598658e-05,
|
136 |
+
"loss": 0.2601,
|
137 |
+
"step": 10000
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.6,
|
141 |
+
"learning_rate": 1.7864513088835088e-05,
|
142 |
+
"loss": 0.2483,
|
143 |
+
"step": 10500
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.68,
|
147 |
+
"learning_rate": 1.7702019791683592e-05,
|
148 |
+
"loss": 0.2532,
|
149 |
+
"step": 11000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.76,
|
153 |
+
"learning_rate": 1.75395264945321e-05,
|
154 |
+
"loss": 0.2455,
|
155 |
+
"step": 11500
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.83,
|
159 |
+
"learning_rate": 1.737703319738061e-05,
|
160 |
+
"loss": 0.2637,
|
161 |
+
"step": 12000
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.91,
|
165 |
+
"learning_rate": 1.7214539900229116e-05,
|
166 |
+
"loss": 0.2391,
|
167 |
+
"step": 12500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.99,
|
171 |
+
"learning_rate": 1.7052046603077624e-05,
|
172 |
+
"loss": 0.243,
|
173 |
+
"step": 13000
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 2.0,
|
177 |
+
"eval_accuracy": 0.9172615778876075,
|
178 |
+
"eval_loss": 0.23211686313152313,
|
179 |
+
"eval_runtime": 9.2969,
|
180 |
+
"eval_samples_per_second": 587.613,
|
181 |
+
"eval_steps_per_second": 73.465,
|
182 |
+
"step": 13094
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 2.06,
|
186 |
+
"learning_rate": 1.6889553305926132e-05,
|
187 |
+
"loss": 0.2067,
|
188 |
+
"step": 13500
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 2.14,
|
192 |
+
"learning_rate": 1.672706000877464e-05,
|
193 |
+
"loss": 0.206,
|
194 |
+
"step": 14000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 2.21,
|
198 |
+
"learning_rate": 1.6564566711623148e-05,
|
199 |
+
"loss": 0.1964,
|
200 |
+
"step": 14500
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.29,
|
204 |
+
"learning_rate": 1.6402073414471655e-05,
|
205 |
+
"loss": 0.1993,
|
206 |
+
"step": 15000
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.37,
|
210 |
+
"learning_rate": 1.6239580117320163e-05,
|
211 |
+
"loss": 0.2153,
|
212 |
+
"step": 15500
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.44,
|
216 |
+
"learning_rate": 1.607708682016867e-05,
|
217 |
+
"loss": 0.2103,
|
218 |
+
"step": 16000
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.52,
|
222 |
+
"learning_rate": 1.5914593523017176e-05,
|
223 |
+
"loss": 0.2023,
|
224 |
+
"step": 16500
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.6,
|
228 |
+
"learning_rate": 1.5752100225865684e-05,
|
229 |
+
"loss": 0.2063,
|
230 |
+
"step": 17000
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 2.67,
|
234 |
+
"learning_rate": 1.558960692871419e-05,
|
235 |
+
"loss": 0.2047,
|
236 |
+
"step": 17500
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 2.75,
|
240 |
+
"learning_rate": 1.54271136315627e-05,
|
241 |
+
"loss": 0.2076,
|
242 |
+
"step": 18000
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 2.83,
|
246 |
+
"learning_rate": 1.5264620334411207e-05,
|
247 |
+
"loss": 0.2086,
|
248 |
+
"step": 18500
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.9,
|
252 |
+
"learning_rate": 1.5102127037259715e-05,
|
253 |
+
"loss": 0.2001,
|
254 |
+
"step": 19000
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 2.98,
|
258 |
+
"learning_rate": 1.4939633740108221e-05,
|
259 |
+
"loss": 0.2048,
|
260 |
+
"step": 19500
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 3.0,
|
264 |
+
"eval_accuracy": 0.9245835621453414,
|
265 |
+
"eval_loss": 0.2992143929004669,
|
266 |
+
"eval_runtime": 9.1061,
|
267 |
+
"eval_samples_per_second": 599.927,
|
268 |
+
"eval_steps_per_second": 75.005,
|
269 |
+
"step": 19641
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 3.05,
|
273 |
+
"learning_rate": 1.477714044295673e-05,
