Upload checkpoint-1000/trainer_state.json with huggingface_hub
Browse files
checkpoint-1000/trainer_state.json
ADDED
@@ -0,0 +1,656 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.8102368116378784,
|
3 |
+
"best_model_checkpoint": "./lora-alpaca/checkpoint-1000",
|
4 |
+
"epoch": 2.5624599615631007,
|
5 |
+
"global_step": 1000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 1e-05,
|
13 |
+
"loss": 2.0409,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 2e-05,
|
19 |
+
"loss": 2.1188,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.08,
|
24 |
+
"learning_rate": 3e-05,
|
25 |
+
"loss": 2.0143,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 4e-05,
|
31 |
+
"loss": 1.8108,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.13,
|
36 |
+
"learning_rate": 5e-05,
|
37 |
+
"loss": 1.4764,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.15,
|
42 |
+
"learning_rate": 6e-05,
|
43 |
+
"loss": 1.1516,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.18,
|
48 |
+
"learning_rate": 6.9e-05,
|
49 |
+
"loss": 0.9453,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.2,
|
54 |
+
"learning_rate": 7.900000000000001e-05,
|
55 |
+
"loss": 0.8503,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.23,
|
60 |
+
"learning_rate": 8.900000000000001e-05,
|
61 |
+
"loss": 0.8352,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.26,
|
66 |
+
"learning_rate": 9.900000000000001e-05,
|
67 |
+
"loss": 0.7635,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.28,
|
72 |
+
"learning_rate": 9.91588785046729e-05,
|
73 |
+
"loss": 0.8399,
|
74 |
+
"step": 110
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.31,
|
78 |
+
"learning_rate": 9.822429906542057e-05,
|
79 |
+
"loss": 0.7683,
|
80 |
+
"step": 120
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.33,
|
84 |
+
"learning_rate": 9.728971962616824e-05,
|
85 |
+
"loss": 0.7774,
|
86 |
+
"step": 130
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.36,
|
90 |
+
"learning_rate": 9.635514018691589e-05,
|
91 |
+
"loss": 0.7912,
|
92 |
+
"step": 140
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.38,
|
96 |
+
"learning_rate": 9.542056074766356e-05,
|
97 |
+
"loss": 0.7262,
|
98 |
+
"step": 150
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.41,
|
102 |
+
"learning_rate": 9.448598130841123e-05,
|
103 |
+
"loss": 0.8167,
|
104 |
+
"step": 160
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.44,
|
108 |
+
"learning_rate": 9.355140186915888e-05,
|
109 |
+
"loss": 0.7634,
|
110 |
+
"step": 170
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.46,
|
114 |
+
"learning_rate": 9.261682242990655e-05,
|
115 |
+
"loss": 0.7666,
|
116 |
+
"step": 180
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.49,
|
120 |
+
"learning_rate": 9.168224299065422e-05,
|
121 |
+
"loss": 0.7943,
|
122 |
+
"step": 190
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.51,
|
126 |
+
"learning_rate": 9.074766355140187e-05,
|
127 |
+
"loss": 0.7366,
|
128 |
+
"step": 200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.51,
|
132 |
+
"eval_loss": 0.8413083553314209,
|
133 |
+
"eval_runtime": 68.432,
|
134 |
+
"eval_samples_per_second": 29.226,
|
135 |
+
"eval_steps_per_second": 0.468,
|
136 |
+
"step": 200
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.54,
|
140 |
+
"learning_rate": 8.981308411214954e-05,
|
141 |
+
"loss": 0.8029,
|
142 |
+
"step": 210
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.56,
|
146 |
+
"learning_rate": 8.88785046728972e-05,
|
147 |
+
"loss": 0.7383,
|
148 |
+
"step": 220
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 0.59,
|
152 |
+
"learning_rate": 8.794392523364486e-05,
|
153 |
+
"loss": 0.7692,
|
154 |
+
"step": 230
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 0.61,
|
158 |
+
"learning_rate": 8.700934579439253e-05,
|
159 |
+
"loss": 0.7673,
|
160 |
+
"step": 240
|
161 |
+
},
|
162 |
+
{
|
163 |
+
"epoch": 0.64,
|
164 |
+
"learning_rate": 8.607476635514018e-05,
|
165 |
+
"loss": 0.7129,
|
166 |
+
"step": 250
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 0.67,
|
170 |
+
"learning_rate": 8.514018691588785e-05,
|
171 |
+
"loss": 0.7913,
|
172 |
+
"step": 260
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 0.69,
|
176 |
+
