Tiago Barbosa de Lima
commited on
Commit
•
ac27b2e
1
Parent(s):
60727ef
End of training
Browse files- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +351 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_bleu": 90.1821,
|
| 4 |
+
"eval_gen_len": 45.9318,
|
| 5 |
+
"eval_loss": 0.043235890567302704,
|
| 6 |
+
"eval_runtime": 489.576,
|
| 7 |
+
"eval_samples": 1041,
|
| 8 |
+
"eval_samples_per_second": 2.126,
|
| 9 |
+
"eval_steps_per_second": 1.064,
|
| 10 |
+
"train_loss": 0.047856709064706696,
|
| 11 |
+
"train_runtime": 5577.5722,
|
| 12 |
+
"train_samples": 9371,
|
| 13 |
+
"train_samples_per_second": 8.401,
|
| 14 |
+
"train_steps_per_second": 4.201
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_bleu": 90.1821,
|
| 4 |
+
"eval_gen_len": 45.9318,
|
| 5 |
+
"eval_loss": 0.043235890567302704,
|
| 6 |
+
"eval_runtime": 489.576,
|
| 7 |
+
"eval_samples": 1041,
|
| 8 |
+
"eval_samples_per_second": 2.126,
|
| 9 |
+
"eval_steps_per_second": 1.064
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.047856709064706696,
|
| 4 |
+
"train_runtime": 5577.5722,
|
| 5 |
+
"train_samples": 9371,
|
| 6 |
+
"train_samples_per_second": 8.401,
|
| 7 |
+
"train_steps_per_second": 4.201
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,351 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 23430,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.11,
|
| 12 |
+
"learning_rate": 4.893299189073837e-05,
|
| 13 |
+
"loss": 0.7701,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.21,
|
| 18 |
+
"learning_rate": 4.786598378147674e-05,
|
| 19 |
+
"loss": 0.1282,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.32,
|
| 24 |
+
"learning_rate": 4.6798975672215114e-05,
|
| 25 |
+
"loss": 0.0914,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.43,
|
| 30 |
+
"learning_rate": 4.5731967562953484e-05,
|
| 31 |
+
"loss": 0.0791,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.53,
|
| 36 |
+
"learning_rate": 4.4664959453691854e-05,
|
| 37 |
+
"loss": 0.0656,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.64,
|
| 42 |
+
"learning_rate": 4.359795134443022e-05,
|
| 43 |
+
"loss": 0.0645,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.75,
|
| 48 |
+
"learning_rate": 4.253094323516859e-05,
|
| 49 |
+
"loss": 0.0573,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.85,
|
| 54 |
+
"learning_rate": 4.146393512590696e-05,
|
| 55 |
+
"loss": 0.0566,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.96,
|
| 60 |
+
"learning_rate": 4.039692701664533e-05,
|
| 61 |
+
"loss": 0.0544,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.0,
|
| 66 |
+
"eval_bleu": 25.989,
|
| 67 |
+
"eval_gen_len": 18.8655,
|
| 68 |
+
"eval_loss": 0.044239919632673264,
|
| 69 |
+
"eval_runtime": 180.9818,
|
| 70 |
+
"eval_samples_per_second": 5.752,
|
| 71 |
+
"eval_steps_per_second": 2.879,
|
| 72 |
+
"step": 4686
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.07,
|
| 76 |
+
"learning_rate": 3.93299189073837e-05,
|
| 77 |
+
"loss": 0.0476,
|
| 78 |
+
"step": 5000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.17,
|
| 82 |
+
"learning_rate": 3.826291079812207e-05,
|
| 83 |
+
"loss": 0.0442,
|
| 84 |
+
"step": 5500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.28,
|
| 88 |
+
"learning_rate": 3.719590268886043e-05,
|
| 89 |
+
"loss": 0.0375,
|
| 90 |
+
"step": 6000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.39,
|
| 94 |
+
"learning_rate": 3.6128894579598804e-05,
|
| 95 |
+
"loss": 0.0408,
|
| 96 |
+
"step": 6500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.49,
|