|
274 |
+
"loss": 0.1717,
|
275 |
+
"step": 20000
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 3.13,
|
279 |
+
"learning_rate": 1.4614647145805237e-05,
|
280 |
+
"loss": 0.1483,
|
281 |
+
"step": 20500
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 3.21,
|
285 |
+
"learning_rate": 1.4452153848653745e-05,
|
286 |
+
"loss": 0.1743,
|
287 |
+
"step": 21000
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 3.28,
|
291 |
+
"learning_rate": 1.4289660551502251e-05,
|
292 |
+
"loss": 0.1442,
|
293 |
+
"step": 21500
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 3.36,
|
297 |
+
"learning_rate": 1.412716725435076e-05,
|
298 |
+
"loss": 0.1744,
|
299 |
+
"step": 22000
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 3.44,
|
303 |
+
"learning_rate": 1.3964673957199267e-05,
|
304 |
+
"loss": 0.1694,
|
305 |
+
"step": 22500
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.51,
|
309 |
+
"learning_rate": 1.3802180660047775e-05,
|
310 |
+
"loss": 0.1669,
|
311 |
+
"step": 23000
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.59,
|
315 |
+
"learning_rate": 1.3639687362896281e-05,
|
316 |
+
"loss": 0.1539,
|
317 |
+
"step": 23500
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 3.67,
|
321 |
+
"learning_rate": 1.347719406574479e-05,
|
322 |
+
"loss": 0.1601,
|
323 |
+
"step": 24000
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 3.74,
|
327 |
+
"learning_rate": 1.3314700768593297e-05,
|
328 |
+
"loss": 0.1689,
|
329 |
+
"step": 24500
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 3.82,
|
333 |
+
"learning_rate": 1.3152207471441804e-05,
|
334 |
+
"loss": 0.1608,
|
335 |
+
"step": 25000
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 3.89,
|
339 |
+
"learning_rate": 1.298971417429031e-05,
|
340 |
+
"loss": 0.1639,
|
341 |
+
"step": 25500
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 3.97,
|
345 |
+
"learning_rate": 1.282722087713882e-05,
|
346 |
+
"loss": 0.1629,
|
347 |
+
"step": 26000
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 4.0,
|
351 |
+
"eval_accuracy": 0.9220208676551346,
|
352 |
+
"eval_loss": 0.3538360595703125,
|
353 |
+
"eval_runtime": 9.1087,
|
354 |
+
"eval_samples_per_second": 599.755,
|
355 |
+
"eval_steps_per_second": 74.983,
|
356 |
+
"step": 26188
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 4.05,
|
360 |
+
"learning_rate": 1.2664727579987326e-05,
|
361 |
+
"loss": 0.1307,
|
362 |
+
"step": 26500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 4.12,
|
366 |
+
"learning_rate": 1.2502234282835834e-05,
|
367 |
+
"loss": 0.1112,
|
368 |
+
"step": 27000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 4.2,
|
372 |
+
"learning_rate": 1.233974098568434e-05,
|
373 |
+
"loss": 0.1243,
|
374 |
+
"step": 27500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 4.28,
|
378 |
+
"learning_rate": 1.217724768853285e-05,
|
379 |
+
"loss": 0.1111,
|
380 |
+
"step": 28000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 4.35,
|
384 |
+
"learning_rate": 1.2014754391381356e-05,
|
385 |
+
"loss": 0.1065,
|
386 |
+
"step": 28500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 4.43,
|
390 |
+
"learning_rate": 1.1852261094229864e-05,
|
391 |
+
"loss": 0.1319,
|
392 |
+
"step": 29000
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 4.51,
|
396 |
+
"learning_rate": 1.168976779707837e-05,
|
397 |
+
"loss": 0.1172,
|
398 |
+
"step": 29500
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 4.58,
|
402 |
+
"learning_rate": 1.152727449992688e-05,
|
403 |
+
"loss": 0.1356,
|
404 |
+
"step": 30000
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 4.66,
|
408 |
+
"learning_rate": 1.1364781202775386e-05,
|
409 |
+
"loss": 0.1262,
|
410 |
+
"step": 30500
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 4.73,
