Upload trainer_state.json
Browse files- trainer_state.json +700 -0
trainer_state.json
ADDED
@@ -0,0 +1,700 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.645846426486969,
|
3 |
+
"best_model_checkpoint": "models/checkpoints/checkpoint-9000",
|
4 |
+
"epoch": 16.2748643761302,
|
5 |
+
"global_step": 9000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.18,
|
12 |
+
"learning_rate": 1.9819168173598556e-05,
|
13 |
+
"loss": 1.9232,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.36,
|
18 |
+
"learning_rate": 1.9638336347197107e-05,
|
19 |
+
"loss": 1.3768,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.54,
|
24 |
+
"learning_rate": 1.9457504520795662e-05,
|
25 |
+
"loss": 1.269,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.72,
|
30 |
+
"learning_rate": 1.9276672694394213e-05,
|
31 |
+
"loss": 1.2114,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.9,
|
36 |
+
"learning_rate": 1.9095840867992768e-05,
|
37 |
+
"loss": 1.1765,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.9,
|
42 |
+
"eval_loss": 1.0532047748565674,
|
43 |
+
"eval_runtime": 12.8374,
|
44 |
+
"eval_samples_per_second": 612.197,
|
45 |
+
"eval_steps_per_second": 4.83,
|
46 |
+
"step": 500
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.08,
|
50 |
+
"learning_rate": 1.8915009041591322e-05,
|
51 |
+
"loss": 1.1195,
|
52 |
+
"step": 600
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.27,
|
56 |
+
"learning_rate": 1.8734177215189874e-05,
|
57 |
+
"loss": 1.1068,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 1.45,
|
62 |
+
"learning_rate": 1.8553345388788428e-05,
|
63 |
+
"loss": 1.0338,
|
64 |
+
"step": 800
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 1.63,
|
68 |
+
"learning_rate": 1.8372513562386983e-05,
|
69 |
+
"loss": 1.0417,
|
70 |
+
"step": 900
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"epoch": 1.81,
|
74 |
+
"learning_rate": 1.8191681735985537e-05,
|
75 |
+
"loss": 1.0382,
|
76 |
+
"step": 1000
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 1.81,
|
80 |
+
"eval_loss": 0.964276909828186,
|
81 |
+
"eval_runtime": 12.8331,
|
82 |
+
"eval_samples_per_second": 612.399,
|
83 |
+
"eval_steps_per_second": 4.831,
|
84 |
+
"step": 1000
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 1.99,
|
88 |
+
"learning_rate": 1.801084990958409e-05,
|
89 |
+
"loss": 1.0084,
|
90 |
+
"step": 1100
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 2.17,
|
94 |
+
"learning_rate": 1.783001808318264e-05,
|
95 |
+
"loss": 0.9976,
|
96 |
+
"step": 1200
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 2.35,
|
100 |
+
"learning_rate": 1.7649186256781194e-05,
|
101 |
+
"loss": 0.9962,
|
102 |
+
"step": 1300
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 2.53,
|
106 |
+
"learning_rate": 1.746835443037975e-05,
|
107 |
+
"loss": 0.953,
|
108 |
+
"step": 1400
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 2.71,
|
112 |
+
"learning_rate": 1.72875226039783e-05,
|
113 |
+
"loss": 0.9687,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 2.71,
|
118 |
+
"eval_loss": 0.9130759239196777,
|
119 |
+
"eval_runtime": 12.8023,
|
120 |
+
"eval_samples_per_second": 613.873,
|
121 |
+
"eval_steps_per_second": 4.843,
|
122 |
+
"step": 1500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 2.89,
|
126 |
+
"learning_rate": 1.7106690777576855e-05,
|
127 |
+
"loss": 0.9551,
|
128 |
+
"step": 1600
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 3.07,
|
132 |
+
"learning_rate": 1.692585895117541e-05,
|
133 |
+
"loss": 0.9156,
|
134 |
+
"step": 1700
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 3.25,
|
138 |
+
"learning_rate": 1.6745027124773964e-05,
|
139 |
+
"loss": 0.9072,
|
140 |
+
"step": 1800
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 3.44,
|
144 |
+
"learning_rate": 1.6564195298372515e-05,
|
145 |
