Commit
•
c5fd10b
1
Parent(s):
83b7fd3
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +242 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.9872670257983565,
|
| 4 |
+
"train_runtime": 36788.8507,
|
| 5 |
+
"train_samples": 391,
|
| 6 |
+
"train_samples_per_second": 0.159,
|
| 7 |
+
"train_steps_per_second": 0.04
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.9872670257983565,
|
| 4 |
+
"train_runtime": 36788.8507,
|
| 5 |
+
"train_samples": 391,
|
| 6 |
+
"train_samples_per_second": 0.159,
|
| 7 |
+
"train_steps_per_second": 0.04
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,242 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"global_step": 1470,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 9e-07,
|
| 13 |
+
"loss": 2.4836,
|
| 14 |
+
"perplexity": 11.98433044208569,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"learning_rate": 9e-07,
|
| 20 |
+
"loss": 2.5819,
|
| 21 |
+
"perplexity": 13.222236558089785,
|
| 22 |
+
"step": 98
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.0,
|
| 26 |
+
"eval_runtime": 0.0033,
|
| 27 |
+
"eval_samples_per_second": 0.0,
|
| 28 |
+
"eval_steps_per_second": 0.0,
|
| 29 |
+
"step": 98
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.0,
|
| 33 |
+
"learning_rate": 9e-07,
|
| 34 |
+
"loss": 2.1794,
|
| 35 |
+
"perplexity": 8.841000066982355,
|
| 36 |
+
"step": 196
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 2.0,
|
| 40 |
+
"eval_runtime": 0.0031,
|
| 41 |
+
"eval_samples_per_second": 0.0,
|
| 42 |
+
"eval_steps_per_second": 0.0,
|
| 43 |
+
"step": 196
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"learning_rate": 9e-07,
|
| 48 |
+
"loss": 1.8523,
|
| 49 |
+
"perplexity": 6.374463942130527,
|
| 50 |
+
"step": 294
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"eval_runtime": 0.0024,
|
| 55 |
+
"eval_samples_per_second": 0.0,
|
| 56 |
+
"eval_steps_per_second": 0.0,
|
| 57 |
+
"step": 294
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"learning_rate": 9e-07,
|
| 62 |
+
"loss": 1.5639,
|
| 63 |
+
"perplexity": 4.777416885923455,
|
| 64 |
+
"step": 392
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_runtime": 0.0026,
|
| 69 |
+
"eval_samples_per_second": 0.0,
|
| 70 |
+
"eval_steps_per_second": 0.0,
|
| 71 |
+
"step": 392
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 5.0,
|
| 75 |
+
"learning_rate": 9e-07,
|
| 76 |
+
"loss": 1.3052,
|
| 77 |
+
"perplexity": 3.688426705282456,
|
| 78 |
+
"step": 490
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.0,
|
| 82 |
+
"eval_runtime": 0.0028,
|
| 83 |
+
"eval_samples_per_second": 0.0,
|
| 84 |
+
"eval_steps_per_second": 0.0,
|
| 85 |
+
"step": 490
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 6.0,
|
| 89 |
+
"learning_rate": 9e-07,
|
| 90 |
+
"loss": 1.0804,
|
| 91 |
+
"perplexity": 2.945857658491727,
|
| 92 |
+
"step": 588
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 6.0,
|
| 96 |
+
"eval_runtime": 0.0025,
|
| 97 |
+
"eval_samples_per_second": 0.0,
|
| 98 |
+
"eval_steps_per_second": 0.0,
|
| 99 |
+
"step": 588
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 7.0,
|
| 103 |
+
"learning_rate": 9e-07,
|
| 104 |
+
"loss": 0.8983,
|
| 105 |
+
"perplexity": 2.455425337981329,
|
| 106 |
+
"step": 686
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 7.0,
|
| 110 |
+
"eval_runtime": 0.0031,
|
| 111 |
+
"eval_samples_per_second": 0.0,
|
| 112 |
+
"eval_steps_per_second": 0.0,
|
| 113 |
+
"step": 686
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 8.0,
|
| 117 |
+
"learning_rate": 9e-07,
|
| 118 |
+
"loss": 0.7469,
|
