End of training
Browse files- all_results.json +11 -0
- eval_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +415 -0
all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 2.353191375732422,
|
| 4 |
+
"eval_runtime": 253.6601,
|
| 5 |
+
"eval_samples_per_second": 195.541,
|
| 6 |
+
"eval_steps_per_second": 6.114,
|
| 7 |
+
"train_loss": 1.4714843301307519,
|
| 8 |
+
"train_runtime": 14279.912,
|
| 9 |
+
"train_samples_per_second": 73.765,
|
| 10 |
+
"train_steps_per_second": 2.305
|
| 11 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 2.353191375732422,
|
| 4 |
+
"eval_runtime": 253.6601,
|
| 5 |
+
"eval_samples_per_second": 195.541,
|
| 6 |
+
"eval_steps_per_second": 6.114
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.4714843301307519,
|
| 4 |
+
"train_runtime": 14279.912,
|
| 5 |
+
"train_samples_per_second": 73.765,
|
| 6 |
+
"train_steps_per_second": 2.305
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,415 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 32919,
|
| 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": 4.92405601628239e-05,
|
| 13 |
+
"loss": 3.2169,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.09,
|
| 18 |
+
"learning_rate": 4.848112032564781e-05,
|
| 19 |
+
"loss": 2.872,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.14,
|
| 24 |
+
"learning_rate": 4.77216804884717e-05,
|
| 25 |
+
"loss": 2.6524,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.18,
|
| 30 |
+
"learning_rate": 4.6962240651295605e-05,
|
| 31 |
+
"loss": 2.5352,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.23,
|
| 36 |
+
"learning_rate": 4.620280081411951e-05,
|
| 37 |
+
"loss": 2.4293,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.27,
|
| 42 |
+
"learning_rate": 4.544336097694341e-05,
|
| 43 |
+
"loss": 2.3569,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.32,
|
| 48 |
+
"learning_rate": 4.468392113976731e-05,
|
| 49 |
+
"loss": 2.2906,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.36,
|
| 54 |
+
"learning_rate": 4.392448130259121e-05,
|
| 55 |
+
"loss": 2.2189,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.41,
|
| 60 |
+
"learning_rate": 4.316504146541511e-05,
|
| 61 |
+
"loss": 2.1845,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.46,
|
| 66 |
+
"learning_rate": 4.240560162823901e-05,
|
| 67 |
+
"loss": 2.1222,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.5,
|
| 72 |
+
"learning_rate": 4.1646161791062915e-05,
|
| 73 |
+
"loss": 2.0704,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.55,
|
| 78 |
+
"learning_rate": 4.088672195388681e-05,
|
| 79 |
+
"loss": 2.0429,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.59,
|
| 84 |
+
"learning_rate": 4.012728211671072e-05,
|
| 85 |
+
"loss": 1.9951,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.64,
|
| 90 |
+
"learning_rate": 3.9367842279534616e-05,
|
| 91 |
+
"loss": 1.9488,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.68,
|
| 96 |
+
"learning_rate": 3.860840244235852e-05,
|
| 97 |
+
"loss": 1.9073,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.73,
|
| 102 |
+
"learning_rate": 3.784896260518242e-05,
|
| 103 |
+
"loss": 1.8649,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.77,
|
| 108 |
+
"learning_rate": 3.7089522768006316e-05,
|
| 109 |
+
"loss": 1.8362,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.82,
|
| 114 |
+
"learning_rate": 3.6330082930830225e-05,
|
| 115 |
+
"loss": 1.8117,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.87,
|
| 120 |
+
"learning_rate": 3.557064309365412e-05,
|
| 121 |
+
"loss": 1.7883,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.91,
|
| 126 |
+
"learning_rate": 3.4811203256478024e-05,
|
| 127 |
+
"loss": 1.7724,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.96,
|
| 132 |
+
"learning_rate": 3.4051763419301926e-05,
|
| 133 |
+
"loss": 1.7215,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.0,
|
| 138 |
+
"learning_rate": 3.329232358212583e-05,
|
| 139 |
+
"loss": 1.671,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.05,
|
| 144 |
+
"learning_rate": 3.2532883744949724e-05,
|
| 145 |
+
"loss": 1.542,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.09,
|
| 150 |
+
"learning_rate": 3.1773443907773626e-05,
|
| 151 |
+
"loss": 1.5211,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.14,
|
| 156 |
+
"learning_rate": 3.101400407059753e-05,
|
| 157 |
+
"loss": 1.537,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.18,
|
| 162 |
+
"learning_rate": 3.0254564233421428e-05,
|
| 163 |
+
"loss": 1.519,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.23,
|
| 168 |
+
"learning_rate": 2.9495124396245334e-05,
|
| 169 |
+
"loss": 1.4805,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.28,
|
| 174 |
+
"learning_rate": 2.8735684559069233e-05,
|
| 175 |
+
"loss": 1.467,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.32,
|
| 180 |
+
"learning_rate": 2.7976244721893135e-05,
|
| 181 |
+
"loss": 1.4591,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.37,
|
| 186 |
+
"learning_rate": 2.7216804884717034e-05,
|
| 187 |
+
"loss": 1.4131,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.41,
|
| 192 |
+
"learning_rate": 2.6457365047540933e-05,
|
| 193 |
+
