Commit
•
5724555
1
Parent(s):
c8ee3bb
End of training
Browse files- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +484 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.99,
|
| 3 |
+
"train_loss": 1.050457198154915,
|
| 4 |
+
"train_runtime": 71670.9422,
|
| 5 |
+
"train_samples_per_second": 22.324,
|
| 6 |
+
"train_steps_per_second": 0.232
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.99,
|
| 3 |
+
"train_loss": 1.050457198154915,
|
| 4 |
+
"train_runtime": 71670.9422,
|
| 5 |
+
"train_samples_per_second": 22.324,
|
| 6 |
+
"train_steps_per_second": 0.232
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.988480921526278,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16656,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.48,
|
| 13 |
+
"learning_rate": 1.9399615754082615e-05,
|
| 14 |
+
"loss": 1.8112,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.96,
|
| 19 |
+
"learning_rate": 1.879923150816523e-05,
|
| 20 |
+
"loss": 1.6328,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 1.0,
|
| 25 |
+
"eval_f1": 0.9029,
|
| 26 |
+
"eval_gen_len": 19.87818181818182,
|
| 27 |
+
"eval_loss": 1.4800708293914795,
|
| 28 |
+
"eval_precision": 0.9134,
|
| 29 |
+
"eval_recall": 0.893,
|
| 30 |
+
"eval_rouge1": 0.448,
|
| 31 |
+
"eval_rouge2": 0.2243,
|
| 32 |
+
"eval_rougeL": 0.385,
|
| 33 |
+
"eval_rougeLsum": 0.385,
|
| 34 |
+
"eval_runtime": 603.3554,
|
| 35 |
+
"eval_samples_per_second": 9.116,
|
| 36 |
+
"eval_steps_per_second": 0.57,
|
| 37 |
+
"step": 1041
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.44,
|
| 41 |
+
"learning_rate": 1.819884726224784e-05,
|
| 42 |
+
"loss": 1.4991,
|
| 43 |
+
"step": 1500
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 1.92,
|
| 47 |
+
"learning_rate": 1.7598463016330453e-05,
|
| 48 |
+
"loss": 1.4598,
|
| 49 |
+
"step": 2000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 2.0,
|
| 53 |
+
"eval_f1": 0.9022,
|
| 54 |
+
"eval_gen_len": 19.934363636363635,
|
| 55 |
+
"eval_loss": 1.405110478401184,
|
| 56 |
+
"eval_precision": 0.9147,
|
| 57 |
+
"eval_recall": 0.8903,
|
| 58 |
+
"eval_rouge1": 0.4428,
|
| 59 |
+
"eval_rouge2": 0.2273,
|
| 60 |
+
"eval_rougeL": 0.3851,
|
| 61 |
+
"eval_rougeLsum": 0.385,
|
| 62 |
+
"eval_runtime": 669.8531,
|
| 63 |
+
"eval_samples_per_second": 8.211,
|
| 64 |
+
"eval_steps_per_second": 0.514,
|
| 65 |
+
"step": 2083
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.4,
|
| 69 |
+
"learning_rate": 1.6998078770413066e-05,
|
| 70 |
+
"loss": 1.3652,
|
| 71 |
+
"step": 2500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.88,
|
| 75 |
+
"learning_rate": 1.6397694524495677e-05,
|
| 76 |
+
"loss": 1.3402,
|
| 77 |
+
"step": 3000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.0,
|
| 81 |
+
"eval_f1": 0.9034,
|
| 82 |
+
"eval_gen_len": 19.95,
|
| 83 |
+
"eval_loss": 1.3839877843856812,
|
| 84 |
+
"eval_precision": 0.9158,
|
| 85 |
+
"eval_recall": 0.8918,
|
| 86 |
+
"eval_rouge1": 0.4498,
|
| 87 |
+
"eval_rouge2": 0.2318,
|
| 88 |
+
"eval_rougeL": 0.3921,
|
| 89 |
+
"eval_rougeLsum": 0.392,
|
| 90 |
+
