Commit
·
54ff6c4
1
Parent(s):
d6c50e2
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +250 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.8873453776041667,
|
| 4 |
+
"train_runtime": 59.0651,
|
| 5 |
+
"train_samples": 156,
|
| 6 |
+
"train_samples_per_second": 7.923,
|
| 7 |
+
"train_steps_per_second": 0.254
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.8873453776041667,
|
| 4 |
+
"train_runtime": 59.0651,
|
| 5 |
+
"train_samples": 156,
|
| 6 |
+
"train_samples_per_second": 7.923,
|
| 7 |
+
"train_steps_per_second": 0.254
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,250 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 15,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.2,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 2.3721,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"eval_accuracy": 0.06414321574844761,
|
| 19 |
+
"eval_loss": 2.21484375,
|
| 20 |
+
"eval_runtime": 1.2171,
|
| 21 |
+
"eval_samples_per_second": 23.828,
|
| 22 |
+
"eval_steps_per_second": 0.822,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.4,
|
| 27 |
+
"learning_rate": 3e-05,
|
| 28 |
+
"loss": 2.241,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.4,
|
| 33 |
+
"eval_accuracy": 0.06414321574844761,
|
| 34 |
+
"eval_loss": 2.21484375,
|
| 35 |
+
"eval_runtime": 1.292,
|
| 36 |
+
"eval_samples_per_second": 22.446,
|
| 37 |
+
"eval_steps_per_second": 0.774,
|
| 38 |
+
"step": 2
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6,
|
| 42 |
+
"learning_rate": 2.9672214011007087e-05,
|
| 43 |
+
"loss": 2.469,
|
| 44 |
+
"step": 3
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6,
|
| 48 |
+
"eval_accuracy": 0.06526621746597966,
|
| 49 |
+
"eval_loss": 2.134765625,
|
| 50 |
+
"eval_runtime": 1.0981,
|
| 51 |
+
"eval_samples_per_second": 26.409,
|
| 52 |
+
"eval_steps_per_second": 0.911,
|
| 53 |
+
"step": 3
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.8,
|
| 57 |
+
"learning_rate": 2.8703181864639013e-05,
|
| 58 |
+
"loss": 2.3735,
|
| 59 |
+
"step": 4
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.8,
|
| 63 |
+
"eval_accuracy": 0.06480380499405469,
|
| 64 |
+
"eval_loss": 2.130859375,
|
| 65 |
+
"eval_runtime": 0.8874,
|
| 66 |
+
"eval_samples_per_second": 32.681,
|
| 67 |
+
"eval_steps_per_second": 1.127,
|
| 68 |
+
"step": 4
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.0,
|
| 72 |
+
"learning_rate": 2.7135254915624213e-05,
|
| 73 |
+
"loss": 2.2755,
|
| 74 |
+
"step": 5
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.0,
|
| 78 |
+
"eval_accuracy": 0.06520015854141895,
|
| 79 |
+
"eval_loss": 2.11328125,
|
| 80 |
+
"eval_runtime": 1.0131,
|
| 81 |
+
"eval_samples_per_second": 28.625,
|
| 82 |
+
"eval_steps_per_second": 0.987,
|
| 83 |
+
"step": 5
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.2,
|
| 87 |
+
"learning_rate": 2.5036959095382875e-05,
|
| 88 |
+
"loss": 2.0428,
|
| 89 |
+
"step": 6
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.2,
|
| 93 |
+
"eval_accuracy": 0.06586074778702602,
|
| 94 |
+
"eval_loss": 2.09375,
|
| 95 |
+
"eval_runtime": 0.9993,
|
| 96 |
+
"eval_samples_per_second": 29.021,
|
| 97 |
+
"eval_steps_per_second": 1.001,
|
| 98 |
+
"step": 6
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.4,
|
| 102 |
+
"learning_rate": 2.25e-05,
|
| 103 |
+
"loss": 1.764,
|
| 104 |
+
"step": 7
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.4,
|
| 108 |
+
"eval_accuracy": 0.06592680671158674,
|
| 109 |
+
"eval_loss": 2.078125,
|
| 110 |
+
"eval_runtime": 0.6991,
|
| 111 |
+
"eval_samples_per_second": 41.485,
|
| 112 |
+
"eval_steps_per_second": 1.431,
|
| 113 |
+
"step": 7
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.6,
|
| 117 |
+
"learning_rate": 1.963525491562421e-05,
|
| 118 |
+
"loss": 1.7458,
|
| 119 |
+
"step": 8
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.6,
|
| 123 |
+
"eval_accuracy": 0.06605892456070815,
