End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9297520661157025,
|
| 4 |
+
"eval_loss": 0.19896750152111053,
|
| 5 |
+
"eval_runtime": 13.631,
|
| 6 |
+
"eval_samples_per_second": 213.043,
|
| 7 |
+
"eval_steps_per_second": 6.676,
|
| 8 |
+
"total_flos": 1.6836842977571635e+18,
|
| 9 |
+
"train_loss": 0.419760063009442,
|
| 10 |
+
"train_runtime": 917.7632,
|
| 11 |
+
"train_samples_per_second": 73.799,
|
| 12 |
+
"train_steps_per_second": 0.577
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9297520661157025,
|
| 4 |
+
"eval_loss": 0.19896750152111053,
|
| 5 |
+
"eval_runtime": 13.631,
|
| 6 |
+
"eval_samples_per_second": 213.043,
|
| 7 |
+
"eval_steps_per_second": 6.676
|
| 8 |
+
}
|
runs/Jul26_02-43-40_DESKTOP-BJ85B6S/events.out.tfevents.1721958773.DESKTOP-BJ85B6S.19880.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e76ba18b441f5781a9fd5329129264c0e39bcc41501c20c5921eba930ce694f1
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.6836842977571635e+18,
|
| 4 |
+
"train_loss": 0.419760063009442,
|
| 5 |
+
"train_runtime": 917.7632,
|
| 6 |
+
"train_samples_per_second": 73.799,
|
| 7 |
+
"train_steps_per_second": 0.577
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,503 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9297520661157025,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-wsdmhar\\checkpoint-530",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 530,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.18867924528301888,
|
| 13 |
+
"grad_norm": 5.205289840698242,
|
| 14 |
+
"learning_rate": 9.433962264150944e-06,
|
| 15 |
+
"loss": 1.7479,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.37735849056603776,
|
| 20 |
+
"grad_norm": 4.392179012298584,
|
| 21 |
+
"learning_rate": 1.8867924528301888e-05,
|
| 22 |
+
"loss": 1.4506,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.5660377358490566,
|
| 27 |
+
"grad_norm": 6.868739604949951,
|
| 28 |
+
"learning_rate": 2.830188679245283e-05,
|
| 29 |
+
"loss": 1.0283,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.7547169811320755,
|
| 34 |
+
"grad_norm": 9.729569435119629,
|
| 35 |
+
"learning_rate": 3.7735849056603776e-05,
|
| 36 |
+
"loss": 0.8064,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.9433962264150944,
|
| 41 |
+
"grad_norm": 12.286840438842773,
|
| 42 |
+
"learning_rate": 4.716981132075472e-05,
|
| 43 |
+
"loss": 0.7388,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_accuracy": 0.7117768595041323,
|
| 49 |
+
"eval_loss": 0.6308234930038452,
|
| 50 |
+
"eval_runtime": 24.0663,
|
| 51 |
+
"eval_samples_per_second": 120.667,
|
| 52 |
+
"eval_steps_per_second": 3.781,
|
| 53 |
+
"step": 53
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.1320754716981132,
|
| 57 |
+
"grad_norm": 9.05235481262207,
|
| 58 |
+
"learning_rate": 4.9266247379454926e-05,
|
| 59 |
+
"loss": 0.6616,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.320754716981132,
|
| 64 |
+
"grad_norm": 10.109884262084961,
|
| 65 |
+
"learning_rate": 4.8218029350104823e-05,
|
| 66 |
+
"loss": 0.6363,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.509433962264151,
|
| 71 |
+
"grad_norm": 10.793712615966797,
|
| 72 |
+
"learning_rate": 4.716981132075472e-05,
|
| 73 |
+
"loss": 0.554,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.6981132075471699,
|
| 78 |
+
