Commit
•
2e00be2
1
Parent(s):
8a02e30
Training in progress, epoch 0
Browse files- all_results.json +13 -0
- eval_results.json +8 -0
- model.safetensors +1 -1
- train_eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +472 -0
- training_args.bin +1 -1
- val_eval_results.json +8 -0
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.9066666666666666,
|
| 4 |
+
"eval_loss": 1.6493659019470215,
|
| 5 |
+
"eval_runtime": 4.846,
|
| 6 |
+
"eval_samples_per_second": 247.626,
|
| 7 |
+
"eval_steps_per_second": 2.064,
|
| 8 |
+
"total_flos": 3.720348931719168e+18,
|
| 9 |
+
"train_loss": 2.217303295135498,
|
| 10 |
+
"train_runtime": 465.0811,
|
| 11 |
+
"train_samples_per_second": 129.01,
|
| 12 |
+
"train_steps_per_second": 0.215
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.8361111111111111,
|
| 4 |
+
"eval_loss": 1.7763370275497437,
|
| 5 |
+
"eval_runtime": 3.2333,
|
| 6 |
+
"eval_samples_per_second": 222.684,
|
| 7 |
+
"eval_steps_per_second": 1.856
|
| 8 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343291648
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f885b5e90f1272edfb2e673aa4df3484bfcf55ae55e01e8d93a99879dbc9622a
|
| 3 |
size 343291648
|
train_eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.9066666666666666,
|
| 4 |
+
"eval_loss": 1.6493659019470215,
|
| 5 |
+
"eval_runtime": 4.846,
|
| 6 |
+
"eval_samples_per_second": 247.626,
|
| 7 |
+
"eval_steps_per_second": 2.064
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 3.720348931719168e+18,
|
| 4 |
+
"train_loss": 2.217303295135498,
|
| 5 |
+
"train_runtime": 465.0811,
|
| 6 |
+
"train_samples_per_second": 129.01,
|
| 7 |
+
"train_steps_per_second": 0.215
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,472 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8361111111111111,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-inaturalist/checkpoint-100",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8,
|
| 13 |
+
"eval_accuracy": 0.05694444444444444,
|
| 14 |
+
"eval_loss": 3.172593116760254,
|
| 15 |
+
"eval_runtime": 3.5125,
|
| 16 |
+
"eval_samples_per_second": 204.98,
|
| 17 |
+
"eval_steps_per_second": 1.708,
|
| 18 |
+
"step": 2
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 2.0,
|
| 22 |
+
"eval_accuracy": 0.11666666666666667,
|
| 23 |
+
"eval_loss": 3.1497302055358887,
|
| 24 |
+
"eval_runtime": 3.1347,
|
| 25 |
+
"eval_samples_per_second": 229.687,
|
| 26 |
+
"eval_steps_per_second": 1.914,
|
| 27 |
+
"step": 5
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.8,
|
| 31 |
+
"eval_accuracy": 0.18333333333333332,
|
| 32 |
+
"eval_loss": 3.121361017227173,
|
| 33 |
+
"eval_runtime": 3.1053,
|
| 34 |
+
"eval_samples_per_second": 231.865,
|
| 35 |
+
"eval_steps_per_second": 1.932,
|
| 36 |
+
"step": 7
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 4.0,
|
| 40 |
+
"grad_norm": 0.7334493398666382,
|
| 41 |
+
"learning_rate": 5e-05,
|
| 42 |
+
"loss": 3.1408,
|
| 43 |
+
"step": 10
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.31527777777777777,
|
| 48 |
+
"eval_loss": 3.0588860511779785,
|
| 49 |
+
"eval_runtime": 3.1331,
|
| 50 |
+
"eval_samples_per_second": 229.807,
|
| 51 |
+
"eval_steps_per_second": 1.915,
|
| 52 |
+
"step": 10
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.8,
|
| 56 |
+
"eval_accuracy": 0.4236111111111111,
|
| 57 |
+
"eval_loss": 3.0067594051361084,
|
| 58 |
+
"eval_runtime": 3.0962,
|
| 59 |
+
"eval_samples_per_second": 232.544,
|
| 60 |
+
"eval_steps_per_second": 1.938,