| 100 |
+
"learning_rate": 3.5061886470337174e-05,
|
| 101 |
+
"loss": 0.0351,
|
| 102 |
+
"step": 7000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.6,
|
| 106 |
+
"learning_rate": 3.3994878361075544e-05,
|
| 107 |
+
"loss": 0.0355,
|
| 108 |
+
"step": 7500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.71,
|
| 112 |
+
"learning_rate": 3.2927870251813915e-05,
|
| 113 |
+
"loss": 0.0379,
|
| 114 |
+
"step": 8000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.81,
|
| 118 |
+
"learning_rate": 3.1860862142552285e-05,
|
| 119 |
+
"loss": 0.0331,
|
| 120 |
+
"step": 8500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.92,
|
| 124 |
+
"learning_rate": 3.0793854033290656e-05,
|
| 125 |
+
"loss": 0.0367,
|
| 126 |
+
"step": 9000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.0,
|
| 130 |
+
"eval_bleu": 25.9358,
|
| 131 |
+
"eval_gen_len": 18.8713,
|
| 132 |
+
"eval_loss": 0.03710315003991127,
|
| 133 |
+
"eval_runtime": 179.552,
|
| 134 |
+
"eval_samples_per_second": 5.798,
|
| 135 |
+
"eval_steps_per_second": 2.902,
|
| 136 |
+
"step": 9372
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.03,
|
| 140 |
+
"learning_rate": 2.9726845924029023e-05,
|
| 141 |
+
"loss": 0.0307,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.13,
|
| 146 |
+
"learning_rate": 2.8659837814767393e-05,
|
| 147 |
+
"loss": 0.0218,
|
| 148 |
+
"step": 10000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 2.24,
|
| 152 |
+
"learning_rate": 2.7592829705505763e-05,
|
| 153 |
+
"loss": 0.0234,
|
| 154 |
+
"step": 10500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.35,
|
| 158 |
+
"learning_rate": 2.6525821596244134e-05,
|
| 159 |
+
"loss": 0.0265,
|
| 160 |
+
"step": 11000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.45,
|
| 164 |
+
"learning_rate": 2.54588134869825e-05,
|
| 165 |
+
"loss": 0.0239,
|
| 166 |
+
"step": 11500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.56,
|
| 170 |
+
"learning_rate": 2.439180537772087e-05,
|
| 171 |
+
"loss": 0.0261,
|
| 172 |
+
"step": 12000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.67,
|
| 176 |
+
"learning_rate": 2.332479726845924e-05,
|
| 177 |
+
"loss": 0.0247,
|
| 178 |
+
"step": 12500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 2.77,
|
| 182 |
+
"learning_rate": 2.2257789159197612e-05,
|
| 183 |
+
"loss": 0.0228,
|
| 184 |
+
"step": 13000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.88,
|
| 188 |
+
"learning_rate": 2.1190781049935982e-05,
|
| 189 |
+
"loss": 0.0233,
|
| 190 |
+
"step": 13500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.99,
|
| 194 |
+
"learning_rate": 2.0123772940674353e-05,
|
| 195 |
+
"loss": 0.0222,
|
| 196 |
+
"step": 14000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.0,
|
| 200 |
+
"eval_bleu": 25.8976,
|
| 201 |
+
"eval_gen_len": 18.8694,
|
| 202 |
+
"eval_loss": 0.037382081151008606,
|
| 203 |
+
"eval_runtime": 178.2936,
|
| 204 |
+
"eval_samples_per_second": 5.839,
|
| 205 |
+
"eval_steps_per_second": 2.922,
|
| 206 |
+
"step": 14058
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.09,
|
| 210 |
+
"learning_rate": 1.905676483141272e-05,
|
| 211 |
+
"loss": 0.0156,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.2,
|
| 216 |
+
"learning_rate": 1.7989756722151087e-05,
|
| 217 |
+
"loss": 0.0153,
|
| 218 |
+
"step": 15000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.31,
|
| 222 |
+
"learning_rate": 1.6922748612889457e-05,
|
| 223 |
+
"loss": 0.0174,
|
| 224 |
+
"step": 15500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.41,
|
| 228 |
+
