Upload trainer_state.json
Browse files- trainer_state.json +211 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,211 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.8,
|
| 5 |
+
"global_step": 5200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.6,
|
| 12 |
+
"learning_rate": 0.0002394,
|
| 13 |
+
"loss": 3.4203,
|
| 14 |
+
"step": 400
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.6,
|
| 18 |
+
"eval_loss": Infinity,
|
| 19 |
+
"eval_runtime": 1205.6526,
|
| 20 |
+
"eval_samples_per_second": 3.822,
|
| 21 |
+
"eval_steps_per_second": 0.478,
|
| 22 |
+
"eval_wer": 0.5028872209091932,
|
| 23 |
+
"step": 400
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 3.2,
|
| 27 |
+
"learning_rate": 0.0002873142857142857,
|
| 28 |
+
"loss": 0.4156,
|
| 29 |
+
"step": 800
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 3.2,
|
| 33 |
+
"eval_loss": Infinity,
|
| 34 |
+
"eval_runtime": 1205.8762,
|
| 35 |
+
"eval_samples_per_second": 3.821,
|
| 36 |
+
"eval_steps_per_second": 0.478,
|
| 37 |
+
"eval_wer": 0.21501354872780476,
|
| 38 |
+
"step": 800
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.8,
|
| 42 |
+
"learning_rate": 0.0002702142857142857,
|
| 43 |
+
"loss": 0.248,
|
| 44 |
+
"step": 1200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.8,
|
| 48 |
+
"eval_loss": Infinity,
|
| 49 |
+
"eval_runtime": 1206.2112,
|
| 50 |
+
"eval_samples_per_second": 3.82,
|
| 51 |
+
"eval_steps_per_second": 0.478,
|
| 52 |
+
"eval_wer": 0.19620116181423092,
|
| 53 |
+
"step": 1200
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 6.4,
|
| 57 |
+
"learning_rate": 0.00025315714285714284,
|
| 58 |
+
"loss": 0.1884,
|
| 59 |
+
"step": 1600
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 6.4,
|
| 63 |
+
"eval_loss": Infinity,
|
| 64 |
+
"eval_runtime": 1198.1731,
|
| 65 |
+
"eval_samples_per_second": 3.846,
|
| 66 |
+
"eval_steps_per_second": 0.481,
|
| 67 |
+
"eval_wer": 0.18049675782839433,
|
| 68 |
+
"step": 1600
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 8.0,
|
| 72 |
+
"learning_rate": 0.00023605714285714283,
|
| 73 |
+
"loss": 0.1488,
|
| 74 |
+
"step": 2000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 8.0,
|
| 78 |
+
"eval_loss": Infinity,
|
| 79 |
+
"eval_runtime": 1195.3799,
|
| 80 |
+
"eval_samples_per_second": 3.855,
|
| 81 |
+
"eval_steps_per_second": 0.482,
|
| 82 |
+
"eval_wer": 0.18608073829744867,
|
| 83 |
+
"step": 2000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 9.6,
|
| 87 |
+
"learning_rate": 0.00021899999999999998,
|
| 88 |
+
"loss": 0.1232,
|
| 89 |
+
"step": 2400
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 9.6,
|
| 93 |
+
"eval_loss": Infinity,
|
| 94 |
+
"eval_runtime": 1196.1097,
|
| 95 |
+
"eval_samples_per_second": 3.852,
|
| 96 |
+
"eval_steps_per_second": 0.482,
|
| 97 |
+
"eval_wer": 0.17618540546624073,
|
| 98 |
+
"step": 2400
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.2,
|
| 102 |
+
"learning_rate": 0.0002019,
|
| 103 |
+
"loss": 0.107,
|
| 104 |
+
"step": 2800
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 11.2,
|
| 108 |
+
"eval_loss": Infinity,
|
| 109 |
+
"eval_runtime": 1192.8335,