"grad_norm": 17.802772521972656,
|
| 79 |
+
"learning_rate": 4.612159329140461e-05,
|
| 80 |
+
"loss": 0.5516,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.8867924528301887,
|
| 85 |
+
"grad_norm": 11.082262992858887,
|
| 86 |
+
"learning_rate": 4.5073375262054504e-05,
|
| 87 |
+
"loss": 0.5099,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.0,
|
| 92 |
+
"eval_accuracy": 0.8484848484848485,
|
| 93 |
+
"eval_loss": 0.36690881848335266,
|
| 94 |
+
"eval_runtime": 13.3175,
|
| 95 |
+
"eval_samples_per_second": 218.059,
|
| 96 |
+
"eval_steps_per_second": 6.833,
|
| 97 |
+
"step": 106
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.0754716981132075,
|
| 101 |
+
"grad_norm": 6.183718204498291,
|
| 102 |
+
"learning_rate": 4.402515723270441e-05,
|
| 103 |
+
"loss": 0.4778,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.2641509433962264,
|
| 108 |
+
"grad_norm": 13.359389305114746,
|
| 109 |
+
"learning_rate": 4.29769392033543e-05,
|
| 110 |
+
"loss": 0.4449,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.452830188679245,
|
| 115 |
+
"grad_norm": 6.166327953338623,
|
| 116 |
+
"learning_rate": 4.192872117400419e-05,
|
| 117 |
+
"loss": 0.4243,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.641509433962264,
|
| 122 |
+
"grad_norm": 6.538959980010986,
|
| 123 |
+
"learning_rate": 4.088050314465409e-05,
|
| 124 |
+
"loss": 0.4265,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.830188679245283,
|
| 129 |
+
"grad_norm": 11.39398193359375,
|
| 130 |
+
"learning_rate": 3.983228511530399e-05,
|
| 131 |
+
"loss": 0.4319,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 3.0,
|
| 136 |
+
"eval_accuracy": 0.8684573002754821,
|
| 137 |
+
"eval_loss": 0.33240532875061035,
|
| 138 |
+
"eval_runtime": 14.2442,
|
| 139 |
+
"eval_samples_per_second": 203.873,
|
| 140 |
+
"eval_steps_per_second": 6.389,
|
| 141 |
+
"step": 159
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.018867924528302,
|
| 145 |
+
"grad_norm": 9.0410795211792,
|
| 146 |
+
"learning_rate": 3.878406708595388e-05,
|
| 147 |
+
"loss": 0.3985,
|
| 148 |
+
"step": 160
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.207547169811321,
|
| 152 |
+
"grad_norm": 6.786131381988525,
|
| 153 |
+
"learning_rate": 3.7735849056603776e-05,
|
| 154 |
+
"loss": 0.3912,
|
| 155 |
+
"step": 170
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.3962264150943398,
|
| 159 |
+
"grad_norm": 8.235289573669434,
|
| 160 |
+
"learning_rate": 3.6687631027253674e-05,
|
| 161 |
+
"loss": 0.3496,
|
| 162 |
+
"step": 180
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.5849056603773586,
|
| 166 |
+
"grad_norm": 6.866382122039795,
|
| 167 |
+
"learning_rate": 3.5639412997903565e-05,
|
| 168 |
+
"loss": 0.3696,
|
| 169 |
+
"step": 190
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.7735849056603774,
|
| 173 |
+
"grad_norm": 8.17050552368164,
|
| 174 |
+
"learning_rate": 3.4591194968553456e-05,
|
| 175 |
+
"loss": 0.3203,
|
| 176 |
+
"step": 200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.9622641509433962,
|
| 180 |
+
"grad_norm": 17.23760414123535,
|
| 181 |
+
"learning_rate": 3.354297693920336e-05,
|
| 182 |
+
"loss": 0.4002,
|
| 183 |
+
"step": 210
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 4.0,
|
| 187 |
+
"eval_accuracy": 0.9028925619834711,
|
| 188 |
+
"eval_loss": 0.2758175730705261,