|
| 124 |
+
"eval_loss": 2.078125,
|
| 125 |
+
"eval_runtime": 0.9029,
|
| 126 |
+
"eval_samples_per_second": 32.119,
|
| 127 |
+
"eval_steps_per_second": 1.108,
|
| 128 |
+
"step": 8
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.8,
|
| 132 |
+
"learning_rate": 1.6567926949014805e-05,
|
| 133 |
+
"loss": 1.868,
|
| 134 |
+
"step": 9
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.8,
|
| 138 |
+
"eval_accuracy": 0.06599286563614744,
|
| 139 |
+
"eval_loss": 2.08203125,
|
| 140 |
+
"eval_runtime": 1.2995,
|
| 141 |
+
"eval_samples_per_second": 22.316,
|
| 142 |
+
"eval_steps_per_second": 0.77,
|
| 143 |
+
"step": 9
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.0,
|
| 147 |
+
"learning_rate": 1.3432073050985201e-05,
|
| 148 |
+
"loss": 1.9548,
|
| 149 |
+
"step": 10
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.0,
|
| 153 |
+
"eval_accuracy": 0.06632316025895098,
|
| 154 |
+
"eval_loss": 2.0703125,
|
| 155 |
+
"eval_runtime": 1.0961,
|
| 156 |
+
"eval_samples_per_second": 26.458,
|
| 157 |
+
"eval_steps_per_second": 0.912,
|
| 158 |
+
"step": 10
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.2,
|
| 162 |
+
"learning_rate": 1.036474508437579e-05,
|
| 163 |
+
"loss": 1.6772,
|
| 164 |
+
"step": 11
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.2,
|
| 168 |
+
"eval_accuracy": 0.0664552781080724,
|
| 169 |
+
"eval_loss": 2.060546875,
|
| 170 |
+
"eval_runtime": 1.0956,
|
| 171 |
+
"eval_samples_per_second": 26.469,
|
| 172 |
+
"eval_steps_per_second": 0.913,
|
| 173 |
+
"step": 11
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.4,
|
| 177 |
+
"learning_rate": 7.500000000000004e-06,
|
| 178 |
+
"loss": 1.3997,
|
| 179 |
+
"step": 12
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.4,
|
| 183 |
+
"eval_accuracy": 0.06678557273087594,
|
| 184 |
+
"eval_loss": 2.056640625,
|
| 185 |
+
"eval_runtime": 1.3048,
|
| 186 |
+
"eval_samples_per_second": 22.226,
|
| 187 |
+
"eval_steps_per_second": 0.766,
|
| 188 |
+
"step": 12
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.6,
|
| 192 |
+
"learning_rate": 4.963040904617131e-06,
|
| 193 |
+
"loss": 1.3717,
|
| 194 |
+
"step": 13
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.6,
|
| 198 |
+
"eval_accuracy": 0.06691769057999736,
|
| 199 |
+
"eval_loss": 2.0546875,
|
| 200 |
+
"eval_runtime": 1.0974,
|
| 201 |
+
"eval_samples_per_second": 26.426,
|
| 202 |
+
"eval_steps_per_second": 0.911,
|
| 203 |
+
"step": 13
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.8,
|
| 207 |
+
"learning_rate": 2.86474508437579e-06,
|
| 208 |
+
"loss": 1.5284,
|
| 209 |
+
"step": 14
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.8,
|
| 213 |
+
"eval_accuracy": 0.06671951380631523,
|
| 214 |
+
"eval_loss": 2.0546875,
|
| 215 |
+
"eval_runtime": 1.2969,
|
| 216 |
+
"eval_samples_per_second": 22.361,
|
| 217 |
+
"eval_steps_per_second": 0.771,
|
| 218 |
+
"step": 14
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.0,
|
| 222 |
+
"learning_rate": 1.2968181353609854e-06,
|
| 223 |
+
"loss": 1.2264,
|
| 224 |
+
"step": 15
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.0,
|
| 228 |
+
"eval_accuracy": 0.06691769057999736,
|
| 229 |
+
"eval_loss": 2.056640625,
|
| 230 |
+
"eval_runtime": 0.8003,
|
| 231 |
+
"eval_samples_per_second": 36.234,
|
| 232 |
+
"eval_steps_per_second": 1.249,
|
| 233 |
+
"step": 15
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.0,
|
| 237 |
+
"step": 15,
|
| 238 |
+
"total_flos": 1415705001984.0,
|
| 239 |
+
"train_loss": 1.8873453776041667,
|
| 240 |
+
"train_runtime": 59.0651,
|
| 241 |
+
"train_samples_per_second": 7.923,
|
| 242 |
+
"train_steps_per_second": 0.254
|
| 243 |
+
}
|
| 244 |
+
],
|
| 245 |
+
"max_steps": 15,
|
| 246 |
+
"num_train_epochs": 3,
|
| 247 |
+
"total_flos": 1415705001984.0,
|
| 248 |
+
"trial_name": null,
|
| 249 |
+
"trial_params": null
|
| 250 |
+
}
|