|
414 |
+
"learning_rate": 1.1202287905623894e-05,
|
415 |
+
"loss": 0.1236,
|
416 |
+
"step": 31000
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 4.81,
|
420 |
+
"learning_rate": 1.10397946084724e-05,
|
421 |
+
"loss": 0.1276,
|
422 |
+
"step": 31500
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 4.89,
|
426 |
+
"learning_rate": 1.087730131132091e-05,
|
427 |
+
"loss": 0.126,
|
428 |
+
"step": 32000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 4.96,
|
432 |
+
"learning_rate": 1.0714808014169416e-05,
|
433 |
+
"loss": 0.1308,
|
434 |
+
"step": 32500
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 5.0,
|
438 |
+
"eval_accuracy": 0.9209225700164745,
|
439 |
+
"eval_loss": 0.35333874821662903,
|
440 |
+
"eval_runtime": 9.1134,
|
441 |
+
"eval_samples_per_second": 599.446,
|
442 |
+
"eval_steps_per_second": 74.944,
|
443 |
+
"step": 32735
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 5.04,
|
447 |
+
"learning_rate": 1.0552314717017924e-05,
|
448 |
+
"loss": 0.1111,
|
449 |
+
"step": 33000
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 5.12,
|
453 |
+
"learning_rate": 1.038982141986643e-05,
|
454 |
+
"loss": 0.0724,
|
455 |
+
"step": 33500
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 5.19,
|
459 |
+
"learning_rate": 1.022732812271494e-05,
|
460 |
+
"loss": 0.0885,
|
461 |
+
"step": 34000
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 5.27,
|
465 |
+
"learning_rate": 1.0064834825563446e-05,
|
466 |
+
"loss": 0.0855,
|
467 |
+
"step": 34500
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 5.35,
|
471 |
+
"learning_rate": 9.902341528411954e-06,
|
472 |
+
"loss": 0.0851,
|
473 |
+
"step": 35000
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 5.42,
|
477 |
+
"learning_rate": 9.739848231260461e-06,
|
478 |
+
"loss": 0.0852,
|
479 |
+
"step": 35500
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 5.5,
|
483 |
+
"learning_rate": 9.57735493410897e-06,
|
484 |
+
"loss": 0.0888,
|
485 |
+
"step": 36000
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 5.58,
|
489 |
+
"learning_rate": 9.414861636957477e-06,
|
490 |
+
"loss": 0.0893,
|
491 |
+
"step": 36500
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 5.65,
|
495 |
+
"learning_rate": 9.252368339805983e-06,
|
496 |
+
"loss": 0.0865,
|
497 |
+
"step": 37000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 5.73,
|
501 |
+
"learning_rate": 9.089875042654491e-06,
|
502 |
+
"loss": 0.0814,
|
503 |
+
"step": 37500
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 5.8,
|
507 |
+
"learning_rate": 8.927381745502999e-06,
|
508 |
+
"loss": 0.0967,
|
509 |
+
"step": 38000
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 5.88,
|
513 |
+
"learning_rate": 8.764888448351507e-06,
|
514 |
+
"loss": 0.0882,
|
515 |
+
"step": 38500
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 5.96,
|
519 |
+
"learning_rate": 8.602395151200013e-06,
|
520 |
+
"loss": 0.0846,
|
521 |
+
"step": 39000
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 6.0,
|
525 |
+
"eval_accuracy": 0.9229361156873512,
|
526 |
+
"eval_loss": 0.427664577960968,
|
527 |
+
"eval_runtime": 9.0686,
|
528 |
+
"eval_samples_per_second": 602.409,
|
529 |
+
"eval_steps_per_second": 75.315,
|
530 |
+
"step": 39282
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 6.0,
|
534 |
+
"step": 39282,
|
535 |
+
"total_flos": 4.133856190735872e+16,
|
536 |
+
"train_loss": 0.20157804863496975,
|
537 |
+
"train_runtime": 4031.9268,
|
538 |
+
"train_samples_per_second": 259.784,
|
539 |
+
"train_steps_per_second": 16.238
|
540 |
+
}
|
541 |
+
],
|
542 |
+
"max_steps": 65470,
|
543 |
+
"num_train_epochs": 10,
|
544 |
+
"total_flos": 4.133856190735872e+16,
|
545 |
+
"trial_name": null,
|
546 |
+
"trial_params": null
|
547 |
+
}
|