Upload trainer_state.json
Browse files- trainer_state.json +691 -0
trainer_state.json
ADDED
@@ -0,0 +1,691 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.998109640831758,
|
5 |
+
"global_step": 99,
|
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": 6.25e-07,
|
13 |
+
"loss": 0.2195,
|
14 |
+
"step": 1
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 1.25e-06,
|
19 |
+
"loss": 0.2205,
|
20 |
+
"step": 2
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.03,
|
24 |
+
"learning_rate": 1.8750000000000003e-06,
|
25 |
+
"loss": 0.22,
|
26 |
+
"step": 3
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.04,
|
30 |
+
"learning_rate": 2.5e-06,
|
31 |
+
"loss": 0.2206,
|
32 |
+
"step": 4
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.05,
|
36 |
+
"learning_rate": 3.125e-06,
|
37 |
+
"loss": 0.2181,
|
38 |
+
"step": 5
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.06,
|
42 |
+
"learning_rate": 3.7500000000000005e-06,
|
43 |
+
"loss": 0.2178,
|
44 |
+
"step": 6
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.07,
|
48 |
+
"learning_rate": 4.3750000000000005e-06,
|
49 |
+
"loss": 0.218,
|
50 |
+
"step": 7
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.08,
|
54 |
+
"learning_rate": 5e-06,
|
55 |
+
"loss": 0.2159,
|
56 |
+
"step": 8
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.09,
|
60 |
+
"learning_rate": 5.625e-06,
|
61 |
+
"loss": 0.2109,
|
62 |
+
"step": 9
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.1,
|
66 |
+
"learning_rate": 6.25e-06,
|
67 |
+
"loss": 0.2079,
|
68 |
+
"step": 10
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.1,
|
72 |
+
"eval_loss": 0.20294944941997528,
|
73 |
+
"eval_runtime": 7.7949,
|
74 |
+
"eval_samples_per_second": 122.387,
|
75 |
+
"eval_steps_per_second": 15.395,
|
76 |
+
"step": 10
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.11,
|
80 |
+
"learning_rate": 6.875e-06,
|
81 |
+
"loss": 0.2064,
|
82 |
+
"step": 11
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.12,
|
86 |
+
"learning_rate": 7.500000000000001e-06,
|
87 |
+
"loss": 0.201,
|
88 |
+
"step": 12
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.13,
|
92 |
+
"learning_rate": 8.125000000000001e-06,
|
93 |
+
"loss": 0.2004,
|
94 |
+
"step": 13
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.14,
|
98 |
+
"learning_rate": 8.750000000000001e-06,
|
99 |
+
"loss": 0.196,
|
100 |
+
"step": 14
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.15,
|
104 |
+
"learning_rate": 9.375000000000001e-06,
|
105 |
+
"loss": 0.193,
|
106 |
+
"step": 15
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.16,
|
110 |
+
"learning_rate": 1e-05,
|
111 |
+
"loss": 0.1892,
|
112 |
+
"step": 16
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.17,
|
116 |
+
"learning_rate": 9.681174353198687e-06,
|
117 |
+
"loss": 0.1893,
|
118 |
+
"step": 17
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.18,
|
122 |
+
"learning_rate": 8.765357330018056e-06,
|
123 |
+
"loss": 0.1845,
|
124 |
+
"step": 18
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.19,
|
128 |
+
"learning_rate": 7.369343312364994e-06,
|
129 |
+
"loss": 0.1832,
|
130 |
+
"step": 19
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.2,
|
134 |
+
"learning_rate": 5.671166329088278e-06,
|
135 |
+
"loss": 0.1822,
|
136 |
+
"step": 20
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.2,
|
140 |
+
"eval_loss": 0.18006880581378937,
|
141 |
+
"eval_runtime": 7.7156,
|
142 |
+
"eval_samples_per_second": 123.646,
|
143 |
+
"eval_steps_per_second": 15.553,
|
144 |
+
"step": 20
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.21,