|
| 61 |
+
"step": 12
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 6.0,
|
| 65 |
+
"eval_accuracy": 0.5347222222222222,
|
| 66 |
+
"eval_loss": 2.9316446781158447,
|
| 67 |
+
"eval_runtime": 3.0379,
|
| 68 |
+
"eval_samples_per_second": 237.005,
|
| 69 |
+
"eval_steps_per_second": 1.975,
|
| 70 |
+
"step": 15
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 6.8,
|
| 74 |
+
"eval_accuracy": 0.5763888888888888,
|
| 75 |
+
"eval_loss": 2.8821420669555664,
|
| 76 |
+
"eval_runtime": 3.238,
|
| 77 |
+
"eval_samples_per_second": 222.356,
|
| 78 |
+
"eval_steps_per_second": 1.853,
|
| 79 |
+
"step": 17
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 8.0,
|
| 83 |
+
"grad_norm": 0.798339307308197,
|
| 84 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 85 |
+
"loss": 2.921,
|
| 86 |
+
"step": 20
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.0,
|
| 90 |
+
"eval_accuracy": 0.6180555555555556,
|
| 91 |
+
"eval_loss": 2.8060097694396973,
|
| 92 |
+
"eval_runtime": 3.1439,
|
| 93 |
+
"eval_samples_per_second": 229.013,
|
| 94 |
+
"eval_steps_per_second": 1.908,
|
| 95 |
+
"step": 20
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.8,
|
| 99 |
+
"eval_accuracy": 0.6458333333333334,
|
| 100 |
+
"eval_loss": 2.7532873153686523,
|
| 101 |
+
"eval_runtime": 3.0396,
|
| 102 |
+
"eval_samples_per_second": 236.87,
|
| 103 |
+
"eval_steps_per_second": 1.974,
|
| 104 |
+
"step": 22
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 10.0,
|
| 108 |
+
"eval_accuracy": 0.6875,
|
| 109 |
+
"eval_loss": 2.6782259941101074,
|
| 110 |
+
"eval_runtime": 3.1324,
|
| 111 |
+
"eval_samples_per_second": 229.852,
|
| 112 |
+
"eval_steps_per_second": 1.915,
|
| 113 |
+
"step": 25
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 10.8,
|
| 117 |
+
"eval_accuracy": 0.7069444444444445,
|
| 118 |
+
"eval_loss": 2.6306703090667725,
|
| 119 |
+
"eval_runtime": 3.0536,
|
| 120 |
+
"eval_samples_per_second": 235.785,
|
| 121 |
+
"eval_steps_per_second": 1.965,
|
| 122 |
+
"step": 27
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 12.0,
|
| 126 |
+
"grad_norm": 0.838556706905365,
|
| 127 |
+
"learning_rate": 3.888888888888889e-05,
|
| 128 |
+
"loss": 2.6444,
|
| 129 |
+
"step": 30
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 12.0,
|
| 133 |
+
"eval_accuracy": 0.7472222222222222,
|
| 134 |
+
"eval_loss": 2.5581698417663574,
|
| 135 |
+
"eval_runtime": 3.3694,
|
| 136 |
+
"eval_samples_per_second": 213.685,
|
| 137 |
+
"eval_steps_per_second": 1.781,
|
| 138 |
+
"step": 30
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 12.8,
|
| 142 |
+
"eval_accuracy": 0.7583333333333333,
|
| 143 |
+
"eval_loss": 2.510007381439209,
|
| 144 |
+
"eval_runtime": 3.0368,
|
| 145 |
+
"eval_samples_per_second": 237.091,
|
| 146 |
+
"eval_steps_per_second": 1.976,
|
| 147 |
+
"step": 32
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 14.0,
|
| 151 |
+
"eval_accuracy": 0.7666666666666667,
|
| 152 |
+
"eval_loss": 2.4404165744781494,
|
| 153 |
+
"eval_runtime": 3.1041,
|
| 154 |
+
"eval_samples_per_second": 231.955,
|
| 155 |
+
"eval_steps_per_second": 1.933,
|
| 156 |
+
"step": 35
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 14.8,
|
| 160 |
+
"eval_accuracy": 0.7708333333333334,
|
| 161 |
+
"eval_loss": 2.3961093425750732,
|
| 162 |
+
"eval_runtime": 2.9999,
|
| 163 |
+
"eval_samples_per_second": 240.006,
|
| 164 |
+
"eval_steps_per_second": 2.0,
|
| 165 |
+
"step": 37
