End of training
Browse files- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +343 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.02,
|
| 3 |
+
"eval_accuracy": 0.5368494657483851,
|
| 4 |
+
"eval_loss": 2.2406654357910156,
|
| 5 |
+
"eval_runtime": 1840.2228,
|
| 6 |
+
"eval_samples": 154161,
|
| 7 |
+
"eval_samples_per_second": 83.773,
|
| 8 |
+
"eval_steps_per_second": 1.309,
|
| 9 |
+
"perplexity": 9.399584026445748,
|
| 10 |
+
"train_loss": 2.2496585121154786,
|
| 11 |
+
"train_runtime": 5217.4701,
|
| 12 |
+
"train_samples": 3578704,
|
| 13 |
+
"train_samples_per_second": 12.266,
|
| 14 |
+
"train_steps_per_second": 0.192
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.02,
|
| 3 |
+
"eval_accuracy": 0.5368494657483851,
|
| 4 |
+
"eval_loss": 2.2406654357910156,
|
| 5 |
+
"eval_runtime": 1840.2228,
|
| 6 |
+
"eval_samples": 154161,
|
| 7 |
+
"eval_samples_per_second": 83.773,
|
| 8 |
+
"eval_steps_per_second": 1.309,
|
| 9 |
+
"perplexity": 9.399584026445748
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.02,
|
| 3 |
+
"train_loss": 2.2496585121154786,
|
| 4 |
+
"train_runtime": 5217.4701,
|
| 5 |
+
"train_samples": 3578704,
|
| 6 |
+
"train_samples_per_second": 12.266,
|
| 7 |
+
"train_steps_per_second": 0.192
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,343 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.017883569023870095,
|
| 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": 0.0,
|
| 12 |
+
"learning_rate": 9.8e-05,
|
| 13 |
+
"loss": 2.2791,
|
| 14 |
+
"step": 20
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.0,
|
| 18 |
+
"learning_rate": 9.6e-05,
|
| 19 |
+
"loss": 2.27,
|
| 20 |
+
"step": 40
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.0,
|
| 24 |
+
"learning_rate": 9.4e-05,
|
| 25 |
+
"loss": 2.2689,
|
| 26 |
+
"step": 60
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.0,
|
| 30 |
+
"learning_rate": 9.200000000000001e-05,
|
| 31 |
+
"loss": 2.2549,
|
| 32 |
+
"step": 80
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.0,
|
| 36 |
+
"learning_rate": 9e-05,
|
| 37 |
+
"loss": 2.2822,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0,
|
| 42 |
+
"learning_rate": 8.800000000000001e-05,
|
| 43 |
+
"loss": 2.2564,
|
| 44 |
+
"step": 120
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0,
|
| 48 |
+
"learning_rate": 8.6e-05,
|
| 49 |
+
"loss": 2.2551,
|
| 50 |
+
"step": 140
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.0,
|
| 54 |
+
"learning_rate": 8.4e-05,
|
| 55 |
+
"loss": 2.2471,
|
| 56 |
+
"step": 160
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.0,
|
| 60 |
+
"learning_rate": 8.2e-05,
|
| 61 |
+
"loss": 2.2515,
|
| 62 |
+
"step": 180
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.0,
|
| 66 |
+
"learning_rate": 8e-05,
|
| 67 |
+
"loss": 2.2825,
|
| 68 |
+
"step": 200
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.0,
|
| 72 |
+
"learning_rate": 7.800000000000001e-05,
|
| 73 |
+
"loss": 2.2584,
|
| 74 |
+
"step": 220
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.0,
|
| 78 |
+
"learning_rate": 7.6e-05,
|
| 79 |
+
"loss": 2.2232,
|
| 80 |
+
"step": 240
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0,
|
| 84 |
+
"learning_rate": 7.4e-05,
|
| 85 |
+
"loss": 2.2463,
|
| 86 |
+
"step": 260
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.01,
|
| 90 |
+
"learning_rate": 7.2e-05,
|
| 91 |
+
"loss": 2.2775,
|
| 92 |
+
"step": 280
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.01,
|
| 96 |
+
"learning_rate": 7e-05,
|
| 97 |
+
"loss": 2.2533,
|
| 98 |
+
"step": 300
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.01,
|
| 102 |
+
"learning_rate": 6.800000000000001e-05,
|
| 103 |
+
"loss": 2.2596,
|
| 104 |
+
"step": 320
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.01,
|
| 108 |
+
"learning_rate": 6.6e-05,
|
| 109 |
+
"loss": 2.2307,
|
| 110 |
+
"step": 340
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.01,
|
| 114 |
+
"learning_rate": 6.400000000000001e-05,
|
| 115 |
+
"loss": 2.2671,
|
| 116 |
+
"step": 360
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.01,
|
| 120 |
+
"learning_rate": 6.2e-05,
|
| 121 |
+
"loss": 2.2191,
|
| 122 |
+
"step": 380
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01,
|
| 126 |
+
"learning_rate": 6e-05,
|
| 127 |
+
"loss": 2.2374,
|
| 128 |
+
"step": 400
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.01,
|
| 132 |
+
"learning_rate": 5.8e-05,
|
| 133 |
+
"loss": 2.2755,
|
| 134 |
+
"step": 420
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01,
|
| 138 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 139 |
+
"loss": 2.249,
|
| 140 |
+
"step": 440
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.01,
|
| 144 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 145 |
+
"loss": 2.2641,
|
| 146 |
+
"step": 460
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.01,
|
| 150 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 151 |
+
"loss": 2.2875,
|
| 152 |
+
"step": 480
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.01,
|
| 156 |
+
"learning_rate": 5e-05,
|
| 157 |