|
| 110 |
+
"eval_samples_per_second": 3.863,
|
| 111 |
+
"eval_steps_per_second": 0.483,
|
| 112 |
+
"eval_wer": 0.17763118025435248,
|
| 113 |
+
"step": 2800
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 12.8,
|
| 117 |
+
"learning_rate": 0.0001848428571428571,
|
| 118 |
+
"loss": 0.2435,
|
| 119 |
+
"step": 3200
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 12.8,
|
| 123 |
+
"eval_loss": Infinity,
|
| 124 |
+
"eval_runtime": 1189.2109,
|
| 125 |
+
"eval_samples_per_second": 3.875,
|
| 126 |
+
"eval_steps_per_second": 0.484,
|
| 127 |
+
"eval_wer": 0.5853483278359262,
|
| 128 |
+
"step": 3200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 14.4,
|
| 132 |
+
"learning_rate": 0.00016774285714285715,
|
| 133 |
+
"loss": 0.7331,
|
| 134 |
+
"step": 3600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 14.4,
|
| 138 |
+
"eval_loss": Infinity,
|
| 139 |
+
"eval_runtime": 1196.7777,
|
| 140 |
+
"eval_samples_per_second": 3.85,
|
| 141 |
+
"eval_steps_per_second": 0.481,
|
| 142 |
+
"eval_wer": 0.5020128301690777,
|
| 143 |
+
"step": 3600
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 16.0,
|
| 147 |
+
"learning_rate": 0.0001506857142857143,
|
| 148 |
+
"loss": 0.5402,
|
| 149 |
+
"step": 4000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.0,
|
| 153 |
+
"eval_loss": Infinity,
|
| 154 |
+
"eval_runtime": 1194.9864,
|
| 155 |
+
"eval_samples_per_second": 3.856,
|
| 156 |
+
"eval_steps_per_second": 0.482,
|
| 157 |
+
"eval_wer": 0.29286895393432544,
|
| 158 |
+
"step": 4000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 17.6,
|
| 162 |
+
"learning_rate": 0.00013358571428571426,
|
| 163 |
+
"loss": 1.3361,
|
| 164 |
+
"step": 4400
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 17.6,
|
| 168 |
+
"eval_loss": Infinity,
|
| 169 |
+
"eval_runtime": 1207.0654,
|
| 170 |
+
"eval_samples_per_second": 3.818,
|
| 171 |
+
"eval_steps_per_second": 0.477,
|
| 172 |
+
"eval_wer": 0.9734566137703555,
|
| 173 |
+
"step": 4400
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 19.2,
|
| 177 |
+
"learning_rate": 0.0001165285714285714,
|
| 178 |
+
"loss": 0.9035,
|
| 179 |
+
"step": 4800
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 19.2,
|
| 183 |
+
"eval_loss": Infinity,
|
| 184 |
+
"eval_runtime": 1210.7223,
|
| 185 |
+
"eval_samples_per_second": 3.806,
|
| 186 |
+
"eval_steps_per_second": 0.476,
|
| 187 |
+
"eval_wer": 0.4126085413257841,
|
| 188 |
+
"step": 4800
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 20.8,
|
| 192 |
+
"learning_rate": 9.947142857142856e-05,
|
| 193 |
+
"loss": 0.5824,
|
| 194 |
+
"step": 5200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 20.8,
|
| 198 |
+
"eval_loss": Infinity,
|
| 199 |
+
"eval_runtime": 1209.2109,
|
| 200 |
+
"eval_samples_per_second": 3.811,
|
| 201 |
+
"eval_steps_per_second": 0.476,
|
| 202 |
+
"eval_wer": 0.2804110502211949,
|
| 203 |
+
"step": 5200
|
| 204 |
+
}
|
| 205 |
+
],
|
| 206 |
+
"max_steps": 7500,
|
| 207 |
+
"num_train_epochs": 30,
|
| 208 |
+
"total_flos": 8.725540278590487e+19,
|
| 209 |
+
"trial_name": null,
|
| 210 |
+
"trial_params": null
|
| 211 |
+
}
|