+
"loss": 0.9038,
|
146 |
+
"step": 1900
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 3.62,
|
150 |
+
"learning_rate": 1.6383363471971066e-05,
|
151 |
+
"loss": 0.9157,
|
152 |
+
"step": 2000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 3.62,
|
156 |
+
"eval_loss": 0.8786566853523254,
|
157 |
+
"eval_runtime": 12.8067,
|
158 |
+
"eval_samples_per_second": 613.662,
|
159 |
+
"eval_steps_per_second": 4.841,
|
160 |
+
"step": 2000
|
161 |
+
},
|
162 |
+
{
|
163 |
+
"epoch": 3.8,
|
164 |
+
"learning_rate": 1.620253164556962e-05,
|
165 |
+
"loss": 0.8994,
|
166 |
+
"step": 2100
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 3.98,
|
170 |
+
"learning_rate": 1.6021699819168176e-05,
|
171 |
+
"loss": 0.8963,
|
172 |
+
"step": 2200
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 4.16,
|
176 |
+
"learning_rate": 1.584086799276673e-05,
|
177 |
+
"loss": 0.8878,
|
178 |
+
"step": 2300
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 4.34,
|
182 |
+
"learning_rate": 1.566003616636528e-05,
|
183 |
+
"loss": 0.8739,
|
184 |
+
"step": 2400
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 4.52,
|
188 |
+
"learning_rate": 1.5479204339963836e-05,
|
189 |
+
"loss": 0.8671,
|
190 |
+
"step": 2500
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 4.52,
|
194 |
+
"eval_loss": 0.8628956079483032,
|
195 |
+
"eval_runtime": 12.8156,
|
196 |
+
"eval_samples_per_second": 613.239,
|
197 |
+
"eval_steps_per_second": 4.838,
|
198 |
+
"step": 2500
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 4.7,
|
202 |
+
"learning_rate": 1.5298372513562387e-05,
|
203 |
+
"loss": 0.8676,
|
204 |
+
"step": 2600
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 4.88,
|
208 |
+
"learning_rate": 1.5117540687160942e-05,
|
209 |
+
"loss": 0.8329,
|
210 |
+
"step": 2700
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 5.06,
|
214 |
+
"learning_rate": 1.4936708860759495e-05,
|
215 |
+
"loss": 0.8438,
|
216 |
+
"step": 2800
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 5.24,
|
220 |
+
"learning_rate": 1.4755877034358048e-05,
|
221 |
+
"loss": 0.8294,
|
222 |
+
"step": 2900
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 5.42,
|
226 |
+
"learning_rate": 1.4575045207956602e-05,
|
227 |
+
"loss": 0.8259,
|
228 |
+
"step": 3000
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 5.42,
|
232 |
+
"eval_loss": 0.825766921043396,
|
233 |
+
"eval_runtime": 12.7989,
|
234 |
+
"eval_samples_per_second": 614.036,
|
235 |
+
"eval_steps_per_second": 4.844,
|
236 |
+
"step": 3000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 5.61,
|
240 |
+
"learning_rate": 1.4394213381555155e-05,
|
241 |
+
"loss": 0.8065,
|
242 |
+
"step": 3100
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 5.79,
|
246 |
+
"learning_rate": 1.421338155515371e-05,
|
247 |
+
"loss": 0.8193,
|
248 |
+
"step": 3200
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 5.97,
|
252 |
+
"learning_rate": 1.403254972875226e-05,
|
253 |
+
"loss": 0.8152,
|
254 |
+
"step": 3300
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 6.15,
|
258 |
+
"learning_rate": 1.3851717902350814e-05,
|
259 |
+
"loss": 0.784,
|
260 |
+
"step": 3400
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 6.33,
|
264 |
+
"learning_rate": 1.3670886075949368e-05,
|
265 |
+
"loss": 0.7941,
|
266 |
+
"step": 3500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 6.33,
|
270 |
+
"eval_loss": 0.8114328384399414,
|
271 |
+
"eval_runtime": 12.8036,
|
272 |
+
"eval_samples_per_second": 613.81,
|
273 |
+
"eval_steps_per_second": 4.842,
|
274 |
+
"step": 3500
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 6.51,
|
278 |
+
"learning_rate": 1.3490054249547921e-05,
|
279 |
+
"loss": 0.7952,
|
280 |
+
"step": 3600
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 6.69,
|
284 |
+