"learning_rate": 8.420560747663552e-05,
|
177 |
+
"loss": 0.7432,
|
178 |
+
"step": 270
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"epoch": 0.72,
|
182 |
+
"learning_rate": 8.327102803738318e-05,
|
183 |
+
"loss": 0.7617,
|
184 |
+
"step": 280
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.74,
|
188 |
+
"learning_rate": 8.233644859813084e-05,
|
189 |
+
"loss": 0.7783,
|
190 |
+
"step": 290
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 0.77,
|
194 |
+
"learning_rate": 8.140186915887851e-05,
|
195 |
+
"loss": 0.7146,
|
196 |
+
"step": 300
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.79,
|
200 |
+
"learning_rate": 8.046728971962617e-05,
|
201 |
+
"loss": 0.7992,
|
202 |
+
"step": 310
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 0.82,
|
206 |
+
"learning_rate": 7.953271028037383e-05,
|
207 |
+
"loss": 0.738,
|
208 |
+
"step": 320
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 0.85,
|
212 |
+
"learning_rate": 7.85981308411215e-05,
|
213 |
+
"loss": 0.7477,
|
214 |
+
"step": 330
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 0.87,
|
218 |
+
"learning_rate": 7.766355140186916e-05,
|
219 |
+
"loss": 0.7698,
|
220 |
+
"step": 340
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 0.9,
|
224 |
+
"learning_rate": 7.672897196261682e-05,
|
225 |
+
"loss": 0.714,
|
226 |
+
"step": 350
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.92,
|
230 |
+
"learning_rate": 7.579439252336449e-05,
|
231 |
+
"loss": 0.7939,
|
232 |
+
"step": 360
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 0.95,
|
236 |
+
"learning_rate": 7.485981308411215e-05,
|
237 |
+
"loss": 0.7394,
|
238 |
+
"step": 370
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 0.97,
|
242 |
+
"learning_rate": 7.392523364485982e-05,
|
243 |
+
"loss": 0.7481,
|
244 |
+
"step": 380
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 1.0,
|
248 |
+
"learning_rate": 7.299065420560748e-05,
|
249 |
+
"loss": 0.7446,
|
250 |
+
"step": 390
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 1.02,
|
254 |
+
"learning_rate": 7.205607476635514e-05,
|
255 |
+
"loss": 0.7805,
|
256 |
+
"step": 400
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 1.02,
|
260 |
+
"eval_loss": 0.8240922093391418,
|
261 |
+
"eval_runtime": 68.3178,
|
262 |
+
"eval_samples_per_second": 29.275,
|
263 |
+
"eval_steps_per_second": 0.468,
|
264 |
+
"step": 400
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.05,
|
268 |
+
"learning_rate": 7.11214953271028e-05,
|
269 |
+
"loss": 0.7309,
|
270 |
+
"step": 410
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 1.08,
|
274 |
+
"learning_rate": 7.018691588785047e-05,
|
275 |
+
"loss": 0.7595,
|
276 |
+
"step": 420
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.1,
|
280 |
+
"learning_rate": 6.925233644859813e-05,
|
281 |
+
"loss": 0.7693,
|
282 |
+
"step": 430
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 1.13,
|
286 |
+
"learning_rate": 6.83177570093458e-05,
|
287 |
+
"loss": 0.7108,
|
288 |
+
"step": 440
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.15,
|
292 |
+
"learning_rate": 6.738317757009346e-05,
|
293 |
+
"loss": 0.7988,
|
294 |
+
"step": 450
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.18,
|
298 |
+
"learning_rate": 6.644859813084112e-05,
|
299 |
+
"loss": 0.7316,
|
300 |
+
"step": 460
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 1.2,
|
304 |
+
"learning_rate": 6.551401869158879e-05,
|
305 |
+
"loss": 0.7518,
|
306 |
+
"step": 470
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 1.23,
|
310 |
+
"learning_rate": 6.457943925233646e-05,
|
311 |
+
"loss": 0.7579,
|
312 |
+
"step": 480
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 1.26,
|
316 |
+
"learning_rate": 6.364485981308411e-05,
|
317 |
+
"loss": 0.7115,
|
318 |
+
"step": 490
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.28,
|
322 |
+
"learning_rate": 6.271028037383178e-05,
|
323 |
+
"loss": 0.7861,
|
324 |
+
"step": 500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.31,
|
328 |
+
"learning_rate": 6.177570093457945e-05,
|
329 |
+
"loss": 0.7292,
|
330 |
+
"step": 510
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.33,
|
334 |
+
"learning_rate": 6.08411214953271e-05,
|
335 |
+
"loss": 0.7471,
|
336 |
+
"step": 520
