Commit
·
fcee3a4
1
Parent(s):
73f2676
Upload trainer_state.json
Browse files- trainer_state.json +428 -0
trainer_state.json
ADDED
@@ -0,0 +1,428 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9999880000239999,
|
5 |
+
"global_step": 33333,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 4.9249992499925e-05,
|
13 |
+
"loss": 1.253,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.03,
|
18 |
+
"learning_rate": 4.849998499985e-05,
|
19 |
+
"loss": 1.2489,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.04,
|
24 |
+
"learning_rate": 4.7749977499775e-05,
|
25 |
+
"loss": 1.2349,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.06,
|
30 |
+
"learning_rate": 4.69999699997e-05,
|
31 |
+
"loss": 1.2296,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.07,
|
36 |
+
"learning_rate": 4.6249962499625e-05,
|
37 |
+
"loss": 1.2276,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.09,
|
42 |
+
"learning_rate": 4.549995499955e-05,
|
43 |
+
"loss": 1.2195,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.1,
|
48 |
+
"learning_rate": 4.4749947499475e-05,
|
49 |
+
"loss": 1.2111,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.12,
|
54 |
+
"learning_rate": 4.39999399994e-05,
|
55 |
+
"loss": 1.2111,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.13,
|
60 |
+
"learning_rate": 4.3249932499324996e-05,
|
61 |
+
"loss": 1.21,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.15,
|
66 |
+
"learning_rate": 4.2499924999249994e-05,
|
67 |
+
"loss": 1.2027,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.16,
|
72 |
+
"learning_rate": 4.174991749917499e-05,
|
73 |
+
"loss": 1.2024,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.18,
|
78 |
+
"learning_rate": 4.09999099991e-05,
|
79 |
+
"loss": 1.2017,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.19,
|
84 |
+
"learning_rate": 4.0249902499024996e-05,
|
85 |
+
"loss": 1.2009,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.21,
|
90 |
+
"learning_rate": 3.9499894998949995e-05,
|
91 |
+
"loss": 1.192,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.22,
|
96 |
+
"learning_rate": 3.8749887498874993e-05,
|
97 |
+
"loss": 1.1969,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.24,
|
102 |
+
"learning_rate": 3.799987999879999e-05,
|
103 |
+
"loss": 1.1949,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.25,
|
108 |
+
"learning_rate": 3.724987249872499e-05,
|
109 |
+
"loss": 1.1974,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.27,
|
114 |
+
"learning_rate": 3.649986499864999e-05,
|
115 |
+
"loss": 1.187,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.28,
|
120 |
+
"learning_rate": 3.574985749857499e-05,
|
121 |
+
"loss": 1.1831,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.3,
|
126 |
+
"learning_rate": 3.4999849998499986e-05,
|
127 |
+
"loss": 1.183,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.31,
|
132 |
+
"learning_rate": 3.4249842498424984e-05,
|
133 |
+
"loss": 1.1819,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.33,
|
138 |
+
"learning_rate": 3.349983499834998e-05,
|
139 |
+
"loss": 1.1824,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.34,
|
144 |
+
"learning_rate": 3.274982749827498e-05,
|
145 |
+
"loss": 1.1832,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.36,
|
150 |
+
"learning_rate": 3.199981999819998e-05,
|
151 |
+
"loss": 1.1781,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.37,
|
156 |
+
"learning_rate": 3.124981249812498e-05,
|
157 |
+
"loss": 1.1775,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.39,
|
162 |
+
"learning_rate": 3.049980499804998e-05,
|
163 |
+
"loss": 1.1699,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.4,
|
168 |
+
"learning_rate": 2.974979749797498e-05,
|
169 |
+
"loss": 1.1687,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.42,
|
174 |
+
"learning_rate": 2.8999789997899977e-05,
|
175 |
+
"loss": 1.1653,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.43,
|
180 |
+
"learning_rate": 2.824978249782498e-05,
|
181 |
+
"loss": 1.1712,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.45,
|
186 |
+
"learning_rate": 2.7499774997749978e-05,
|
187 |
+
"loss": 1.1671,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.46,
|
192 |
+
"learning_rate": 2.6749767497674976e-05,
|
193 |
+
"loss": 1.1634,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.48,
|
198 |
+
"learning_rate": 2.5999759997599975e-05,
|
199 |
+
"loss": 1.1667,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.49,
|
204 |
+
"learning_rate": 2.5249752497524977e-05,
|
205 |
+
"loss": 1.159,
|
206 |
+
"step": 16500
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.51,
|
210 |
+
"learning_rate": 2.4499744997449975e-05,
|
211 |
+
"loss": 1.1569,
|
212 |
+
"step": 17000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.52,
|
216 |
+
"learning_rate": 2.3749737497374974e-05,
|
217 |
+
"loss": 1.1556,
|
218 |
+
"step": 17500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.54,
|
222 |
+
"learning_rate": 2.2999729997299972e-05,
|
223 |
+
"loss": 1.1603,
|
224 |
+
"step": 18000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.55,
|
228 |
+
"learning_rate": 2.2249722497224974e-05,
|
229 |
+
"loss": 1.1533,
|
230 |
+
"step": 18500
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.57,