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 16.0,
|
| 169 |
+
"grad_norm": 0.858555018901825,
|
| 170 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 171 |
+
"loss": 2.3812,
|
| 172 |
+
"step": 40
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 16.0,
|
| 176 |
+
"eval_accuracy": 0.7680555555555556,
|
| 177 |
+
"eval_loss": 2.3315865993499756,
|
| 178 |
+
"eval_runtime": 3.1014,
|
| 179 |
+
"eval_samples_per_second": 232.15,
|
| 180 |
+
"eval_steps_per_second": 1.935,
|
| 181 |
+
"step": 40
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 16.8,
|
| 185 |
+
"eval_accuracy": 0.7763888888888889,
|
| 186 |
+
"eval_loss": 2.290663957595825,
|
| 187 |
+
"eval_runtime": 3.0562,
|
| 188 |
+
"eval_samples_per_second": 235.585,
|
| 189 |
+
"eval_steps_per_second": 1.963,
|
| 190 |
+
"step": 42
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 18.0,
|
| 194 |
+
"eval_accuracy": 0.7791666666666667,
|
| 195 |
+
"eval_loss": 2.2317349910736084,
|
| 196 |
+
"eval_runtime": 3.075,
|
| 197 |
+
"eval_samples_per_second": 234.149,
|
| 198 |
+
"eval_steps_per_second": 1.951,
|
| 199 |
+
"step": 45
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 18.8,
|
| 203 |
+
"eval_accuracy": 0.7888888888888889,
|
| 204 |
+
"eval_loss": 2.1949777603149414,
|
| 205 |
+
"eval_runtime": 3.2371,
|
| 206 |
+
"eval_samples_per_second": 222.42,
|
| 207 |
+
"eval_steps_per_second": 1.854,
|
| 208 |
+
"step": 47
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 20.0,
|
| 212 |
+
"grad_norm": 0.8867940306663513,
|
| 213 |
+
"learning_rate": 2.777777777777778e-05,
|
| 214 |
+
"loss": 2.1612,
|
| 215 |
+
"step": 50
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 20.0,
|
| 219 |
+
"eval_accuracy": 0.7958333333333333,
|
| 220 |
+
"eval_loss": 2.1451728343963623,
|
| 221 |
+
"eval_runtime": 3.1048,
|
| 222 |
+
"eval_samples_per_second": 231.902,
|
| 223 |
+
"eval_steps_per_second": 1.933,
|
| 224 |
+
"step": 50
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 20.8,
|
| 228 |
+
"eval_accuracy": 0.7930555555555555,
|
| 229 |
+
"eval_loss": 2.1138837337493896,
|
| 230 |
+
"eval_runtime": 3.0589,
|
| 231 |
+
"eval_samples_per_second": 235.381,
|
| 232 |
+
"eval_steps_per_second": 1.962,
|
| 233 |
+
"step": 52
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 22.0,
|
| 237 |
+
"eval_accuracy": 0.7930555555555555,
|
| 238 |
+
"eval_loss": 2.0691888332366943,
|
| 239 |
+
"eval_runtime": 3.1824,
|
| 240 |
+
"eval_samples_per_second": 226.244,
|
| 241 |
+
"eval_steps_per_second": 1.885,
|
| 242 |
+
"step": 55
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 22.8,
|
| 246 |
+
"eval_accuracy": 0.7986111111111112,
|
| 247 |
+
"eval_loss": 2.041097402572632,
|
| 248 |
+
"eval_runtime": 3.0557,
|
| 249 |
+
"eval_samples_per_second": 235.626,
|
| 250 |
+
"eval_steps_per_second": 1.964,
|
| 251 |
+
"step": 57
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 24.0,
|
| 255 |
+
"grad_norm": 0.9011107683181763,
|
| 256 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 257 |
+
"loss": 1.9801,
|
| 258 |
+
"step": 60
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 24.0,
|
| 262 |
+
"eval_accuracy": 0.7986111111111112,
|
| 263 |
+
"eval_loss": 2.003042221069336,
|
| 264 |
+
"eval_runtime": 3.2184,
|
| 265 |
+
"eval_samples_per_second": 223.713,
|
| 266 |
+
"eval_steps_per_second": 1.864,
|
| 267 |
+
"step": 60
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 24.8,
|
| 271 |
+
"eval_accuracy": 0.8083333333333333,
|
| 272 |
+