"learning_rate": 1.3309222423146476e-05,
|
285 |
+
"loss": 0.8168,
|
286 |
+
"step": 3700
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 6.87,
|
290 |
+
"learning_rate": 1.3128390596745029e-05,
|
291 |
+
"loss": 0.7917,
|
292 |
+
"step": 3800
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 7.05,
|
296 |
+
"learning_rate": 1.2947558770343582e-05,
|
297 |
+
"loss": 0.7582,
|
298 |
+
"step": 3900
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 7.23,
|
302 |
+
"learning_rate": 1.2766726943942136e-05,
|
303 |
+
"loss": 0.7605,
|
304 |
+
"step": 4000
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 7.23,
|
308 |
+
"eval_loss": 0.7806207537651062,
|
309 |
+
"eval_runtime": 12.8016,
|
310 |
+
"eval_samples_per_second": 613.906,
|
311 |
+
"eval_steps_per_second": 4.843,
|
312 |
+
"step": 4000
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 7.41,
|
316 |
+
"learning_rate": 1.2585895117540687e-05,
|
317 |
+
"loss": 0.7651,
|
318 |
+
"step": 4100
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 7.59,
|
322 |
+
"learning_rate": 1.240506329113924e-05,
|
323 |
+
"loss": 0.7716,
|
324 |
+
"step": 4200
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 7.78,
|
328 |
+
"learning_rate": 1.2224231464737795e-05,
|
329 |
+
"loss": 0.7704,
|
330 |
+
"step": 4300
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 7.96,
|
334 |
+
"learning_rate": 1.2043399638336348e-05,
|
335 |
+
"loss": 0.7483,
|
336 |
+
"step": 4400
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 8.14,
|
340 |
+
"learning_rate": 1.1862567811934902e-05,
|
341 |
+
"loss": 0.7323,
|
342 |
+
"step": 4500
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 8.14,
|
346 |
+
"eval_loss": 0.7683702707290649,
|
347 |
+
"eval_runtime": 12.8031,
|
348 |
+
"eval_samples_per_second": 613.836,
|
349 |
+
"eval_steps_per_second": 4.843,
|
350 |
+
"step": 4500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 8.32,
|
354 |
+
"learning_rate": 1.1681735985533455e-05,
|
355 |
+
"loss": 0.7432,
|
356 |
+
"step": 4600
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 8.5,
|
360 |
+
"learning_rate": 1.150090415913201e-05,
|
361 |
+
"loss": 0.7486,
|
362 |
+
"step": 4700
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 8.68,
|
366 |
+
"learning_rate": 1.1320072332730561e-05,
|
367 |
+
"loss": 0.7479,
|
368 |
+
"step": 4800
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 8.86,
|
372 |
+
"learning_rate": 1.1139240506329114e-05,
|
373 |
+
"loss": 0.7192,
|
374 |
+
"step": 4900
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 9.04,
|
378 |
+
"learning_rate": 1.0958408679927669e-05,
|
379 |
+
"loss": 0.7292,
|
380 |
+
"step": 5000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 9.04,
|
384 |
+
"eval_loss": 0.7645160555839539,
|
385 |
+
"eval_runtime": 12.802,
|
386 |
+
"eval_samples_per_second": 613.888,
|
387 |
+
"eval_steps_per_second": 4.843,
|
388 |
+
"step": 5000
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 9.22,
|
392 |
+
"learning_rate": 1.0777576853526221e-05,
|
393 |
+
"loss": 0.7055,
|
394 |
+
"step": 5100
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 9.4,
|
398 |
+
"learning_rate": 1.0596745027124774e-05,
|
399 |
+
"loss": 0.7094,
|
400 |
+
"step": 5200
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 9.58,
|
404 |
+
"learning_rate": 1.0415913200723329e-05,
|
405 |
+
"loss": 0.7297,
|
406 |
+
"step": 5300
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 9.76,
|
410 |
+
"learning_rate": 1.0235081374321882e-05,
|
411 |
+
"loss": 0.7096,
|
412 |
+
"step": 5400
|
413 |
+
},
|
414 |
+
{
|
415 |
+
"epoch": 9.95,
|
416 |
+
"learning_rate": 1.0054249547920433e-05,
|
417 |
+
"loss": 0.7259,
|
418 |
+
"step": 5500
|
419 |
+
},
|
420 |
+
{
|
421 |
+
"epoch": 9.95,
|
422 |
+
"eval_loss": 0.7333822846412659,
|
423 |
+
"eval_runtime": 12.806,
|
424 |
+