|
| 189 |
+
"eval_runtime": 14.4978,
|
| 190 |
+
"eval_samples_per_second": 200.306,
|
| 191 |
+
"eval_steps_per_second": 6.277,
|
| 192 |
+
"step": 212
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.150943396226415,
|
| 196 |
+
"grad_norm": 17.225961685180664,
|
| 197 |
+
"learning_rate": 3.249475890985325e-05,
|
| 198 |
+
"loss": 0.3484,
|
| 199 |
+
"step": 220
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.339622641509434,
|
| 203 |
+
"grad_norm": 12.645174980163574,
|
| 204 |
+
"learning_rate": 3.144654088050314e-05,
|
| 205 |
+
"loss": 0.353,
|
| 206 |
+
"step": 230
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.528301886792453,
|
| 210 |
+
"grad_norm": 5.725615978240967,
|
| 211 |
+
"learning_rate": 3.0398322851153044e-05,
|
| 212 |
+
"loss": 0.2883,
|
| 213 |
+
"step": 240
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.716981132075472,
|
| 217 |
+
"grad_norm": 5.914320468902588,
|
| 218 |
+
"learning_rate": 2.935010482180294e-05,
|
| 219 |
+
"loss": 0.3067,
|
| 220 |
+
"step": 250
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 4.90566037735849,
|
| 224 |
+
"grad_norm": 6.547543525695801,
|
| 225 |
+
"learning_rate": 2.830188679245283e-05,
|
| 226 |
+
"loss": 0.3589,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 5.0,
|
| 231 |
+
"eval_accuracy": 0.9132231404958677,
|
| 232 |
+
"eval_loss": 0.2503372132778168,
|
| 233 |
+
"eval_runtime": 14.406,
|
| 234 |
+
"eval_samples_per_second": 201.582,
|
| 235 |
+
"eval_steps_per_second": 6.317,
|
| 236 |
+
"step": 265
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.09433962264151,
|
| 240 |
+
"grad_norm": 8.123590469360352,
|
| 241 |
+
"learning_rate": 2.7253668763102725e-05,
|
| 242 |
+
"loss": 0.3272,
|
| 243 |
+
"step": 270
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.283018867924528,
|
| 247 |
+
"grad_norm": 5.585710048675537,
|
| 248 |
+
"learning_rate": 2.6205450733752623e-05,
|
| 249 |
+
"loss": 0.3535,
|
| 250 |
+
"step": 280
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 5.471698113207547,
|
| 254 |
+
"grad_norm": 8.529882431030273,
|
| 255 |
+
"learning_rate": 2.5157232704402517e-05,
|
| 256 |
+
"loss": 0.2797,
|
| 257 |
+
"step": 290
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 5.660377358490566,
|
| 261 |
+
"grad_norm": 6.009942054748535,
|
| 262 |
+
"learning_rate": 2.4109014675052412e-05,
|
| 263 |
+
"loss": 0.3004,
|
| 264 |
+
"step": 300
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 5.849056603773585,
|
| 268 |
+
"grad_norm": 4.354741096496582,
|
| 269 |
+
"learning_rate": 2.3060796645702306e-05,
|
| 270 |
+
"loss": 0.3096,
|
| 271 |
+
"step": 310
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 6.0,
|
| 275 |
+
"eval_accuracy": 0.9135674931129476,
|
| 276 |
+
"eval_loss": 0.24190402030944824,
|
| 277 |
+
"eval_runtime": 14.5262,
|
| 278 |
+
"eval_samples_per_second": 199.915,
|
| 279 |
+
"eval_steps_per_second": 6.265,
|
| 280 |
+
"step": 318
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 6.037735849056604,
|
| 284 |
+
"grad_norm": 6.793937683105469,
|
| 285 |
+
"learning_rate": 2.2012578616352204e-05,
|
| 286 |
+
"loss": 0.3024,
|
| 287 |
+
"step": 320
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 6.226415094339623,
|
| 291 |
+
"grad_norm": 4.6357221603393555,
|
| 292 |
+
"learning_rate": 2.0964360587002095e-05,