|
148 |
+
"learning_rate": 3.887395330218429e-06,
|
149 |
+
"loss": 0.182,
|
150 |
+
"step": 21
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.22,
|
154 |
+
"learning_rate": 2.245515092739488e-06,
|
155 |
+
"loss": 0.1818,
|
156 |
+
"step": 22
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.23,
|
160 |
+
"learning_rate": 9.549150281252633e-07,
|
161 |
+
"loss": 0.1828,
|
162 |
+
"step": 23
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.24,
|
166 |
+
"learning_rate": 1.801856965207338e-07,
|
167 |
+
"loss": 0.1803,
|
168 |
+
"step": 24
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.25,
|
172 |
+
"learning_rate": 2.012853002380466e-08,
|
173 |
+
"loss": 0.181,
|
174 |
+
"step": 25
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.26,
|
178 |
+
"learning_rate": 4.951556604879043e-07,
|
179 |
+
"loss": 0.1795,
|
180 |
+
"step": 26
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.27,
|
184 |
+
"learning_rate": 1.5446867550656763e-06,
|
185 |
+
"loss": 0.1838,
|
186 |
+
"step": 27
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 0.28,
|
190 |
+
"learning_rate": 3.0348748417303813e-06,
|
191 |
+
"loss": 0.1815,
|
192 |
+
"step": 28
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 0.29,
|
196 |
+
"learning_rate": 4.775675848247425e-06,
|
197 |
+
"loss": 0.1799,
|
198 |
+
"step": 29
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.3,
|
202 |
+
"learning_rate": 6.545084971874737e-06,
|
203 |
+
"loss": 0.1816,
|
204 |
+
"step": 30
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.3,
|
208 |
+
"eval_loss": 0.17900113761425018,
|
209 |
+
"eval_runtime": 7.6855,
|
210 |
+
"eval_samples_per_second": 124.129,
|
211 |
+
"eval_steps_per_second": 15.614,
|
212 |
+
"step": 30
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.31,
|
216 |
+
"learning_rate": 8.117449009293668e-06,
|
217 |
+
"loss": 0.18,
|
218 |
+
"step": 31
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.32,
|
222 |
+
"learning_rate": 9.292243968009332e-06,
|
223 |
+
"loss": 0.1796,
|
224 |
+
"step": 32
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.33,
|
228 |
+
"learning_rate": 9.91964794299315e-06,
|
229 |
+
"loss": 0.1809,
|
230 |
+
"step": 33
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.34,
|
234 |
+
"learning_rate": 9.91964794299315e-06,
|
235 |
+
"loss": 0.1782,
|
236 |
+
"step": 34
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.35,
|
240 |
+
"learning_rate": 9.292243968009333e-06,
|
241 |
+
"loss": 0.1781,
|
242 |
+
"step": 35
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.36,
|
246 |
+
"learning_rate": 8.11744900929367e-06,
|
247 |
+
"loss": 0.1759,
|
248 |
+
"step": 36
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.37,
|
252 |
+
"learning_rate": 6.545084971874739e-06,
|
253 |
+
"loss": 0.1752,
|
254 |
+
"step": 37
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.38,
|
258 |
+
"learning_rate": 4.775675848247427e-06,
|
259 |
+
"loss": 0.1786,
|
260 |
+
"step": 38
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.39,
|
264 |
+
"learning_rate": 3.0348748417303834e-06,
|
265 |
+
"loss": 0.1758,
|
266 |
+
"step": 39
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.4,
|
270 |
+
"learning_rate": 1.544686755065678e-06,
|
271 |
+
"loss": 0.1756,
|
272 |
+
"step": 40
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.4,
|
276 |
+
"eval_loss": 0.17418844997882843,
|
277 |
+
"eval_runtime": 7.8587,
|
278 |
+
"eval_samples_per_second": 121.394,
|
279 |
+
"eval_steps_per_second": 15.27,
|
280 |
+
"step": 40
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 0.41,
|
284 |
+