| 119 |
+
"perplexity": 2.1104474782431386,
|
| 120 |
+
"step": 784
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_runtime": 0.0028,
|
| 125 |
+
"eval_samples_per_second": 0.0,
|
| 126 |
+
"eval_steps_per_second": 0.0,
|
| 127 |
+
"step": 784
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 9.0,
|
| 131 |
+
"learning_rate": 9e-07,
|
| 132 |
+
"loss": 0.6196,
|
| 133 |
+
"perplexity": 1.8581846193240203,
|
| 134 |
+
"step": 882
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 9.0,
|
| 138 |
+
"eval_runtime": 0.0024,
|
| 139 |
+
"eval_samples_per_second": 0.0,
|
| 140 |
+
"eval_steps_per_second": 0.0,
|
| 141 |
+
"step": 882
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.0,
|
| 145 |
+
"learning_rate": 9e-07,
|
| 146 |
+
"loss": 0.5154,
|
| 147 |
+
"perplexity": 1.6743080911071826,
|
| 148 |
+
"step": 980
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.0,
|
| 152 |
+
"eval_runtime": 0.0034,
|
| 153 |
+
"eval_samples_per_second": 0.0,
|
| 154 |
+
"eval_steps_per_second": 0.0,
|
| 155 |
+
"step": 980
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 11.0,
|
| 159 |
+
"learning_rate": 9e-07,
|
| 160 |
+
"loss": 0.4301,
|
| 161 |
+
"perplexity": 1.53741125698718,
|
| 162 |
+
"step": 1078
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 11.0,
|
| 166 |
+
"eval_runtime": 0.0026,
|
| 167 |
+
"eval_samples_per_second": 0.0,
|
| 168 |
+
"eval_steps_per_second": 0.0,
|
| 169 |
+
"step": 1078
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 12.0,
|
| 173 |
+
"learning_rate": 9e-07,
|
| 174 |
+
"loss": 0.3466,
|
| 175 |
+
"perplexity": 1.4142509118505286,
|
| 176 |
+
"step": 1176
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 12.0,
|
| 180 |
+
"eval_runtime": 0.0032,
|
| 181 |
+
"eval_samples_per_second": 0.0,
|
| 182 |
+
"eval_steps_per_second": 0.0,
|
| 183 |
+
"step": 1176
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 13.0,
|
| 187 |
+
"learning_rate": 9e-07,
|
| 188 |
+
"loss": 0.2822,
|
| 189 |
+
"perplexity": 1.3260439022481316,
|
| 190 |
+
"step": 1274
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 13.0,
|
| 194 |
+
"eval_runtime": 0.0031,
|
| 195 |
+
"eval_samples_per_second": 0.0,
|
| 196 |
+
"eval_steps_per_second": 0.0,
|
| 197 |
+
"step": 1274
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 14.0,
|
| 201 |
+
"learning_rate": 9e-07,
|
| 202 |
+
"loss": 0.2263,
|
| 203 |
+
"perplexity": 1.2539517943943168,
|
| 204 |
+
"step": 1372
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 14.0,
|
| 208 |
+
"eval_runtime": 0.0032,
|
| 209 |
+
"eval_samples_per_second": 0.0,
|
| 210 |
+
"eval_steps_per_second": 0.0,
|
| 211 |
+
"step": 1372
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 15.0,
|
| 215 |
+
"learning_rate": 9e-07,
|
| 216 |
+
"loss": 0.1815,
|
| 217 |
+
"perplexity": 1.1990145367097138,
|
| 218 |
+
"step": 1470
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 15.0,
|
| 222 |
+
"eval_runtime": 0.0022,
|
| 223 |
+
"eval_samples_per_second": 0.0,
|
| 224 |
+
"eval_steps_per_second": 0.0,
|
| 225 |
+
"step": 1470
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 15.0,
|
| 229 |
+
"step": 1470,
|
| 230 |
+
"total_flos": 49376758071296.0,
|
| 231 |
+
"train_loss": 0.9872670257983565,
|
| 232 |
+
"train_runtime": 36788.8507,
|
| 233 |
+
"train_samples_per_second": 0.159,
|
| 234 |
+
"train_steps_per_second": 0.04
|
| 235 |
+
}
|
| 236 |
+
],
|
| 237 |
+
"max_steps": 1470,
|
| 238 |
+
"num_train_epochs": 15,
|
| 239 |
+
"total_flos": 49376758071296.0,
|
| 240 |
+
"trial_name": null,
|
| 241 |
+
"trial_params": null
|
| 242 |
+
}
|