|
| 293 |
+
"loss": 0.3013,
|
| 294 |
+
"step": 330
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.415094339622642,
|
| 298 |
+
"grad_norm": 5.533959865570068,
|
| 299 |
+
"learning_rate": 1.9916142557651993e-05,
|
| 300 |
+
"loss": 0.311,
|
| 301 |
+
"step": 340
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 6.60377358490566,
|
| 305 |
+
"grad_norm": 5.0147600173950195,
|
| 306 |
+
"learning_rate": 1.8867924528301888e-05,
|
| 307 |
+
"loss": 0.2865,
|
| 308 |
+
"step": 350
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.7924528301886795,
|
| 312 |
+
"grad_norm": 4.817899227142334,
|
| 313 |
+
"learning_rate": 1.7819706498951782e-05,
|
| 314 |
+
"loss": 0.2813,
|
| 315 |
+
"step": 360
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.981132075471698,
|
| 319 |
+
"grad_norm": 6.032566547393799,
|
| 320 |
+
"learning_rate": 1.677148846960168e-05,
|
| 321 |
+
"loss": 0.2708,
|
| 322 |
+
"step": 370
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 7.0,
|
| 326 |
+
"eval_accuracy": 0.9232093663911846,
|
| 327 |
+
"eval_loss": 0.22773109376430511,
|
| 328 |
+
"eval_runtime": 15.0479,
|
| 329 |
+
"eval_samples_per_second": 192.984,
|
| 330 |
+
"eval_steps_per_second": 6.047,
|
| 331 |
+
"step": 371
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 7.169811320754717,
|
| 335 |
+
"grad_norm": 5.316783428192139,
|
| 336 |
+
"learning_rate": 1.572327044025157e-05,
|
| 337 |
+
"loss": 0.3009,
|
| 338 |
+
"step": 380
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 7.3584905660377355,
|
| 342 |
+
"grad_norm": 5.2730607986450195,
|
| 343 |
+
"learning_rate": 1.467505241090147e-05,
|
| 344 |
+
"loss": 0.2422,
|
| 345 |
+
"step": 390
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 7.547169811320755,
|
| 349 |
+
"grad_norm": 5.739044189453125,
|
| 350 |
+
"learning_rate": 1.3626834381551362e-05,
|
| 351 |
+
"loss": 0.2667,
|
| 352 |
+
"step": 400
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 7.735849056603773,
|
| 356 |
+
"grad_norm": 5.689853668212891,
|
| 357 |
+
"learning_rate": 1.2578616352201259e-05,
|
| 358 |
+
"loss": 0.2833,
|
| 359 |
+
"step": 410
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 7.9245283018867925,
|
| 363 |
+
"grad_norm": 6.43410062789917,
|
| 364 |
+
"learning_rate": 1.1530398322851153e-05,
|
| 365 |
+
"loss": 0.261,
|
| 366 |
+
"step": 420
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 8.0,
|
| 370 |
+
"eval_accuracy": 0.925275482093664,
|
| 371 |
+
"eval_loss": 0.21677668392658234,
|
| 372 |
+
"eval_runtime": 14.7132,
|
| 373 |
+
"eval_samples_per_second": 197.373,
|
| 374 |
+
"eval_steps_per_second": 6.185,
|
| 375 |
+
"step": 424
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 8.11320754716981,
|
| 379 |
+
"grad_norm": 5.5042643547058105,
|
| 380 |
+
"learning_rate": 1.0482180293501048e-05,
|
| 381 |
+
"loss": 0.2675,
|
| 382 |
+
"step": 430
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 8.30188679245283,
|
| 386 |
+
"grad_norm": 4.880862236022949,
|
| 387 |
+
"learning_rate": 9.433962264150944e-06,
|
| 388 |
+
"loss": 0.2413,
|
| 389 |
+
"step": 440
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 8.49056603773585,
|
| 393 |
+
"grad_norm": 3.9594244956970215,
|
| 394 |
+
"learning_rate": 8.38574423480084e-06,
|
| 395 |
+
"loss": 0.2577,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 8.679245283018869,