"eval_loss": 1.97919499874115,
|
| 273 |
+
"eval_runtime": 3.1013,
|
| 274 |
+
"eval_samples_per_second": 232.157,
|
| 275 |
+
"eval_steps_per_second": 1.935,
|
| 276 |
+
"step": 62
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 26.0,
|
| 280 |
+
"eval_accuracy": 0.8097222222222222,
|
| 281 |
+
"eval_loss": 1.9466288089752197,
|
| 282 |
+
"eval_runtime": 3.1382,
|
| 283 |
+
"eval_samples_per_second": 229.434,
|
| 284 |
+
"eval_steps_per_second": 1.912,
|
| 285 |
+
"step": 65
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 26.8,
|
| 289 |
+
"eval_accuracy": 0.8083333333333333,
|
| 290 |
+
"eval_loss": 1.9272509813308716,
|
| 291 |
+
"eval_runtime": 3.2793,
|
| 292 |
+
"eval_samples_per_second": 219.559,
|
| 293 |
+
"eval_steps_per_second": 1.83,
|
| 294 |
+
"step": 67
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 28.0,
|
| 298 |
+
"grad_norm": 0.8933854699134827,
|
| 299 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 300 |
+
"loss": 1.8437,
|
| 301 |
+
"step": 70
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 28.0,
|
| 305 |
+
"eval_accuracy": 0.8152777777777778,
|
| 306 |
+
"eval_loss": 1.90021550655365,
|
| 307 |
+
"eval_runtime": 3.1022,
|
| 308 |
+
"eval_samples_per_second": 232.097,
|
| 309 |
+
"eval_steps_per_second": 1.934,
|
| 310 |
+
"step": 70
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 28.8,
|
| 314 |
+
"eval_accuracy": 0.8180555555555555,
|
| 315 |
+
"eval_loss": 1.8828704357147217,
|
| 316 |
+
"eval_runtime": 3.0817,
|
| 317 |
+
"eval_samples_per_second": 233.636,
|
| 318 |
+
"eval_steps_per_second": 1.947,
|
| 319 |
+
"step": 72
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 30.0,
|
| 323 |
+
"eval_accuracy": 0.825,
|
| 324 |
+
"eval_loss": 1.860001802444458,
|
| 325 |
+
"eval_runtime": 3.3637,
|
| 326 |
+
"eval_samples_per_second": 214.053,
|
| 327 |
+
"eval_steps_per_second": 1.784,
|
| 328 |
+
"step": 75
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 30.8,
|
| 332 |
+
"eval_accuracy": 0.8291666666666667,
|
| 333 |
+
"eval_loss": 1.8465533256530762,
|
| 334 |
+
"eval_runtime": 3.0565,
|
| 335 |
+
"eval_samples_per_second": 235.565,
|
| 336 |
+
"eval_steps_per_second": 1.963,
|
| 337 |
+
"step": 77
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 32.0,
|
| 341 |
+
"grad_norm": 0.9217949509620667,
|
| 342 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 343 |
+
"loss": 1.7518,
|
| 344 |
+
"step": 80
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 32.0,
|
| 348 |
+
"eval_accuracy": 0.8277777777777777,
|
| 349 |
+
"eval_loss": 1.827829360961914,
|
| 350 |
+
"eval_runtime": 3.1649,
|
| 351 |
+
"eval_samples_per_second": 227.494,
|
| 352 |
+
"eval_steps_per_second": 1.896,
|
| 353 |
+
"step": 80
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 32.8,
|
| 357 |
+
"eval_accuracy": 0.8319444444444445,
|
| 358 |
+
"eval_loss": 1.8170872926712036,
|
| 359 |
+
"eval_runtime": 3.4533,
|
| 360 |
+
"eval_samples_per_second": 208.494,
|
| 361 |
+
"eval_steps_per_second": 1.737,
|
| 362 |
+
"step": 82
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 34.0,
|
| 366 |
+
"eval_accuracy": 0.8333333333333334,
|
| 367 |
+
"eval_loss": 1.8042806386947632,
|
| 368 |
+
"eval_runtime": 3.7195,
|
| 369 |
+
"eval_samples_per_second": 193.576,
|
| 370 |
+
"eval_steps_per_second": 1.613,
|
| 371 |
+
"step": 85
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 34.8,
|
| 375 |
+
"eval_accuracy": 0.8347222222222223,
|
| 376 |
+
"eval_loss": 1.7976477146148682,