"eval_runtime": 670.3562,
|
| 91 |
+
"eval_samples_per_second": 8.205,
|
| 92 |
+
"eval_steps_per_second": 0.513,
|
| 93 |
+
"step": 3125
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 3.36,
|
| 97 |
+
"learning_rate": 1.579731027857829e-05,
|
| 98 |
+
"loss": 1.2679,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.84,
|
| 103 |
+
"learning_rate": 1.5196926032660904e-05,
|
| 104 |
+
"loss": 1.2446,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 4.0,
|
| 109 |
+
"eval_f1": 0.9054,
|
| 110 |
+
"eval_gen_len": 19.884,
|
| 111 |
+
"eval_loss": 1.3682185411453247,
|
| 112 |
+
"eval_precision": 0.9169,
|
| 113 |
+
"eval_recall": 0.8944,
|
| 114 |
+
"eval_rouge1": 0.4604,
|
| 115 |
+
"eval_rouge2": 0.2405,
|
| 116 |
+
"eval_rougeL": 0.4014,
|
| 117 |
+
"eval_rougeLsum": 0.4014,
|
| 118 |
+
"eval_runtime": 577.6339,
|
| 119 |
+
"eval_samples_per_second": 9.522,
|
| 120 |
+
"eval_steps_per_second": 0.596,
|
| 121 |
+
"step": 4167
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 4.32,
|
| 125 |
+
"learning_rate": 1.4596541786743516e-05,
|
| 126 |
+
"loss": 1.1877,
|
| 127 |
+
"step": 4500
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 4.8,
|
| 131 |
+
"learning_rate": 1.399615754082613e-05,
|
| 132 |
+
"loss": 1.1651,
|
| 133 |
+
"step": 5000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 5.0,
|
| 137 |
+
"eval_f1": 0.9055,
|
| 138 |
+
"eval_gen_len": 19.894,
|
| 139 |
+
"eval_loss": 1.3695330619812012,
|
| 140 |
+
"eval_precision": 0.9173,
|
| 141 |
+
"eval_recall": 0.8942,
|
| 142 |
+
"eval_rouge1": 0.4594,
|
| 143 |
+
"eval_rouge2": 0.2401,
|
| 144 |
+
"eval_rougeL": 0.3995,
|
| 145 |
+
"eval_rougeLsum": 0.3995,
|
| 146 |
+
"eval_runtime": 669.362,
|
| 147 |
+
"eval_samples_per_second": 8.217,
|
| 148 |
+
"eval_steps_per_second": 0.514,
|
| 149 |
+
"step": 5208
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 5.28,
|
| 153 |
+
"learning_rate": 1.3395773294908743e-05,
|
| 154 |
+
"loss": 1.1201,
|
| 155 |
+
"step": 5500
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.76,
|
| 159 |
+
"learning_rate": 1.2795389048991355e-05,
|
| 160 |
+
"loss": 1.1002,
|
| 161 |
+
"step": 6000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.0,
|
| 165 |
+
"eval_f1": 0.9053,
|
| 166 |
+
"eval_gen_len": 19.91181818181818,
|
| 167 |
+
"eval_loss": 1.3782570362091064,
|
| 168 |
+
"eval_precision": 0.9166,
|
| 169 |
+
"eval_recall": 0.8945,
|
| 170 |
+
"eval_rouge1": 0.4607,
|
| 171 |
+
"eval_rouge2": 0.2423,
|
| 172 |
+
"eval_rougeL": 0.4014,
|
| 173 |
+
"eval_rougeLsum": 0.4014,
|
| 174 |
+
"eval_runtime": 671.1543,
|
| 175 |
+
"eval_samples_per_second": 8.195,
|
| 176 |
+
"eval_steps_per_second": 0.513,
|
| 177 |
+
"step": 6250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 6.24,
|
| 181 |
+
"learning_rate": 1.2195004803073969e-05,
|
| 182 |
+
"loss": 1.0653,
|
| 183 |
+
"step": 6500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.72,
|
| 187 |
+
"learning_rate": 1.1594620557156582e-05,
|
| 188 |
+
"loss": 1.0427,
|
| 189 |
+
"step": 7000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 7.0,