"learning_rate": 4.951556604879054e-07,
|
285 |
+
"loss": 0.1749,
|
286 |
+
"step": 41
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 0.42,
|
290 |
+
"learning_rate": 2.0128530023805216e-08,
|
291 |
+
"loss": 0.1764,
|
292 |
+
"step": 42
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 0.43,
|
296 |
+
"learning_rate": 1.8018569652073326e-07,
|
297 |
+
"loss": 0.1761,
|
298 |
+
"step": 43
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 0.44,
|
302 |
+
"learning_rate": 9.549150281252616e-07,
|
303 |
+
"loss": 0.1782,
|
304 |
+
"step": 44
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 0.45,
|
308 |
+
"learning_rate": 2.245515092739486e-06,
|
309 |
+
"loss": 0.176,
|
310 |
+
"step": 45
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.46,
|
314 |
+
"learning_rate": 3.887395330218426e-06,
|
315 |
+
"loss": 0.1769,
|
316 |
+
"step": 46
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 0.47,
|
320 |
+
"learning_rate": 5.671166329088275e-06,
|
321 |
+
"loss": 0.1768,
|
322 |
+
"step": 47
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 0.48,
|
326 |
+
"learning_rate": 7.369343312364992e-06,
|
327 |
+
"loss": 0.1751,
|
328 |
+
"step": 48
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.49,
|
332 |
+
"learning_rate": 8.765357330018054e-06,
|
333 |
+
"loss": 0.1744,
|
334 |
+
"step": 49
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 0.5,
|
338 |
+
"learning_rate": 9.681174353198685e-06,
|
339 |
+
"loss": 0.1751,
|
340 |
+
"step": 50
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 0.5,
|
344 |
+
"eval_loss": 0.17353740334510803,
|
345 |
+
"eval_runtime": 7.7334,
|
346 |
+
"eval_samples_per_second": 123.361,
|
347 |
+
"eval_steps_per_second": 15.517,
|
348 |
+
"step": 50
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 0.51,
|
352 |
+
"learning_rate": 1e-05,
|
353 |
+
"loss": 0.1758,
|
354 |
+
"step": 51
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 0.52,
|
358 |
+
"learning_rate": 9.68117435319869e-06,
|
359 |
+
"loss": 0.1768,
|
360 |
+
"step": 52
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.53,
|
364 |
+
"learning_rate": 8.765357330018058e-06,
|
365 |
+
"loss": 0.1752,
|
366 |
+
"step": 53
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.54,
|
370 |
+
"learning_rate": 7.369343312365004e-06,
|
371 |
+
"loss": 0.1746,
|
372 |
+
"step": 54
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 0.55,
|
376 |
+
"learning_rate": 5.6711663290882805e-06,
|
377 |
+
"loss": 0.1722,
|
378 |
+
"step": 55
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 0.56,
|
382 |
+
"learning_rate": 3.887395330218431e-06,
|
383 |
+
"loss": 0.1736,
|
384 |
+
"step": 56
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 0.57,
|
388 |
+
"learning_rate": 2.24551509273949e-06,
|
389 |
+
"loss": 0.1727,
|
390 |
+
"step": 57
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 0.58,
|
394 |
+
"learning_rate": 9.549150281252646e-07,
|
395 |
+
"loss": 0.1713,
|
396 |
+
"step": 58
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 0.59,
|
400 |
+
"learning_rate": 1.8018569652073214e-07,
|
401 |
+
"loss": 0.1732,
|
402 |
+
"step": 59
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 0.6,
|
406 |
+
"learning_rate": 2.012853002380466e-08,
|
407 |
+
"loss": 0.1738,
|
408 |
+
"step": 60
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.6,
|
412 |
+
"eval_loss": 0.17227524518966675,
|
413 |
+
"eval_runtime": 7.7647,
|
414 |
+
"eval_samples_per_second": 122.863,
|
415 |
+
"eval_steps_per_second": 15.455,
|
416 |
+
"step": 60
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.61,
|
420 |
+