"eval_samples_per_second": 613.699,
|
425 |
+
"eval_steps_per_second": 4.841,
|
426 |
+
"step": 5500
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 10.13,
|
430 |
+
"learning_rate": 9.87341772151899e-06,
|
431 |
+
"loss": 0.7138,
|
432 |
+
"step": 5600
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 10.31,
|
436 |
+
"learning_rate": 9.69258589511754e-06,
|
437 |
+
"loss": 0.7118,
|
438 |
+
"step": 5700
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 10.49,
|
442 |
+
"learning_rate": 9.511754068716095e-06,
|
443 |
+
"loss": 0.7091,
|
444 |
+
"step": 5800
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 10.67,
|
448 |
+
"learning_rate": 9.330922242314648e-06,
|
449 |
+
"loss": 0.6839,
|
450 |
+
"step": 5900
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 10.85,
|
454 |
+
"learning_rate": 9.150090415913203e-06,
|
455 |
+
"loss": 0.7045,
|
456 |
+
"step": 6000
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 10.85,
|
460 |
+
"eval_loss": 0.7454974055290222,
|
461 |
+
"eval_runtime": 12.7967,
|
462 |
+
"eval_samples_per_second": 614.144,
|
463 |
+
"eval_steps_per_second": 4.845,
|
464 |
+
"step": 6000
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 11.03,
|
468 |
+
"learning_rate": 8.969258589511754e-06,
|
469 |
+
"loss": 0.7059,
|
470 |
+
"step": 6100
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 11.21,
|
474 |
+
"learning_rate": 8.788426763110308e-06,
|
475 |
+
"loss": 0.6759,
|
476 |
+
"step": 6200
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 11.39,
|
480 |
+
"learning_rate": 8.607594936708861e-06,
|
481 |
+
"loss": 0.6806,
|
482 |
+
"step": 6300
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 11.57,
|
486 |
+
"learning_rate": 8.426763110307414e-06,
|
487 |
+
"loss": 0.6734,
|
488 |
+
"step": 6400
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 11.75,
|
492 |
+
"learning_rate": 8.245931283905967e-06,
|
493 |
+
"loss": 0.677,
|
494 |
+
"step": 6500
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 11.75,
|
498 |
+
"eval_loss": 0.7277078032493591,
|
499 |
+
"eval_runtime": 12.8036,
|
500 |
+
"eval_samples_per_second": 613.809,
|
501 |
+
"eval_steps_per_second": 4.842,
|
502 |
+
"step": 6500
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 11.93,
|
506 |
+
"learning_rate": 8.065099457504522e-06,
|
507 |
+
"loss": 0.6811,
|
508 |
+
"step": 6600
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 12.12,
|
512 |
+
"learning_rate": 7.884267631103075e-06,
|
513 |
+
"loss": 0.669,
|
514 |
+
"step": 6700
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 12.3,
|
518 |
+
"learning_rate": 7.703435804701628e-06,
|
519 |
+
"loss": 0.6586,
|
520 |
+
"step": 6800
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 12.48,
|
524 |
+
"learning_rate": 7.522603978300181e-06,
|
525 |
+
"loss": 0.6428,
|
526 |
+
"step": 6900
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 12.66,
|
530 |
+
"learning_rate": 7.341772151898735e-06,
|
531 |
+
"loss": 0.6618,
|
532 |
+
"step": 7000
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 12.66,
|
536 |
+
"eval_loss": 0.7127390503883362,
|
537 |
+
"eval_runtime": 12.8054,
|
538 |
+
"eval_samples_per_second": 613.727,
|
539 |
+
"eval_steps_per_second": 4.842,
|
540 |
+
"step": 7000
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 12.84,
|
544 |
+
"learning_rate": 7.160940325497288e-06,
|
545 |
+
"loss": 0.6703,
|
546 |
+
"step": 7100
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 13.02,
|
550 |
+
"learning_rate": 6.980108499095841e-06,
|
551 |
+
"loss": 0.6577,
|
552 |
+
"step": 7200
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 13.2,
|
556 |
+
"learning_rate": 6.799276672694395e-06,
|
557 |
+
"loss": 0.6431,
|
558 |
+
"step": 7300
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 13.38,
|
562 |
+
"learning_rate": 6.618444846292948e-06,