|
| 193 |
+
"eval_f1": 0.9056,
|
| 194 |
+
"eval_gen_len": 19.907454545454545,
|
| 195 |
+
"eval_loss": 1.3850913047790527,
|
| 196 |
+
"eval_precision": 0.9172,
|
| 197 |
+
"eval_recall": 0.8946,
|
| 198 |
+
"eval_rouge1": 0.462,
|
| 199 |
+
"eval_rouge2": 0.2432,
|
| 200 |
+
"eval_rougeL": 0.4028,
|
| 201 |
+
"eval_rougeLsum": 0.4028,
|
| 202 |
+
"eval_runtime": 669.8936,
|
| 203 |
+
"eval_samples_per_second": 8.21,
|
| 204 |
+
"eval_steps_per_second": 0.514,
|
| 205 |
+
"step": 7292
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.2,
|
| 209 |
+
"learning_rate": 1.0994236311239194e-05,
|
| 210 |
+
"loss": 1.0163,
|
| 211 |
+
"step": 7500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 7.68,
|
| 215 |
+
"learning_rate": 1.0393852065321808e-05,
|
| 216 |
+
"loss": 0.9881,
|
| 217 |
+
"step": 8000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.0,
|
| 221 |
+
"eval_f1": 0.9059,
|
| 222 |
+
"eval_gen_len": 19.907090909090908,
|
| 223 |
+
"eval_loss": 1.3910883665084839,
|
| 224 |
+
"eval_precision": 0.9177,
|
| 225 |
+
"eval_recall": 0.8947,
|
| 226 |
+
"eval_rouge1": 0.4635,
|
| 227 |
+
"eval_rouge2": 0.2442,
|
| 228 |
+
"eval_rougeL": 0.4038,
|
| 229 |
+
"eval_rougeLsum": 0.4037,
|
| 230 |
+
"eval_runtime": 573.3321,
|
| 231 |
+
"eval_samples_per_second": 9.593,
|
| 232 |
+
"eval_steps_per_second": 0.6,
|
| 233 |
+
"step": 8334
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 8.16,
|
| 237 |
+
"learning_rate": 9.79346781940442e-06,
|
| 238 |
+
"loss": 0.9742,
|
| 239 |
+
"step": 8500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 8.64,
|
| 243 |
+
"learning_rate": 9.193083573487034e-06,
|
| 244 |
+
"loss": 0.9435,
|
| 245 |
+
"step": 9000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 9.0,
|
| 249 |
+
"eval_f1": 0.9067,
|
| 250 |
+
"eval_gen_len": 19.880545454545455,
|
| 251 |
+
"eval_loss": 1.4075220823287964,
|
| 252 |
+
"eval_precision": 0.918,
|
| 253 |
+
"eval_recall": 0.8959,
|
| 254 |
+
"eval_rouge1": 0.468,
|
| 255 |
+
"eval_rouge2": 0.2471,
|
| 256 |
+
"eval_rougeL": 0.4085,
|
| 257 |
+
"eval_rougeLsum": 0.4084,
|
| 258 |
+
"eval_runtime": 599.9366,
|
| 259 |
+
"eval_samples_per_second": 9.168,
|
| 260 |
+
"eval_steps_per_second": 0.573,
|
| 261 |
+
"step": 9375
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 9.12,
|
| 265 |
+
"learning_rate": 8.592699327569645e-06,
|
| 266 |
+
"loss": 0.9362,
|
| 267 |
+
"step": 9500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 9.6,
|
| 271 |
+
"learning_rate": 7.992315081652257e-06,
|
| 272 |
+
"loss": 0.9035,
|
| 273 |
+
"step": 10000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 10.0,
|
| 277 |
+
"eval_f1": 0.9064,
|
| 278 |
+
"eval_gen_len": 19.881090909090908,
|
| 279 |
+
"eval_loss": 1.412468671798706,
|
| 280 |
+
"eval_precision": 0.9178,
|
| 281 |
+
"eval_recall": 0.8957,
|
| 282 |
+
"eval_rouge1": 0.4675,
|
| 283 |
+
"eval_rouge2": 0.248,
|
| 284 |
+
"eval_rougeL": 0.4085,
|
| 285 |
+
"eval_rougeLsum": 0.4086,
|
| 286 |
+
"eval_runtime": 566.1377,
|
| 287 |
+
"eval_samples_per_second": 9.715,
|
| 288 |
+