+
"loss": 2.2802,
|
| 158 |
+
"step": 500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.01,
|
| 162 |
+
"eval_accuracy": 0.5345466022376205,
|
| 163 |
+
"eval_loss": 2.2552647590637207,
|
| 164 |
+
"eval_runtime": 2008.2531,
|
| 165 |
+
"eval_samples_per_second": 76.764,
|
| 166 |
+
"eval_steps_per_second": 1.2,
|
| 167 |
+
"step": 500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.01,
|
| 171 |
+
"learning_rate": 4.8e-05,
|
| 172 |
+
"loss": 2.2486,
|
| 173 |
+
"step": 520
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.01,
|
| 177 |
+
"learning_rate": 4.600000000000001e-05,
|
| 178 |
+
"loss": 2.2124,
|
| 179 |
+
"step": 540
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.01,
|
| 183 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 184 |
+
"loss": 2.2253,
|
| 185 |
+
"step": 560
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01,
|
| 189 |
+
"learning_rate": 4.2e-05,
|
| 190 |
+
"loss": 2.2211,
|
| 191 |
+
"step": 580
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.01,
|
| 195 |
+
"learning_rate": 4e-05,
|
| 196 |
+
"loss": 2.2316,
|
| 197 |
+
"step": 600
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.01,
|
| 201 |
+
"learning_rate": 3.8e-05,
|
| 202 |
+
"loss": 2.2463,
|
| 203 |
+
"step": 620
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.01,
|
| 207 |
+
"learning_rate": 3.6e-05,
|
| 208 |
+
"loss": 2.252,
|
| 209 |
+
"step": 640
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.01,
|
| 213 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 214 |
+
"loss": 2.267,
|
| 215 |
+
"step": 660
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.01,
|
| 219 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 220 |
+
"loss": 2.2416,
|
| 221 |
+
"step": 680
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.01,
|
| 225 |
+
"learning_rate": 3e-05,
|
| 226 |
+
"loss": 2.2452,
|
| 227 |
+
"step": 700
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.01,
|
| 231 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 232 |
+
"loss": 2.242,
|
| 233 |
+
"step": 720
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.01,
|
| 237 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 238 |
+
"loss": 2.2572,
|
| 239 |
+
"step": 740
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.01,
|
| 243 |
+
"learning_rate": 2.4e-05,
|
| 244 |
+
"loss": 2.2346,
|
| 245 |
+
"step": 760
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.01,
|
| 249 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 250 |
+
"loss": 2.2022,
|
| 251 |
+
"step": 780
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.01,
|
| 255 |
+
"learning_rate": 2e-05,
|
| 256 |
+
"loss": 2.2317,
|
| 257 |
+
"step": 800
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.01,
|
| 261 |
+
"learning_rate": 1.8e-05,
|
| 262 |
+
"loss": 2.2348,
|
| 263 |
+
"step": 820
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.02,
|
| 267 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 268 |
+
"loss": 2.2453,
|
| 269 |
+
"step": 840
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02,
|
| 273 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 274 |
+
"loss": 2.2165,
|
| 275 |
+
"step": 860
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.02,
|
| 279 |
+
"learning_rate": 1.2e-05,
|
| 280 |
+
"loss": 2.2548,
|
| 281 |
+
"step": 880
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.02,
|
| 285 |
+
"learning_rate": 1e-05,
|
| 286 |
+
"loss": 2.2867,
|
| 287 |
+
"step": 900
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.02,
|
| 291 |
+
"learning_rate": 8.000000000000001e-06,
|
| 292 |
+
"loss": 2.2381,
|
| 293 |
+
"step": 920
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.02,
|
| 297 |
+
"learning_rate": 6e-06,
|
| 298 |
+
"loss": 2.242,
|
| 299 |
+
"step": 940
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.02,
|
| 303 |
+
"learning_rate": 4.000000000000001e-06,
|
| 304 |
+
"loss": 2.267,
|
| 305 |
+
"step": 960
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.02,
|
| 309 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 310 |
+
"loss": 2.2344,
|
| 311 |
+
"step": 980
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02,
|
| 315 |
+
"learning_rate": 0.0,
|
| 316 |
+
"loss": 2.2277,
|
| 317 |
+
"step": 1000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.02,
|
| 321 |
+
"eval_accuracy": 0.5368494657483851,
|
| 322 |
+
"eval_loss": 2.2406654357910156,
|
| 323 |
+
"eval_runtime": 1837.3548,
|
| 324 |
+
"eval_samples_per_second": 83.904,
|
| 325 |
+
"eval_steps_per_second": 1.311,
|
| 326 |
+
"step": 1000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.02,
|
| 330 |
+
"step": 1000,
|
| 331 |
+
"total_flos": 1.7866929733632e+17,
|
| 332 |
+
"train_loss": 2.2496585121154786,
|
| 333 |
+
"train_runtime": 5217.4701,
|
| 334 |
+
"train_samples_per_second": 12.266,
|
| 335 |
+
"train_steps_per_second": 0.192
|
| 336 |
+
}
|
| 337 |
+
],
|
| 338 |
+
"max_steps": 1000,
|
| 339 |
+
"num_train_epochs": 1,
|
| 340 |
+
"total_flos": 1.7866929733632e+17,
|
| 341 |
+
"trial_name": null,
|
| 342 |
+
"trial_params": null
|
| 343 |
+
}
|