"learning_rate": 1.5855740503627827e-05,
|
| 229 |
+
"loss": 0.0171,
|
| 230 |
+
"step": 16000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.52,
|
| 234 |
+
"learning_rate": 1.4788732394366198e-05,
|
| 235 |
+
"loss": 0.0176,
|
| 236 |
+
"step": 16500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.63,
|
| 240 |
+
"learning_rate": 1.3721724285104568e-05,
|
| 241 |
+
"loss": 0.0187,
|
| 242 |
+
"step": 17000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 3.73,
|
| 246 |
+
"learning_rate": 1.2654716175842937e-05,
|
| 247 |
+
"loss": 0.0157,
|
| 248 |
+
"step": 17500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.84,
|
| 252 |
+
"learning_rate": 1.1587708066581307e-05,
|
| 253 |
+
"loss": 0.0169,
|
| 254 |
+
"step": 18000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.95,
|
| 258 |
+
"learning_rate": 1.0520699957319676e-05,
|
| 259 |
+
"loss": 0.0152,
|
| 260 |
+
"step": 18500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.0,
|
| 264 |
+
"eval_bleu": 26.1575,
|
| 265 |
+
"eval_gen_len": 18.8694,
|
| 266 |
+
"eval_loss": 0.040931396186351776,
|
| 267 |
+
"eval_runtime": 182.3446,
|
| 268 |
+
"eval_samples_per_second": 5.709,
|
| 269 |
+
"eval_steps_per_second": 2.857,
|
| 270 |
+
"step": 18744
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 4.05,
|
| 274 |
+
"learning_rate": 9.453691848058044e-06,
|
| 275 |
+
"loss": 0.0148,
|
| 276 |
+
"step": 19000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.16,
|
| 280 |
+
"learning_rate": 8.386683738796415e-06,
|
| 281 |
+
"loss": 0.0124,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.27,
|
| 286 |
+
"learning_rate": 7.319675629534785e-06,
|
| 287 |
+
"loss": 0.0148,
|
| 288 |
+
"step": 20000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.37,
|
| 292 |
+
"learning_rate": 6.252667520273155e-06,
|
| 293 |
+
"loss": 0.0123,
|
| 294 |
+
"step": 20500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 4.48,
|
| 298 |
+
"learning_rate": 5.1856594110115235e-06,
|
| 299 |
+
"loss": 0.0139,
|
| 300 |
+
"step": 21000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.59,
|
| 304 |
+
"learning_rate": 4.118651301749894e-06,
|
| 305 |
+
"loss": 0.0116,
|
| 306 |
+
"step": 21500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 4.69,
|
| 310 |
+
"learning_rate": 3.051643192488263e-06,
|
| 311 |
+
"loss": 0.014,
|
| 312 |
+
"step": 22000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 4.8,
|
| 316 |
+
"learning_rate": 1.9846350832266325e-06,
|
| 317 |
+
"loss": 0.0143,
|
| 318 |
+
"step": 22500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.91,
|
| 322 |
+
"learning_rate": 9.176269739650021e-07,
|
| 323 |
+
"loss": 0.0147,
|
| 324 |
+
"step": 23000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.0,
|
| 328 |
+
"eval_bleu": 26.0973,
|
| 329 |
+
"eval_gen_len": 18.8694,
|
| 330 |
+
"eval_loss": 0.043235890567302704,
|
| 331 |
+
"eval_runtime": 184.0328,
|
| 332 |
+
"eval_samples_per_second": 5.657,
|
| 333 |
+
"eval_steps_per_second": 2.831,
|
| 334 |
+
"step": 23430
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 5.0,
|
| 338 |
+
"step": 23430,
|
| 339 |
+
"total_flos": 2019762272885760.0,
|
| 340 |
+
"train_loss": 0.047856709064706696,
|
| 341 |
+
"train_runtime": 5577.5722,
|
| 342 |
+
"train_samples_per_second": 8.401,
|
| 343 |
+
"train_steps_per_second": 4.201
|
| 344 |
+
}
|
| 345 |
+
],
|
| 346 |
+
"max_steps": 23430,
|
| 347 |
+
"num_train_epochs": 5,
|
| 348 |
+
"total_flos": 2019762272885760.0,
|
| 349 |
+
"trial_name": null,
|
| 350 |
+
"trial_params": null
|
| 351 |
+
}
|