"eval_steps_per_second": 0.608,
|
| 289 |
+
"step": 10417
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 10.08,
|
| 293 |
+
"learning_rate": 7.391930835734871e-06,
|
| 294 |
+
"loss": 0.9014,
|
| 295 |
+
"step": 10500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 10.56,
|
| 299 |
+
"learning_rate": 6.791546589817484e-06,
|
| 300 |
+
"loss": 0.8702,
|
| 301 |
+
"step": 11000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 11.0,
|
| 305 |
+
"eval_f1": 0.9063,
|
| 306 |
+
"eval_gen_len": 19.894727272727273,
|
| 307 |
+
"eval_loss": 1.4218909740447998,
|
| 308 |
+
"eval_precision": 0.9181,
|
| 309 |
+
"eval_recall": 0.895,
|
| 310 |
+
"eval_rouge1": 0.4646,
|
| 311 |
+
"eval_rouge2": 0.2455,
|
| 312 |
+
"eval_rougeL": 0.405,
|
| 313 |
+
"eval_rougeLsum": 0.4051,
|
| 314 |
+
"eval_runtime": 670.3799,
|
| 315 |
+
"eval_samples_per_second": 8.204,
|
| 316 |
+
"eval_steps_per_second": 0.513,
|
| 317 |
+
"step": 11459
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 11.04,
|
| 321 |
+
"learning_rate": 6.191162343900097e-06,
|
| 322 |
+
"loss": 0.8741,
|
| 323 |
+
"step": 11500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 11.52,
|
| 327 |
+
"learning_rate": 5.590778097982709e-06,
|
| 328 |
+
"loss": 0.8395,
|
| 329 |
+
"step": 12000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 12.0,
|
| 333 |
+
"learning_rate": 4.990393852065322e-06,
|
| 334 |
+
"loss": 0.8458,
|
| 335 |
+
"step": 12500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_f1": 0.9061,
|
| 340 |
+
"eval_gen_len": 19.898545454545456,
|
| 341 |
+
"eval_loss": 1.4338867664337158,
|
| 342 |
+
"eval_precision": 0.9177,
|
| 343 |
+
"eval_recall": 0.8952,
|
| 344 |
+
"eval_rouge1": 0.4643,
|
| 345 |
+
"eval_rouge2": 0.2447,
|
| 346 |
+
"eval_rougeL": 0.4055,
|
| 347 |
+
"eval_rougeLsum": 0.4055,
|
| 348 |
+
"eval_runtime": 670.6829,
|
| 349 |
+
"eval_samples_per_second": 8.201,
|
| 350 |
+
"eval_steps_per_second": 0.513,
|
| 351 |
+
"step": 12501
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 12.48,
|
| 355 |
+
"learning_rate": 4.390009606147935e-06,
|
| 356 |
+
"loss": 0.8172,
|
| 357 |
+
"step": 13000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.96,
|
| 361 |
+
"learning_rate": 3.7896253602305477e-06,
|
| 362 |
+
"loss": 0.8207,
|
| 363 |
+
"step": 13500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 13.0,
|
| 367 |
+
"eval_f1": 0.9064,
|
| 368 |
+
"eval_gen_len": 19.905272727272727,
|
| 369 |
+
"eval_loss": 1.44303560256958,
|
| 370 |
+
"eval_precision": 0.9182,
|
| 371 |
+
"eval_recall": 0.8952,
|
| 372 |
+
"eval_rouge1": 0.4671,
|
| 373 |
+
"eval_rouge2": 0.2463,
|
| 374 |
+
"eval_rougeL": 0.4068,
|
| 375 |
+
"eval_rougeLsum": 0.4069,
|
| 376 |
+
"eval_runtime": 650.7057,
|
| 377 |
+
"eval_samples_per_second": 8.452,
|
| 378 |
+
"eval_steps_per_second": 0.529,
|
| 379 |
+
"step": 13542
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.44,
|
| 383 |
+
"learning_rate": 3.189241114313161e-06,
|
| 384 |
+
"loss": 0.8006,
|
| 385 |
+
"step": 14000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 13.92,
|
| 389 |
+