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 1.36,
|
340 |
+
"learning_rate": 5.990654205607477e-05,
|
341 |
+
"loss": 0.7458,
|
342 |
+
"step": 530
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.38,
|
346 |
+
"learning_rate": 5.897196261682243e-05,
|
347 |
+
"loss": 0.7164,
|
348 |
+
"step": 540
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 1.41,
|
352 |
+
"learning_rate": 5.803738317757009e-05,
|
353 |
+
"loss": 0.7867,
|
354 |
+
"step": 550
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.43,
|
358 |
+
"learning_rate": 5.710280373831776e-05,
|
359 |
+
"loss": 0.7165,
|
360 |
+
"step": 560
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.46,
|
364 |
+
"learning_rate": 5.616822429906542e-05,
|
365 |
+
"loss": 0.7424,
|
366 |
+
"step": 570
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 1.49,
|
370 |
+
"learning_rate": 5.523364485981308e-05,
|
371 |
+
"loss": 0.7438,
|
372 |
+
"step": 580
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 1.51,
|
376 |
+
"learning_rate": 5.429906542056075e-05,
|
377 |
+
"loss": 0.702,
|
378 |
+
"step": 590
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 1.54,
|
382 |
+
"learning_rate": 5.336448598130841e-05,
|
383 |
+
"loss": 0.7791,
|
384 |
+
"step": 600
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.54,
|
388 |
+
"eval_loss": 0.8157330751419067,
|
389 |
+
"eval_runtime": 68.2707,
|
390 |
+
"eval_samples_per_second": 29.295,
|
391 |
+
"eval_steps_per_second": 0.469,
|
392 |
+
"step": 600
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 1.56,
|
396 |
+
"learning_rate": 5.242990654205607e-05,
|
397 |
+
"loss": 0.7248,
|
398 |
+
"step": 610
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 1.59,
|
402 |
+
"learning_rate": 5.1495327102803733e-05,
|
403 |
+
"loss": 0.7558,
|
404 |
+
"step": 620
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 1.61,
|
408 |
+
"learning_rate": 5.05607476635514e-05,
|
409 |
+
"loss": 0.7438,
|
410 |
+
"step": 630
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 1.64,
|
414 |
+
"learning_rate": 4.962616822429907e-05,
|
415 |
+
"loss": 0.7094,
|
416 |
+
"step": 640
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.67,
|
420 |
+
"learning_rate": 4.869158878504673e-05,
|
421 |
+
"loss": 0.7733,
|
422 |
+
"step": 650
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 1.69,
|
426 |
+
"learning_rate": 4.77570093457944e-05,
|
427 |
+
"loss": 0.7162,
|
428 |
+
"step": 660
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 1.72,
|
432 |
+
"learning_rate": 4.682242990654206e-05,
|
433 |
+
"loss": 0.7397,
|
434 |
+
"step": 670
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 1.74,
|
438 |
+
"learning_rate": 4.588785046728972e-05,
|
439 |
+
"loss": 0.7579,
|
440 |
+
"step": 680
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 1.77,
|
444 |
+
"learning_rate": 4.495327102803738e-05,
|
445 |
+
"loss": 0.7055,
|
446 |
+
"step": 690
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 1.79,
|
450 |
+
"learning_rate": 4.401869158878505e-05,
|
451 |
+
"loss": 0.7685,
|
452 |
+
"step": 700
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 1.82,
|
456 |
+
"learning_rate": 4.308411214953271e-05,
|
457 |
+
"loss": 0.7266,
|
458 |
+
"step": 710
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.84,
|
462 |
+
"learning_rate": 4.2149532710280373e-05,
|
463 |
+
"loss": 0.7469,
|
464 |
+
"step": 720
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 1.87,
|
468 |
+
"learning_rate": 4.121495327102804e-05,
|
469 |
+
"loss": 0.7559,
|
470 |
+
"step": 730
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 1.9,
|
474 |
+
"learning_rate": 4.02803738317757e-05,
|
475 |
+
"loss": 0.6953,
|
476 |
+
"step": 740
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 1.92,
|
480 |
+
"learning_rate": 3.9345794392523364e-05,
|
481 |
+
"loss": 0.7702,
|
482 |
+
"step": 750
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 1.95,
|
486 |
+
"learning_rate": 3.841121495327103e-05,
|
487 |
+
"loss": 0.7111,
|
488 |
+
"step": 760
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 1.97,
|
492 |
+
"learning_rate": 3.7476635514018693e-05,
|
493 |
+
"loss": 0.7348,
|
494 |
+
"step": 770
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.0,
|