"learning_rate": 4.951556604879032e-07,
|
421 |
+
"loss": 0.1744,
|
422 |
+
"step": 61
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.63,
|
426 |
+
"learning_rate": 1.5446867550656746e-06,
|
427 |
+
"loss": 0.1741,
|
428 |
+
"step": 62
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.64,
|
432 |
+
"learning_rate": 3.034874841730379e-06,
|
433 |
+
"loss": 0.1749,
|
434 |
+
"step": 63
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.65,
|
438 |
+
"learning_rate": 4.775675848247422e-06,
|
439 |
+
"loss": 0.1745,
|
440 |
+
"step": 64
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.66,
|
444 |
+
"learning_rate": 6.545084971874734e-06,
|
445 |
+
"loss": 0.1751,
|
446 |
+
"step": 65
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.67,
|
450 |
+
"learning_rate": 8.117449009293666e-06,
|
451 |
+
"loss": 0.1751,
|
452 |
+
"step": 66
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.68,
|
456 |
+
"learning_rate": 9.292243968009328e-06,
|
457 |
+
"loss": 0.1758,
|
458 |
+
"step": 67
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.69,
|
462 |
+
"learning_rate": 9.91964794299315e-06,
|
463 |
+
"loss": 0.174,
|
464 |
+
"step": 68
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.7,
|
468 |
+
"learning_rate": 9.91964794299315e-06,
|
469 |
+
"loss": 0.1723,
|
470 |
+
"step": 69
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.71,
|
474 |
+
"learning_rate": 9.292243968009333e-06,
|
475 |
+
"loss": 0.1734,
|
476 |
+
"step": 70
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.71,
|
480 |
+
"eval_loss": 0.17139491438865662,
|
481 |
+
"eval_runtime": 8.0933,
|
482 |
+
"eval_samples_per_second": 117.875,
|
483 |
+
"eval_steps_per_second": 14.827,
|
484 |
+
"step": 70
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 0.72,
|
488 |
+
"learning_rate": 8.117449009293671e-06,
|
489 |
+
"loss": 0.1725,
|
490 |
+
"step": 71
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 0.73,
|
494 |
+
"learning_rate": 6.545084971874741e-06,
|
495 |
+
"loss": 0.1739,
|
496 |
+
"step": 72
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 0.74,
|
500 |
+
"learning_rate": 4.77567584824743e-06,
|
501 |
+
"loss": 0.1728,
|
502 |
+
"step": 73
|
503 |
+
},
|
504 |
+
{
|
505 |
+
"epoch": 0.75,
|
506 |
+
"learning_rate": 3.034874841730386e-06,
|
507 |
+
"loss": 0.1723,
|
508 |
+
"step": 74
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 0.76,
|
512 |
+
"learning_rate": 1.5446867550656797e-06,
|
513 |
+
"loss": 0.1757,
|
514 |
+
"step": 75
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 0.77,
|
518 |
+
"learning_rate": 4.951556604879059e-07,
|
519 |
+
"loss": 0.1741,
|
520 |
+
"step": 76
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.78,
|
524 |
+
"learning_rate": 2.0128530023805216e-08,
|
525 |
+
"loss": 0.1739,
|
526 |
+
"step": 77
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 0.79,
|
530 |
+
"learning_rate": 1.8018569652073214e-07,
|
531 |
+
"loss": 0.1728,
|
532 |
+
"step": 78
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.8,
|
536 |
+
"learning_rate": 9.549150281252605e-07,
|
537 |
+
"loss": 0.1723,
|
538 |
+
"step": 79
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 0.81,
|
542 |
+
"learning_rate": 2.2455150927394835e-06,
|
543 |
+
"loss": 0.1727,
|
544 |
+
"step": 80
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 0.81,
|
548 |
+
"eval_loss": 0.17101411521434784,
|
549 |
+
"eval_runtime": 7.7464,
|
550 |
+
"eval_samples_per_second": 123.153,
|
551 |
+
"eval_steps_per_second": 15.491,
|
552 |
+
"step": 80
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 0.82,
|
556 |
+
"learning_rate": 3.8873953302184244e-06,