"learning_rate": 2.5888568683957737e-06,
|
| 390 |
+
"loss": 0.7987,
|
| 391 |
+
"step": 14500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 14.0,
|
| 395 |
+
"eval_f1": 0.9059,
|
| 396 |
+
"eval_gen_len": 19.918,
|
| 397 |
+
"eval_loss": 1.449475646018982,
|
| 398 |
+
"eval_precision": 0.9179,
|
| 399 |
+
"eval_recall": 0.8944,
|
| 400 |
+
"eval_rouge1": 0.4633,
|
| 401 |
+
"eval_rouge2": 0.2455,
|
| 402 |
+
"eval_rougeL": 0.4046,
|
| 403 |
+
"eval_rougeLsum": 0.4047,
|
| 404 |
+
"eval_runtime": 661.0314,
|
| 405 |
+
"eval_samples_per_second": 8.32,
|
| 406 |
+
"eval_steps_per_second": 0.52,
|
| 407 |
+
"step": 14584
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 14.4,
|
| 411 |
+
"learning_rate": 1.988472622478386e-06,
|
| 412 |
+
"loss": 0.7843,
|
| 413 |
+
"step": 15000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 14.88,
|
| 417 |
+
"learning_rate": 1.3880883765609993e-06,
|
| 418 |
+
"loss": 0.787,
|
| 419 |
+
"step": 15500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 15.0,
|
| 423 |
+
"eval_f1": 0.9064,
|
| 424 |
+
"eval_gen_len": 19.895636363636363,
|
| 425 |
+
"eval_loss": 1.4560260772705078,
|
| 426 |
+
"eval_precision": 0.9182,
|
| 427 |
+
"eval_recall": 0.8953,
|
| 428 |
+
"eval_rouge1": 0.4666,
|
| 429 |
+
"eval_rouge2": 0.2471,
|
| 430 |
+
"eval_rougeL": 0.407,
|
| 431 |
+
"eval_rougeLsum": 0.4072,
|
| 432 |
+
"eval_runtime": 670.9962,
|
| 433 |
+
"eval_samples_per_second": 8.197,
|
| 434 |
+
"eval_steps_per_second": 0.513,
|
| 435 |
+
"step": 15626
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 15.36,
|
| 439 |
+
"learning_rate": 7.87704130643612e-07,
|
| 440 |
+
"loss": 0.7775,
|
| 441 |
+
"step": 16000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 15.84,
|
| 445 |
+
"learning_rate": 1.8731988472622478e-07,
|
| 446 |
+
"loss": 0.772,
|
| 447 |
+
"step": 16500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 15.99,
|
| 451 |
+
"eval_f1": 0.9068,
|
| 452 |
+
"eval_gen_len": 19.881636363636364,
|
| 453 |
+
"eval_loss": 1.4622657299041748,
|
| 454 |
+
"eval_precision": 0.9185,
|
| 455 |
+
"eval_recall": 0.8957,
|
| 456 |
+
"eval_rouge1": 0.4678,
|
| 457 |
+
"eval_rouge2": 0.2472,
|
| 458 |
+
"eval_rougeL": 0.4081,
|
| 459 |
+
"eval_rougeLsum": 0.4082,
|
| 460 |
+
"eval_runtime": 669.6134,
|
| 461 |
+
"eval_samples_per_second": 8.214,
|
| 462 |
+
"eval_steps_per_second": 0.514,
|
| 463 |
+
"step": 16656
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 15.99,
|
| 467 |
+
"step": 16656,
|
| 468 |
+
"total_flos": 3.421567656204632e+18,
|
| 469 |
+
"train_loss": 1.050457198154915,
|
| 470 |
+
"train_runtime": 71670.9422,
|
| 471 |
+
"train_samples_per_second": 22.324,
|
| 472 |
+
"train_steps_per_second": 0.232
|
| 473 |
+
}
|
| 474 |
+
],
|
| 475 |
+
"logging_steps": 500,
|
| 476 |
+
"max_steps": 16656,
|
| 477 |
+
"num_input_tokens_seen": 0,
|
| 478 |
+
"num_train_epochs": 16,
|
| 479 |
+
"save_steps": 500,
|
| 480 |
+
"total_flos": 3.421567656204632e+18,
|
| 481 |
+
"train_batch_size": 24,
|
| 482 |
+
"trial_name": null,
|
| 483 |
+
"trial_params": null
|
| 484 |
+
}
|