Commit
•
7736669
1
Parent(s):
188adc4
End of training
Browse files- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +230 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"train_loss": 1.7074453571236545,
|
| 4 |
+
"train_runtime": 28108.9431,
|
| 5 |
+
"train_samples_per_second": 28.461,
|
| 6 |
+
"train_steps_per_second": 0.222
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"train_loss": 1.7074453571236545,
|
| 4 |
+
"train_runtime": 28108.9431,
|
| 5 |
+
"train_samples_per_second": 28.461,
|
| 6 |
+
"train_steps_per_second": 0.222
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,230 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.99744,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 6248,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.64,
|
| 13 |
+
"learning_rate": 1.8399487836107557e-05,
|
| 14 |
+
"loss": 2.1501,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"eval_f1": 0.9072,
|
| 20 |
+
"eval_gen_len": 25.465454545454545,
|
| 21 |
+
"eval_loss": 1.7061578035354614,
|
| 22 |
+
"eval_precision": 0.9082,
|
| 23 |
+
"eval_recall": 0.9065,
|
| 24 |
+
"eval_rouge1": 0.4566,
|
| 25 |
+
"eval_rouge2": 0.209,
|
| 26 |
+
"eval_rougeL": 0.3745,
|
| 27 |
+
"eval_rougeLsum": 0.3744,
|
| 28 |
+
"eval_runtime": 882.5908,
|
| 29 |
+
"eval_samples_per_second": 6.232,
|
| 30 |
+
"eval_steps_per_second": 0.39,
|
| 31 |
+
"step": 781
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.28,
|
| 35 |
+
"learning_rate": 1.679897567221511e-05,
|
| 36 |
+
"loss": 1.839,
|
| 37 |
+
"step": 1000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.92,
|
| 41 |
+
"learning_rate": 1.5198463508322665e-05,
|
| 42 |
+
"loss": 1.7722,
|
| 43 |
+
"step": 1500
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"eval_f1": 0.9097,
|
| 48 |
+
"eval_gen_len": 25.42981818181818,
|
| 49 |
+
"eval_loss": 1.6313680410385132,
|
| 50 |
+
"eval_precision": 0.9107,
|
| 51 |
+
"eval_recall": 0.909,
|
| 52 |
+
"eval_rouge1": 0.4712,
|
| 53 |
+
"eval_rouge2": 0.2226,
|
| 54 |
+
"eval_rougeL": 0.3906,
|
| 55 |
+
"eval_rougeLsum": 0.3904,
|
| 56 |
+
"eval_runtime": 869.6253,
|
| 57 |
+
"eval_samples_per_second": 6.325,
|
| 58 |
+
"eval_steps_per_second": 0.396,
|
| 59 |
+
"step": 1562
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.56,
|
| 63 |
+
"learning_rate": 1.3597951344430219e-05,
|
| 64 |
+
"loss": 1.7218,
|
| 65 |
+
"step": 2000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 3.0,
|
| 69 |
+
"eval_f1": 0.9106,
|
| 70 |
+
"eval_gen_len": 25.656909090909092,
|
| 71 |
+
"eval_loss": 1.594840168952942,
|
| 72 |
+
"eval_precision": 0.9112,
|
| 73 |
+
"eval_recall": 0.9103,
|
| 74 |
+
"eval_rouge1": 0.4776,
|
| 75 |
+
"eval_rouge2": 0.2284,
|
| 76 |
+
"eval_rougeL": 0.3965,
|
| 77 |
+
"eval_rougeLsum": 0.3963,
|
| 78 |
+
"eval_runtime": 866.3975,
|
| 79 |
+
"eval_samples_per_second": 6.348,
|
| 80 |
+
"eval_steps_per_second": 0.397,
|
| 81 |
+
"step": 2343
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 3.2,
|
| 85 |
+
"learning_rate": 1.1997439180537773e-05,
|
| 86 |
+
"loss": 1.6923,
|
| 87 |
+
"step": 2500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 3.84,
|
| 91 |
+
"learning_rate": 1.0396927016645328e-05,
|
| 92 |
+
"loss": 1.6668,
|
| 93 |
+
"step": 3000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"eval_f1": 0.9112,
|
| 98 |
+
"eval_gen_len": 25.34509090909091,
|
| 99 |
+
"eval_loss": 1.5707985162734985,
|
| 100 |
+
"eval_precision": 0.9122,
|
| 101 |
+
"eval_recall": 0.9107,
|
| 102 |
+
"eval_rouge1": 0.481,
|
| 103 |
+
"eval_rouge2": 0.2316,
|
| 104 |
+
"eval_rougeL": 0.4002,
|
| 105 |
+
"eval_rougeLsum": 0.4,
|
| 106 |
+
"eval_runtime": 864.0634,
|
| 107 |
+
"eval_samples_per_second": 6.365,
|
| 108 |
+
"eval_steps_per_second": 0.398,
|
| 109 |
+