|
| 377 |
+
"eval_runtime": 3.0512,
|
| 378 |
+
"eval_samples_per_second": 235.975,
|
| 379 |
+
"eval_steps_per_second": 1.966,
|
| 380 |
+
"step": 87
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 36.0,
|
| 384 |
+
"grad_norm": 0.8925730586051941,
|
| 385 |
+
"learning_rate": 5.555555555555556e-06,
|
| 386 |
+
"loss": 1.6966,
|
| 387 |
+
"step": 90
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 36.0,
|
| 391 |
+
"eval_accuracy": 0.8319444444444445,
|
| 392 |
+
"eval_loss": 1.7895526885986328,
|
| 393 |
+
"eval_runtime": 3.0775,
|
| 394 |
+
"eval_samples_per_second": 233.957,
|
| 395 |
+
"eval_steps_per_second": 1.95,
|
| 396 |
+
"step": 90
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 36.8,
|
| 400 |
+
"eval_accuracy": 0.8347222222222223,
|
| 401 |
+
"eval_loss": 1.7851399183273315,
|
| 402 |
+
"eval_runtime": 3.1908,
|
| 403 |
+
"eval_samples_per_second": 225.647,
|
| 404 |
+
"eval_steps_per_second": 1.88,
|
| 405 |
+
"step": 92
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 38.0,
|
| 409 |
+
"eval_accuracy": 0.8347222222222223,
|
| 410 |
+
"eval_loss": 1.7800432443618774,
|
| 411 |
+
"eval_runtime": 3.2053,
|
| 412 |
+
"eval_samples_per_second": 224.627,
|
| 413 |
+
"eval_steps_per_second": 1.872,
|
| 414 |
+
"step": 95
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 38.8,
|
| 418 |
+
"eval_accuracy": 0.8347222222222223,
|
| 419 |
+
"eval_loss": 1.7778292894363403,
|
| 420 |
+
"eval_runtime": 3.0496,
|
| 421 |
+
"eval_samples_per_second": 236.093,
|
| 422 |
+
"eval_steps_per_second": 1.967,
|
| 423 |
+
"step": 97
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 40.0,
|
| 427 |
+
"grad_norm": 0.8934192061424255,
|
| 428 |
+
"learning_rate": 0.0,
|
| 429 |
+
"loss": 1.6522,
|
| 430 |
+
"step": 100
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 40.0,
|
| 434 |
+
"eval_accuracy": 0.8361111111111111,
|
| 435 |
+
"eval_loss": 1.7763370275497437,
|
| 436 |
+
"eval_runtime": 3.345,
|
| 437 |
+
"eval_samples_per_second": 215.246,
|
| 438 |
+
"eval_steps_per_second": 1.794,
|
| 439 |
+
"step": 100
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 40.0,
|
| 443 |
+
"step": 100,
|
| 444 |
+
"total_flos": 3.720348931719168e+18,
|
| 445 |
+
"train_loss": 2.217303295135498,
|
| 446 |
+
"train_runtime": 465.0811,
|
| 447 |
+
"train_samples_per_second": 129.01,
|
| 448 |
+
"train_steps_per_second": 0.215
|
| 449 |
+
}
|
| 450 |
+
],
|
| 451 |
+
"logging_steps": 10,
|
| 452 |
+
"max_steps": 100,
|
| 453 |
+
"num_input_tokens_seen": 0,
|
| 454 |
+
"num_train_epochs": 50,
|
| 455 |
+
"save_steps": 500,
|
| 456 |
+
"stateful_callbacks": {
|
| 457 |
+
"TrainerControl": {
|
| 458 |
+
"args": {
|
| 459 |
+
"should_epoch_stop": false,
|
| 460 |
+
"should_evaluate": false,
|
| 461 |
+
"should_log": false,
|
| 462 |
+
"should_save": true,
|
| 463 |
+
"should_training_stop": true
|
| 464 |
+
},
|
| 465 |
+
"attributes": {}
|
| 466 |
+
}
|
| 467 |
+
},
|
| 468 |
+
"total_flos": 3.720348931719168e+18,
|
| 469 |
+
"train_batch_size": 128,
|
| 470 |
+
"trial_name": null,
|
| 471 |
+
"trial_params": null
|
| 472 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af0140e2af8a1f2250efdd62a9ff3a56a0f8aa77d7a3cf01cafc87a67d4282ce
|
| 3 |
size 5304
|
val_eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.8361111111111111,
|
| 4 |
+
"eval_loss": 1.7763370275497437,
|
| 5 |
+
"eval_runtime": 3.2414,
|
| 6 |
+
"eval_samples_per_second": 222.127,
|
| 7 |
+
"eval_steps_per_second": 1.851
|
| 8 |
+
}
|