|
| 400 |
+
"grad_norm": 6.740973949432373,
|
| 401 |
+
"learning_rate": 7.337526205450735e-06,
|
| 402 |
+
"loss": 0.2595,
|
| 403 |
+
"step": 460
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 8.867924528301886,
|
| 407 |
+
"grad_norm": 7.188548564910889,
|
| 408 |
+
"learning_rate": 6.289308176100629e-06,
|
| 409 |
+
"loss": 0.2526,
|
| 410 |
+
"step": 470
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 9.0,
|
| 414 |
+
"eval_accuracy": 0.9245867768595041,
|
| 415 |
+
"eval_loss": 0.20991356670856476,
|
| 416 |
+
"eval_runtime": 14.793,
|
| 417 |
+
"eval_samples_per_second": 196.31,
|
| 418 |
+
"eval_steps_per_second": 6.152,
|
| 419 |
+
"step": 477
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 9.056603773584905,
|
| 423 |
+
"grad_norm": 8.5458402633667,
|
| 424 |
+
"learning_rate": 5.241090146750524e-06,
|
| 425 |
+
"loss": 0.2465,
|
| 426 |
+
"step": 480
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 9.245283018867925,
|
| 430 |
+
"grad_norm": 4.695093154907227,
|
| 431 |
+
"learning_rate": 4.19287211740042e-06,
|
| 432 |
+
"loss": 0.2172,
|
| 433 |
+
"step": 490
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 9.433962264150944,
|
| 437 |
+
"grad_norm": 3.67899751663208,
|
| 438 |
+
"learning_rate": 3.1446540880503146e-06,
|
| 439 |
+
"loss": 0.2635,
|
| 440 |
+
"step": 500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 9.622641509433961,
|
| 444 |
+
"grad_norm": 4.685281276702881,
|
| 445 |
+
"learning_rate": 2.09643605870021e-06,
|
| 446 |
+
"loss": 0.254,
|
| 447 |
+
"step": 510
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 9.81132075471698,
|
| 451 |
+
"grad_norm": 5.518146991729736,
|
| 452 |
+
"learning_rate": 1.048218029350105e-06,
|
| 453 |
+
"loss": 0.2574,
|
| 454 |
+
"step": 520
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 10.0,
|
| 458 |
+
"grad_norm": 5.643453598022461,
|
| 459 |
+
"learning_rate": 0.0,
|
| 460 |
+
"loss": 0.2767,
|
| 461 |
+
"step": 530
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 10.0,
|
| 465 |
+
"eval_accuracy": 0.9297520661157025,
|
| 466 |
+
"eval_loss": 0.19896750152111053,
|
| 467 |
+
"eval_runtime": 15.5471,
|
| 468 |
+
"eval_samples_per_second": 186.787,
|
| 469 |
+
"eval_steps_per_second": 5.853,
|
| 470 |
+
"step": 530
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 10.0,
|
| 474 |
+
"step": 530,
|
| 475 |
+
"total_flos": 1.6836842977571635e+18,
|
| 476 |
+
"train_loss": 0.419760063009442,
|
| 477 |
+
"train_runtime": 917.7632,
|
| 478 |
+
"train_samples_per_second": 73.799,
|
| 479 |
+
"train_steps_per_second": 0.577
|
| 480 |
+
}
|
| 481 |
+
],
|
| 482 |
+
"logging_steps": 10,
|
| 483 |
+
"max_steps": 530,
|
| 484 |
+
"num_input_tokens_seen": 0,
|
| 485 |
+
"num_train_epochs": 10,
|
| 486 |
+
"save_steps": 500,
|
| 487 |
+
"stateful_callbacks": {
|
| 488 |
+
"TrainerControl": {
|
| 489 |
+
"args": {
|
| 490 |
+
"should_epoch_stop": false,
|
| 491 |
+
"should_evaluate": false,
|
| 492 |
+
"should_log": false,
|
| 493 |
+
"should_save": true,
|
| 494 |
+
"should_training_stop": true
|
| 495 |
+
},
|
| 496 |
+
"attributes": {}
|
| 497 |
+
}
|
| 498 |
+
},
|
| 499 |
+
"total_flos": 1.6836842977571635e+18,
|
| 500 |
+
"train_batch_size": 32,
|
| 501 |
+
"trial_name": null,
|
| 502 |
+
"trial_params": null
|
| 503 |
+
}
|