"loss": 1.4217,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.46,
|
| 198 |
+
"learning_rate": 2.5697925210364836e-05,
|
| 199 |
+
"loss": 1.3952,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.5,
|
| 204 |
+
"learning_rate": 2.4938485373188738e-05,
|
| 205 |
+
"loss": 1.3792,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.55,
|
| 210 |
+
"learning_rate": 2.4179045536012637e-05,
|
| 211 |
+
"loss": 1.3583,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.59,
|
| 216 |
+
"learning_rate": 2.341960569883654e-05,
|
| 217 |
+
"loss": 1.3674,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.64,
|
| 222 |
+
"learning_rate": 2.266016586166044e-05,
|
| 223 |
+
"loss": 1.3277,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.69,
|
| 228 |
+
"learning_rate": 2.190072602448434e-05,
|
| 229 |
+
"loss": 1.2849,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.73,
|
| 234 |
+
"learning_rate": 2.1141286187308243e-05,
|
| 235 |
+
"loss": 1.2953,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.78,
|
| 240 |
+
"learning_rate": 2.0381846350132146e-05,
|
| 241 |
+
"loss": 1.2479,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.82,
|
| 246 |
+
"learning_rate": 1.9622406512956045e-05,
|
| 247 |
+
"loss": 1.2357,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.87,
|
| 252 |
+
"learning_rate": 1.8862966675779944e-05,
|
| 253 |
+
"loss": 1.2418,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.91,
|
| 258 |
+
"learning_rate": 1.8103526838603846e-05,
|
| 259 |
+
"loss": 1.2233,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.96,
|
| 264 |
+
"learning_rate": 1.734408700142775e-05,
|
| 265 |
+
"loss": 1.1791,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.0,
|
| 270 |
+
"learning_rate": 1.6584647164251648e-05,
|
| 271 |
+
"loss": 1.1664,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.05,
|
| 276 |
+
"learning_rate": 1.582520732707555e-05,
|
| 277 |
+
"loss": 0.9787,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.1,
|
| 282 |
+
"learning_rate": 1.5065767489899452e-05,
|
| 283 |
+
"loss": 0.9935,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.14,
|
| 288 |
+
"learning_rate": 1.4306327652723353e-05,
|
| 289 |
+
"loss": 0.9884,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.19,
|
| 294 |
+
"learning_rate": 1.3546887815547252e-05,
|
| 295 |
+
"loss": 0.9676,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.23,
|
| 300 |
+
"learning_rate": 1.2787447978371153e-05,
|
| 301 |
+
"loss": 0.9653,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.28,
|
| 306 |
+
"learning_rate": 1.2028008141195055e-05,
|
| 307 |
+
"loss": 0.9492,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.32,
|
| 312 |
+
"learning_rate": 1.1268568304018956e-05,
|
| 313 |
+
"loss": 0.9345,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.37,
|
| 318 |
+
"learning_rate": 1.0509128466842857e-05,
|
| 319 |
+
"loss": 0.9182,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.42,
|
| 324 |
+
"learning_rate": 9.749688629666759e-06,
|
| 325 |
+
"loss": 0.9127,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.46,
|
| 330 |
+
"learning_rate": 8.990248792490658e-06,
|
| 331 |
+
"loss": 0.9039,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.51,
|
| 336 |
+
"learning_rate": 8.23080895531456e-06,
|
| 337 |
+
"loss": 0.8962,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.55,
|
| 342 |
+
"learning_rate": 7.471369118138462e-06,
|
| 343 |
+
"loss": 0.8984,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.6,
|
| 348 |
+
"learning_rate": 6.711929280962362e-06,
|
| 349 |
+
"loss": 0.8718,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.64,
|
| 354 |
+
"learning_rate": 5.9524894437862635e-06,
|
| 355 |
+
"loss": 0.8662,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 2.69,
|
| 360 |
+
"learning_rate": 5.193049606610165e-06,
|
| 361 |
+
"loss": 0.864,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 2.73,
|
| 366 |
+
"learning_rate": 4.433609769434066e-06,
|
| 367 |
+
"loss": 0.8497,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.78,
|
| 372 |
+
"learning_rate": 3.674169932257967e-06,
|
| 373 |
+
"loss": 0.8242,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.83,
|
| 378 |
+
"learning_rate": 2.9147300950818677e-06,
|
| 379 |
+
"loss": 0.8302,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.87,
|
| 384 |
+
"learning_rate": 2.155290257905769e-06,
|
| 385 |
+
"loss": 0.7988,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 2.92,
|
| 390 |
+
"learning_rate": 1.3958504207296697e-06,
|
| 391 |
+
"loss": 0.8169,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.96,
|
| 396 |
+
"learning_rate": 6.364105835535709e-07,
|
| 397 |
+
"loss": 0.8069,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.0,
|
| 402 |
+
"step": 32919,
|
| 403 |
+
"total_flos": 1.4064216261264e+17,
|
| 404 |
+
"train_loss": 1.4714843301307519,
|
| 405 |
+
"train_runtime": 14279.912,
|
| 406 |
+
"train_samples_per_second": 73.765,
|
| 407 |
+
"train_steps_per_second": 2.305
|
| 408 |
+
}
|
| 409 |
+
],
|
| 410 |
+
"max_steps": 32919,
|
| 411 |
+
"num_train_epochs": 3,
|
| 412 |
+
"total_flos": 1.4064216261264e+17,
|
| 413 |
+
"trial_name": null,
|
| 414 |
+
"trial_params": null
|
| 415 |
+
}
|