|
234 |
+
"learning_rate": 2.1499714997149973e-05,
|
235 |
+
"loss": 1.1536,
|
236 |
+
"step": 19000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.58,
|
240 |
+
"learning_rate": 2.074970749707497e-05,
|
241 |
+
"loss": 1.1563,
|
242 |
+
"step": 19500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.6,
|
246 |
+
"learning_rate": 1.999969999699997e-05,
|
247 |
+
"loss": 1.1494,
|
248 |
+
"step": 20000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.61,
|
252 |
+
"learning_rate": 1.924969249692497e-05,
|
253 |
+
"loss": 1.1511,
|
254 |
+
"step": 20500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.63,
|
258 |
+
"learning_rate": 1.849968499684997e-05,
|
259 |
+
"loss": 1.1432,
|
260 |
+
"step": 21000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.64,
|
264 |
+
"learning_rate": 1.774967749677497e-05,
|
265 |
+
"loss": 1.1525,
|
266 |
+
"step": 21500
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.66,
|
270 |
+
"learning_rate": 1.6999669996699967e-05,
|
271 |
+
"loss": 1.1437,
|
272 |
+
"step": 22000
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.67,
|
276 |
+
"learning_rate": 1.624966249662497e-05,
|
277 |
+
"loss": 1.1452,
|
278 |
+
"step": 22500
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.69,
|
282 |
+
"learning_rate": 1.5499654996549967e-05,
|
283 |
+
"loss": 1.1411,
|
284 |
+
"step": 23000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.7,
|
288 |
+
"learning_rate": 1.4749647496474966e-05,
|
289 |
+
"loss": 1.1381,
|
290 |
+
"step": 23500
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.72,
|
294 |
+
"learning_rate": 1.3999639996399966e-05,
|
295 |
+
"loss": 1.1412,
|
296 |
+
"step": 24000
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.73,
|
300 |
+
"learning_rate": 1.3249632496324964e-05,
|
301 |
+
"loss": 1.1405,
|
302 |
+
"step": 24500
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.75,
|
306 |
+
"learning_rate": 1.2499624996249963e-05,
|
307 |
+
"loss": 1.1418,
|
308 |
+
"step": 25000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.76,
|
312 |
+
"learning_rate": 1.1749617496174963e-05,
|
313 |
+
"loss": 1.1351,
|
314 |
+
"step": 25500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.78,
|
318 |
+
"learning_rate": 1.0999609996099962e-05,
|
319 |
+
"loss": 1.1364,
|
320 |
+
"step": 26000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.79,
|
324 |
+
"learning_rate": 1.0249602496024962e-05,
|
325 |
+
"loss": 1.1322,
|
326 |
+
"step": 26500
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.81,
|
330 |
+
"learning_rate": 9.49959499594996e-06,
|
331 |
+
"loss": 1.1337,
|
332 |
+
"step": 27000
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.82,
|
336 |
+
"learning_rate": 8.749587495874959e-06,
|
337 |
+
"loss": 1.1322,
|
338 |
+
"step": 27500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.84,
|
342 |
+
"learning_rate": 7.999579995799957e-06,
|
343 |
+
"loss": 1.128,
|
344 |
+
"step": 28000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.85,
|
348 |
+
"learning_rate": 7.249572495724958e-06,
|
349 |
+
"loss": 1.1318,
|
350 |
+
"step": 28500
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 0.87,
|
354 |
+
"learning_rate": 6.499564995649957e-06,
|
355 |
+
"loss": 1.1283,
|
356 |
+
"step": 29000
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.88,
|
360 |
+
"learning_rate": 5.749557495574956e-06,
|
361 |
+
"loss": 1.1262,
|
362 |
+
"step": 29500
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 0.9,
|
366 |
+
"learning_rate": 4.999549995499956e-06,
|
367 |
+
"loss": 1.1295,
|
368 |
+
"step": 30000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.91,
|
372 |
+
"learning_rate": 4.249542495424955e-06,
|
373 |
+
"loss": 1.125,
|
374 |
+
"step": 30500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.93,
|
378 |
+
"learning_rate": 3.499534995349954e-06,
|
379 |
+
"loss": 1.1241,
|
380 |
+
"step": 31000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.94,
|
384 |
+
"learning_rate": 2.749527495274953e-06,
|
385 |
+
"loss": 1.1155,
|
386 |
+
"step": 31500
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 0.96,
|
390 |
+
"learning_rate": 1.999519995199952e-06,
|
391 |
+
"loss": 1.1226,
|
392 |
+
"step": 32000
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 0.97,
|
396 |
+
"learning_rate": 1.2495124951249513e-06,
|
397 |
+
"loss": 1.1281,
|
398 |
+
"step": 32500
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 0.99,
|
402 |
+
"learning_rate": 4.995049950499505e-07,
|
403 |
+
"loss": 1.1218,
|
404 |
+
"step": 33000
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 1.0,
|
408 |
+
"eval_bleu": 33.5469,
|
409 |
+
"eval_gen_len": 17.7505,
|
410 |
+
"eval_loss": 1.3453229665756226,
|
411 |
+
"eval_runtime": 449.9191,
|
412 |
+
"eval_samples_per_second": 4.445,
|
413 |
+
"step": 33333
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 1.0,
|
417 |
+
"step": 33333,
|
418 |
+
"total_flos": 2.7716100970398106e+18,
|
419 |
+
"train_runtime": 119333.5912,
|
420 |
+
"train_samples_per_second": 0.279
|
421 |
+
}
|
422 |
+
],
|
423 |
+
"max_steps": 33333,
|
424 |
+
"num_train_epochs": 1,
|
425 |
+
"total_flos": 2.7716100970398106e+18,
|
426 |
+
"trial_name": null,
|
427 |
+
"trial_params": null
|
428 |
+
}
|