"step": 3125
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 4.48,
|
| 113 |
+
"learning_rate": 8.796414852752882e-06,
|
| 114 |
+
"loss": 1.6437,
|
| 115 |
+
"step": 3500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 5.0,
|
| 119 |
+
"eval_f1": 0.9118,
|
| 120 |
+
"eval_gen_len": 25.482,
|
| 121 |
+
"eval_loss": 1.5564521551132202,
|
| 122 |
+
"eval_precision": 0.9127,
|
| 123 |
+
"eval_recall": 0.9113,
|
| 124 |
+
"eval_rouge1": 0.4844,
|
| 125 |
+
"eval_rouge2": 0.2346,
|
| 126 |
+
"eval_rougeL": 0.4034,
|
| 127 |
+
"eval_rougeLsum": 0.4031,
|
| 128 |
+
"eval_runtime": 868.5536,
|
| 129 |
+
"eval_samples_per_second": 6.332,
|
| 130 |
+
"eval_steps_per_second": 0.396,
|
| 131 |
+
"step": 3906
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.12,
|
| 135 |
+
"learning_rate": 7.195902688860435e-06,
|
| 136 |
+
"loss": 1.6338,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.76,
|
| 141 |
+
"learning_rate": 5.595390524967991e-06,
|
| 142 |
+
"loss": 1.6186,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"eval_f1": 0.912,
|
| 148 |
+
"eval_gen_len": 25.419090909090908,
|
| 149 |
+
"eval_loss": 1.5476245880126953,
|
| 150 |
+
"eval_precision": 0.9129,
|
| 151 |
+
"eval_recall": 0.9115,
|
| 152 |
+
"eval_rouge1": 0.4852,
|
| 153 |
+
"eval_rouge2": 0.236,
|
| 154 |
+
"eval_rougeL": 0.4047,
|
| 155 |
+
"eval_rougeLsum": 0.4044,
|
| 156 |
+
"eval_runtime": 867.5381,
|
| 157 |
+
"eval_samples_per_second": 6.34,
|
| 158 |
+
"eval_steps_per_second": 0.397,
|
| 159 |
+
"step": 4687
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.4,
|
| 163 |
+
"learning_rate": 3.9948783610755446e-06,
|
| 164 |
+
"loss": 1.607,
|
| 165 |
+
"step": 5000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 7.0,
|
| 169 |
+
"eval_f1": 0.9122,
|
| 170 |
+
"eval_gen_len": 25.49490909090909,
|
| 171 |
+
"eval_loss": 1.5425922870635986,
|
| 172 |
+
"eval_precision": 0.9129,
|
| 173 |
+
"eval_recall": 0.9118,
|
| 174 |
+
"eval_rouge1": 0.486,
|
| 175 |
+
"eval_rouge2": 0.2367,
|
| 176 |
+
"eval_rougeL": 0.4052,
|
| 177 |
+
"eval_rougeLsum": 0.405,
|
| 178 |
+
"eval_runtime": 866.1986,
|
| 179 |
+
"eval_samples_per_second": 6.35,
|
| 180 |
+
"eval_steps_per_second": 0.397,
|
| 181 |
+
"step": 5468
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 7.04,
|
| 185 |
+
"learning_rate": 2.3943661971830984e-06,
|
| 186 |
+
"loss": 1.6029,
|
| 187 |
+
"step": 5500
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 7.68,
|
| 191 |
+
"learning_rate": 7.93854033290653e-07,
|
| 192 |
+
"loss": 1.5972,
|
| 193 |
+
"step": 6000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"eval_f1": 0.9123,
|
| 198 |
+
"eval_gen_len": 25.38509090909091,
|
| 199 |
+
"eval_loss": 1.540870189666748,
|
| 200 |
+
"eval_precision": 0.9131,
|
| 201 |
+
"eval_recall": 0.9117,
|
| 202 |
+
"eval_rouge1": 0.4869,
|
| 203 |
+
"eval_rouge2": 0.2373,
|
| 204 |
+
"eval_rougeL": 0.406,
|
| 205 |
+
"eval_rougeLsum": 0.4058,
|
| 206 |
+
"eval_runtime": 863.7281,
|
| 207 |
+
"eval_samples_per_second": 6.368,
|
| 208 |
+
"eval_steps_per_second": 0.398,
|
| 209 |
+
"step": 6248
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 8.0,
|
| 213 |
+
"step": 6248,
|
| 214 |
+
"total_flos": 1.1554159142381814e+18,
|
| 215 |
+
"train_loss": 1.7074453571236545,
|
| 216 |
+
"train_runtime": 28108.9431,
|
| 217 |
+
"train_samples_per_second": 28.461,
|
| 218 |
+
"train_steps_per_second": 0.222
|
| 219 |
+
}
|
| 220 |
+
],
|
| 221 |
+
"logging_steps": 500,
|
| 222 |
+
"max_steps": 6248,
|
| 223 |
+
"num_input_tokens_seen": 0,
|
| 224 |
+
"num_train_epochs": 8,
|
| 225 |
+
"save_steps": 500,
|
| 226 |
+
"total_flos": 1.1554159142381814e+18,
|
| 227 |
+
"train_batch_size": 32,
|
| 228 |
+
"trial_name": null,
|
| 229 |
+
"trial_params": null
|
| 230 |
+
}
|