|
563 |
+
"loss": 0.6525,
|
564 |
+
"step": 7400
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 13.56,
|
568 |
+
"learning_rate": 6.437613019891501e-06,
|
569 |
+
"loss": 0.6817,
|
570 |
+
"step": 7500
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 13.56,
|
574 |
+
"eval_loss": 0.7023378014564514,
|
575 |
+
"eval_runtime": 12.8161,
|
576 |
+
"eval_samples_per_second": 613.215,
|
577 |
+
"eval_steps_per_second": 4.838,
|
578 |
+
"step": 7500
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 13.74,
|
582 |
+
"learning_rate": 6.256781193490055e-06,
|
583 |
+
"loss": 0.6523,
|
584 |
+
"step": 7600
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 13.92,
|
588 |
+
"learning_rate": 6.075949367088608e-06,
|
589 |
+
"loss": 0.6679,
|
590 |
+
"step": 7700
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 14.1,
|
594 |
+
"learning_rate": 5.895117540687162e-06,
|
595 |
+
"loss": 0.6368,
|
596 |
+
"step": 7800
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 14.29,
|
600 |
+
"learning_rate": 5.7142857142857145e-06,
|
601 |
+
"loss": 0.641,
|
602 |
+
"step": 7900
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 14.47,
|
606 |
+
"learning_rate": 5.533453887884268e-06,
|
607 |
+
"loss": 0.6337,
|
608 |
+
"step": 8000
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 14.47,
|
612 |
+
"eval_loss": 0.7140629887580872,
|
613 |
+
"eval_runtime": 12.8021,
|
614 |
+
"eval_samples_per_second": 613.882,
|
615 |
+
"eval_steps_per_second": 4.843,
|
616 |
+
"step": 8000
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 14.65,
|
620 |
+
"learning_rate": 5.352622061482822e-06,
|
621 |
+
"loss": 0.6359,
|
622 |
+
"step": 8100
|
623 |
+
},
|
624 |
+
{
|
625 |
+
"epoch": 14.83,
|
626 |
+
"learning_rate": 5.171790235081374e-06,
|
627 |
+
"loss": 0.6298,
|
628 |
+
"step": 8200
|
629 |
+
},
|
630 |
+
{
|
631 |
+
"epoch": 15.01,
|
632 |
+
"learning_rate": 4.990958408679928e-06,
|
633 |
+
"loss": 0.6444,
|
634 |
+
"step": 8300
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"epoch": 15.19,
|
638 |
+
"learning_rate": 4.8101265822784815e-06,
|
639 |
+
"loss": 0.639,
|
640 |
+
"step": 8400
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 15.37,
|
644 |
+
"learning_rate": 4.6292947558770344e-06,
|
645 |
+
"loss": 0.6182,
|
646 |
+
"step": 8500
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 15.37,
|
650 |
+
"eval_loss": 0.7226015329360962,
|
651 |
+
"eval_runtime": 12.7692,
|
652 |
+
"eval_samples_per_second": 615.466,
|
653 |
+
"eval_steps_per_second": 4.855,
|
654 |
+
"step": 8500
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 15.55,
|
658 |
+
"learning_rate": 4.448462929475588e-06,
|
659 |
+
"loss": 0.6425,
|
660 |
+
"step": 8600
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 15.73,
|
664 |
+
"learning_rate": 4.267631103074141e-06,
|
665 |
+
"loss": 0.6255,
|
666 |
+
"step": 8700
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 15.91,
|
670 |
+
"learning_rate": 4.086799276672695e-06,
|
671 |
+
"loss": 0.6223,
|
672 |
+
"step": 8800
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 16.09,
|
676 |
+
"learning_rate": 3.905967450271248e-06,
|
677 |
+
"loss": 0.6303,
|
678 |
+
"step": 8900
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 16.27,
|
682 |
+
"learning_rate": 3.7251356238698015e-06,
|
683 |
+
"loss": 0.6367,
|
684 |
+
"step": 9000
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 16.27,
|
688 |
+
"eval_loss": 0.645846426486969,
|
689 |
+
"eval_runtime": 12.785,
|
690 |
+
"eval_samples_per_second": 614.705,
|
691 |
+
"eval_steps_per_second": 4.849,
|
692 |
+
"step": 9000
|
693 |
+
}
|
694 |
+
],
|
695 |
+
"max_steps": 11060,
|
696 |
+
"num_train_epochs": 20,
|
697 |
+
"total_flos": 5.947714605600499e+16,
|
698 |
+
"trial_name": null,
|
699 |
+
"trial_params": null
|
700 |
+
}
|