End of training
Browse files- all_results.json +11 -0
- eval_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +355 -0
all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 1.6998499631881714,
|
| 4 |
+
"eval_runtime": 78.0149,
|
| 5 |
+
"eval_samples_per_second": 320.631,
|
| 6 |
+
"eval_steps_per_second": 5.012,
|
| 7 |
+
"train_loss": 0.395306228830761,
|
| 8 |
+
"train_runtime": 12286.1682,
|
| 9 |
+
"train_samples_per_second": 144.492,
|
| 10 |
+
"train_steps_per_second": 2.258
|
| 11 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 1.6998499631881714,
|
| 4 |
+
"eval_runtime": 78.0149,
|
| 5 |
+
"eval_samples_per_second": 320.631,
|
| 6 |
+
"eval_steps_per_second": 5.012
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.395306228830761,
|
| 4 |
+
"train_runtime": 12286.1682,
|
| 5 |
+
"train_samples_per_second": 144.492,
|
| 6 |
+
"train_steps_per_second": 2.258
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,355 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 27741,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.05,
|
| 12 |
+
"learning_rate": 4.909880682022999e-05,
|
| 13 |
+
"loss": 2.0352,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.11,
|
| 18 |
+
"learning_rate": 4.819761364045997e-05,
|
| 19 |
+
"loss": 1.0236,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.16,
|
| 24 |
+
"learning_rate": 4.7296420460689956e-05,
|
| 25 |
+
"loss": 0.8962,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.22,
|
| 30 |
+
"learning_rate": 4.639522728091994e-05,
|
| 31 |
+
"loss": 0.8171,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.27,
|
| 36 |
+
"learning_rate": 4.5494034101149925e-05,
|
| 37 |
+
"loss": 0.7493,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.32,
|
| 42 |
+
"learning_rate": 4.459284092137991e-05,
|
| 43 |
+
"loss": 0.7156,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.38,
|
| 48 |
+
"learning_rate": 4.3691647741609894e-05,
|
| 49 |
+
"loss": 0.6699,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.43,
|
| 54 |
+
"learning_rate": 4.279045456183988e-05,
|
| 55 |
+
"loss": 0.6495,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.49,
|
| 60 |
+
"learning_rate": 4.188926138206986e-05,
|
| 61 |
+
"loss": 0.6145,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.54,
|
| 66 |
+
"learning_rate": 4.098806820229985e-05,
|
| 67 |
+
"loss": 0.5858,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.59,
|
| 72 |
+
"learning_rate": 4.008687502252983e-05,
|
| 73 |
+
"loss": 0.5629,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.65,
|
| 78 |
+
"learning_rate": 3.9185681842759816e-05,
|
| 79 |
+
"loss": 0.5392,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.7,
|
| 84 |
+
"learning_rate": 3.82844886629898e-05,
|
| 85 |
+
"loss": 0.5323,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.76,
|
| 90 |
+
"learning_rate": 3.7383295483219785e-05,
|
| 91 |
+
"loss": 0.5048,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.81,
|
| 96 |
+
"learning_rate": 3.648210230344977e-05,
|
| 97 |
+
"loss": 0.4895,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.87,
|
| 102 |
+
"learning_rate": 3.558090912367975e-05,
|
| 103 |
+
"loss": 0.4698,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.92,
|
| 108 |
+
"learning_rate": 3.467971594390974e-05,
|
| 109 |
+
"loss": 0.4545,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.97,
|
| 114 |
+
"learning_rate": 3.3778522764139723e-05,
|
| 115 |
+
"loss": 0.4527,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.03,
|
| 120 |
+
"learning_rate": 3.287732958436971e-05,
|
| 121 |
+
"loss": 0.3951,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.08,
|
| 126 |
+
"learning_rate": 3.197613640459969e-05,
|
| 127 |
+
"loss": 0.3576,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.14,
|
| 132 |
+
"learning_rate": 3.107494322482968e-05,
|
| 133 |
+
"loss": 0.3595,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.19,
|
| 138 |
+
"learning_rate": 3.017375004505966e-05,
|
| 139 |
+
"loss": 0.346,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.24,
|
| 144 |
+
"learning_rate": 2.9272556865289646e-05,
|
| 145 |
+
"loss": 0.3441,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.3,
|
| 150 |
+
"learning_rate": 2.837136368551963e-05,
|
| 151 |
+
"loss": 0.333,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.35,
|
| 156 |
+
"learning_rate": 2.7470170505749615e-05,
|
| 157 |
+
"loss": 0.3288,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.41,
|
| 162 |
+
"learning_rate": 2.65689773259796e-05,
|
| 163 |