498 |
+
"learning_rate": 3.6542056074766355e-05,
|
499 |
+
"loss": 0.7146,
|
500 |
+
"step": 780
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 2.02,
|
504 |
+
"learning_rate": 3.560747663551402e-05,
|
505 |
+
"loss": 0.7753,
|
506 |
+
"step": 790
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 2.05,
|
510 |
+
"learning_rate": 3.4672897196261684e-05,
|
511 |
+
"loss": 0.717,
|
512 |
+
"step": 800
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 2.05,
|
516 |
+
"eval_loss": 0.8136950731277466,
|
517 |
+
"eval_runtime": 68.3249,
|
518 |
+
"eval_samples_per_second": 29.272,
|
519 |
+
"eval_steps_per_second": 0.468,
|
520 |
+
"step": 800
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 2.08,
|
524 |
+
"learning_rate": 3.3738317757009345e-05,
|
525 |
+
"loss": 0.7384,
|
526 |
+
"step": 810
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 2.1,
|
530 |
+
"learning_rate": 3.2803738317757014e-05,
|
531 |
+
"loss": 0.7444,
|
532 |
+
"step": 820
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 2.13,
|
536 |
+
"learning_rate": 3.1869158878504675e-05,
|
537 |
+
"loss": 0.713,
|
538 |
+
"step": 830
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 2.15,
|
542 |
+
"learning_rate": 3.0934579439252336e-05,
|
543 |
+
"loss": 0.7782,
|
544 |
+
"step": 840
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 2.18,
|
548 |
+
"learning_rate": 3e-05,
|
549 |
+
"loss": 0.7091,
|
550 |
+
"step": 850
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 2.2,
|
554 |
+
"learning_rate": 2.9065420560747662e-05,
|
555 |
+
"loss": 0.7428,
|
556 |
+
"step": 860
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 2.23,
|
560 |
+
"learning_rate": 2.8130841121495327e-05,
|
561 |
+
"loss": 0.7546,
|
562 |
+
"step": 870
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 2.25,
|
566 |
+
"learning_rate": 2.719626168224299e-05,
|
567 |
+
"loss": 0.7122,
|
568 |
+
"step": 880
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 2.28,
|
572 |
+
"learning_rate": 2.6261682242990653e-05,
|
573 |
+
"loss": 0.764,
|
574 |
+
"step": 890
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 2.31,
|
578 |
+
"learning_rate": 2.5327102803738317e-05,
|
579 |
+
"loss": 0.7097,
|
580 |
+
"step": 900
|
581 |
+
},
|
582 |
+
{
|
583 |
+
"epoch": 2.33,
|
584 |
+
"learning_rate": 2.4392523364485982e-05,
|
585 |
+
"loss": 0.7546,
|
586 |
+
"step": 910
|
587 |
+
},
|
588 |
+
{
|
589 |
+
"epoch": 2.36,
|
590 |
+
"learning_rate": 2.3457943925233647e-05,
|
591 |
+
"loss": 0.7304,
|
592 |
+
"step": 920
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"epoch": 2.38,
|
596 |
+
"learning_rate": 2.252336448598131e-05,
|
597 |
+
"loss": 0.7117,
|
598 |
+
"step": 930
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 2.41,
|
602 |
+
"learning_rate": 2.1588785046728973e-05,
|
603 |
+
"loss": 0.7642,
|
604 |
+
"step": 940
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 2.43,
|
608 |
+
"learning_rate": 2.0654205607476637e-05,
|
609 |
+
"loss": 0.7092,
|
610 |
+
"step": 950
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 2.46,
|
614 |
+
"learning_rate": 1.9719626168224302e-05,
|
615 |
+
"loss": 0.7555,
|
616 |
+
"step": 960
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 2.49,
|
620 |
+
"learning_rate": 1.8785046728971963e-05,
|
621 |
+
"loss": 0.7411,
|
622 |
+
"step": 970
|
623 |
+
},
|
624 |
+
{
|
625 |
+
"epoch": 2.51,
|
626 |
+
"learning_rate": 1.7850467289719628e-05,
|
627 |
+
"loss": 0.7126,
|
628 |
+
"step": 980
|
629 |
+
},
|
630 |
+
{
|
631 |
+
"epoch": 2.54,
|
632 |
+
"learning_rate": 1.691588785046729e-05,
|
633 |
+
"loss": 0.7549,
|
634 |
+
"step": 990
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"epoch": 2.56,
|
638 |
+
"learning_rate": 1.5981308411214954e-05,
|
639 |
+
"loss": 0.7036,
|
640 |
+
"step": 1000
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 2.56,
|
644 |
+
"eval_loss": 0.8102368116378784,
|
645 |
+
"eval_runtime": 68.2256,
|
646 |
+
"eval_samples_per_second": 29.315,
|
647 |
+
"eval_steps_per_second": 0.469,
|
648 |
+
"step": 1000
|
649 |
+
}
|
650 |
+
],
|
651 |
+
"max_steps": 1170,
|
652 |
+
"num_train_epochs": 3,
|
653 |
+
"total_flos": 6.919323374783037e+18,
|
654 |
+
"trial_name": null,
|
655 |
+
"trial_params": null
|
656 |
+
}
|