|
557 |
+
"loss": 0.1726,
|
558 |
+
"step": 81
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 0.83,
|
562 |
+
"learning_rate": 5.671166329088274e-06,
|
563 |
+
"loss": 0.1726,
|
564 |
+
"step": 82
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 0.84,
|
568 |
+
"learning_rate": 7.369343312364989e-06,
|
569 |
+
"loss": 0.1728,
|
570 |
+
"step": 83
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 0.85,
|
574 |
+
"learning_rate": 8.765357330018051e-06,
|
575 |
+
"loss": 0.1724,
|
576 |
+
"step": 84
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.86,
|
580 |
+
"learning_rate": 9.681174353198685e-06,
|
581 |
+
"loss": 0.171,
|
582 |
+
"step": 85
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 0.87,
|
586 |
+
"learning_rate": 1e-05,
|
587 |
+
"loss": 0.1717,
|
588 |
+
"step": 86
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 0.88,
|
592 |
+
"learning_rate": 9.681174353198689e-06,
|
593 |
+
"loss": 0.174,
|
594 |
+
"step": 87
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 0.89,
|
598 |
+
"learning_rate": 8.76535733001807e-06,
|
599 |
+
"loss": 0.1716,
|
600 |
+
"step": 88
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 0.9,
|
604 |
+
"learning_rate": 7.369343312364998e-06,
|
605 |
+
"loss": 0.1715,
|
606 |
+
"step": 89
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 0.91,
|
610 |
+
"learning_rate": 5.671166329088282e-06,
|
611 |
+
"loss": 0.1712,
|
612 |
+
"step": 90
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 0.91,
|
616 |
+
"eval_loss": 0.17012353241443634,
|
617 |
+
"eval_runtime": 7.792,
|
618 |
+
"eval_samples_per_second": 122.433,
|
619 |
+
"eval_steps_per_second": 15.4,
|
620 |
+
"step": 90
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.92,
|
624 |
+
"learning_rate": 3.887395330218433e-06,
|
625 |
+
"loss": 0.173,
|
626 |
+
"step": 91
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.93,
|
630 |
+
"learning_rate": 2.245515092739507e-06,
|
631 |
+
"loss": 0.1731,
|
632 |
+
"step": 92
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.94,
|
636 |
+
"learning_rate": 9.549150281252557e-07,
|
637 |
+
"loss": 0.1727,
|
638 |
+
"step": 93
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 0.95,
|
642 |
+
"learning_rate": 1.8018569652073492e-07,
|
643 |
+
"loss": 0.1724,
|
644 |
+
"step": 94
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 0.96,
|
648 |
+
"learning_rate": 2.0128530023804104e-08,
|
649 |
+
"loss": 0.1695,
|
650 |
+
"step": 95
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 0.97,
|
654 |
+
"learning_rate": 4.951556604879021e-07,
|
655 |
+
"loss": 0.1711,
|
656 |
+
"step": 96
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.98,
|
660 |
+
"learning_rate": 1.5446867550656852e-06,
|
661 |
+
"loss": 0.1726,
|
662 |
+
"step": 97
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.99,
|
666 |
+
"learning_rate": 3.034874841730377e-06,
|
667 |
+
"loss": 0.1722,
|
668 |
+
"step": 98
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.0,
|
672 |
+
"learning_rate": 4.7756758482474195e-06,
|
673 |
+
"loss": 0.171,
|
674 |
+
"step": 99
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 1.0,
|
678 |
+
"step": 99,
|
679 |
+
"total_flos": 6.416685198968095e+18,
|
680 |
+
"train_loss": 0.18112580719018223,
|
681 |
+
"train_runtime": 16993.0978,
|
682 |
+
"train_samples_per_second": 47.816,
|
683 |
+
"train_steps_per_second": 0.006
|
684 |
+
}
|
685 |
+
],
|
686 |
+
"max_steps": 99,
|
687 |
+
"num_train_epochs": 1,
|
688 |
+
"total_flos": 6.416685198968095e+18,
|
689 |
+
"trial_name": null,
|
690 |
+
"trial_params": null
|
691 |
+
}
|