+
"loss": 0.3256,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.46,
|
| 168 |
+
"learning_rate": 2.5667784146209584e-05,
|
| 169 |
+
"loss": 0.3191,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.51,
|
| 174 |
+
"learning_rate": 2.476659096643957e-05,
|
| 175 |
+
"loss": 0.3123,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.57,
|
| 180 |
+
"learning_rate": 2.3865397786669553e-05,
|
| 181 |
+
"loss": 0.3036,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.62,
|
| 186 |
+
"learning_rate": 2.2964204606899537e-05,
|
| 187 |
+
"loss": 0.2941,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.68,
|
| 192 |
+
"learning_rate": 2.2063011427129522e-05,
|
| 193 |
+
"loss": 0.2934,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.73,
|
| 198 |
+
"learning_rate": 2.1161818247359506e-05,
|
| 199 |
+
"loss": 0.2858,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.78,
|
| 204 |
+
"learning_rate": 2.026062506758949e-05,
|
| 205 |
+
"loss": 0.2779,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.84,
|
| 210 |
+
"learning_rate": 1.9359431887819472e-05,
|
| 211 |
+
"loss": 0.2736,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.89,
|
| 216 |
+
"learning_rate": 1.8458238708049457e-05,
|
| 217 |
+
"loss": 0.2674,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.95,
|
| 222 |
+
"learning_rate": 1.755704552827944e-05,
|
| 223 |
+
"loss": 0.2601,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.0,
|
| 228 |
+
"learning_rate": 1.6655852348509426e-05,
|
| 229 |
+
"loss": 0.2507,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.05,
|
| 234 |
+
"learning_rate": 1.575465916873941e-05,
|
| 235 |
+
"loss": 0.2024,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.11,
|
| 240 |
+
"learning_rate": 1.4853465988969395e-05,
|
| 241 |
+
"loss": 0.1933,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.16,
|
| 246 |
+
"learning_rate": 1.3952272809199379e-05,
|
| 247 |
+
"loss": 0.1951,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.22,
|
| 252 |
+
"learning_rate": 1.3051079629429367e-05,
|
| 253 |
+
"loss": 0.19,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.27,
|
| 258 |
+
"learning_rate": 1.214988644965935e-05,
|
| 259 |
+
"loss": 0.1939,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.33,
|
| 264 |
+
"learning_rate": 1.1248693269889334e-05,
|
| 265 |
+
"loss": 0.1842,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.38,
|
| 270 |
+
"learning_rate": 1.0347500090119317e-05,
|
| 271 |
+
"loss": 0.1852,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.43,
|
| 276 |
+
"learning_rate": 9.446306910349303e-06,
|
| 277 |
+
"loss": 0.1792,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.49,
|
| 282 |
+
"learning_rate": 8.545113730579288e-06,
|
| 283 |
+
"loss": 0.1793,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.54,
|
| 288 |
+
"learning_rate": 7.643920550809272e-06,
|
| 289 |
+
"loss": 0.1758,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.6,
|
| 294 |
+
"learning_rate": 6.742727371039257e-06,
|
| 295 |
+
"loss": 0.1698,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.65,
|
| 300 |
+
"learning_rate": 5.84153419126924e-06,
|
| 301 |
+
"loss": 0.1674,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.7,
|
| 306 |
+
"learning_rate": 4.940341011499225e-06,
|
| 307 |
+
"loss": 0.1635,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.76,
|
| 312 |
+
"learning_rate": 4.03914783172921e-06,
|
| 313 |
+
"loss": 0.1583,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.81,
|
| 318 |
+
"learning_rate": 3.1379546519591943e-06,
|
| 319 |
+
"loss": 0.1613,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.87,
|
| 324 |
+
"learning_rate": 2.2367614721891784e-06,
|
| 325 |
+
"loss": 0.1548,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.92,
|
| 330 |
+
"learning_rate": 1.335568292419163e-06,
|
| 331 |
+
"loss": 0.1547,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.97,
|
| 336 |
+
"learning_rate": 4.3437511264914753e-07,
|
| 337 |
+
"loss": 0.1597,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.0,
|
| 342 |
+
"step": 27741,
|
| 343 |
+
"total_flos": 2.3702775297552e+17,
|
| 344 |
+
"train_loss": 0.395306228830761,
|
| 345 |
+
"train_runtime": 12286.1682,
|
| 346 |
+
"train_samples_per_second": 144.492,
|
| 347 |
+
"train_steps_per_second": 2.258
|
| 348 |
+
}
|
| 349 |
+
],
|
| 350 |
+
"max_steps": 27741,
|
| 351 |
+
"num_train_epochs": 3,
|
| 352 |
+
"total_flos": 2.3702775297552e+17,
|
| 353 |
+
"trial_name": null,
|
| 354 |
+
"trial_params": null
|
| 355 |
+
}
|