Commit
•
fa0d9ae
1
Parent(s):
a024413
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr01_23-07-37_e71c489085e0/events.out.tfevents.1712014345.e71c489085e0.1159.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +503 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
- Accuracy: 1.0
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0367
|
| 21 |
- Accuracy: 1.0
|
| 22 |
|
| 23 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.03671738877892494,
|
| 5 |
+
"eval_runtime": 2.014,
|
| 6 |
+
"eval_samples_per_second": 72.988,
|
| 7 |
+
"eval_steps_per_second": 2.483,
|
| 8 |
+
"total_flos": 2.366647785100247e+18,
|
| 9 |
+
"train_loss": 0.2210820493598779,
|
| 10 |
+
"train_runtime": 1366.588,
|
| 11 |
+
"train_samples_per_second": 24.206,
|
| 12 |
+
"train_steps_per_second": 0.176
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.03671738877892494,
|
| 5 |
+
"eval_runtime": 2.014,
|
| 6 |
+
"eval_samples_per_second": 72.988,
|
| 7 |
+
"eval_steps_per_second": 2.483
|
| 8 |
+
}
|
runs/Apr01_23-07-37_e71c489085e0/events.out.tfevents.1712014345.e71c489085e0.1159.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c22bed83682a84dc9c1e9d63dcc1f8bb35a5b306a6cae962085f4c4497b6f2e8
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"total_flos": 2.366647785100247e+18,
|
| 4 |
+
"train_loss": 0.2210820493598779,
|
| 5 |
+
"train_runtime": 1366.588,
|
| 6 |
+
"train_samples_per_second": 24.206,
|
| 7 |
+
"train_steps_per_second": 0.176
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,503 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-dmae-va-U5-40/checkpoint-65",
|
| 4 |
+
"epoch": 36.92307692307692,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.92,
|
| 13 |
+
"eval_accuracy": 0.36054421768707484,
|
| 14 |
+
"eval_loss": 1.4235920906066895,
|
| 15 |
+
"eval_runtime": 2.4199,
|
| 16 |
+
"eval_samples_per_second": 60.746,
|
| 17 |
+
"eval_steps_per_second": 2.066,
|
| 18 |
+
"step": 6
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.85,
|
| 22 |
+
"grad_norm": 4.381593227386475,
|
| 23 |
+
"learning_rate": 2.5e-05,
|
| 24 |
+
"loss": 1.3165,
|
| 25 |
+
"step": 12
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_accuracy": 0.5306122448979592,
|
| 30 |
+
"eval_loss": 1.107216715812683,
|
| 31 |
+
"eval_runtime": 1.8757,
|
| 32 |
+
"eval_samples_per_second": 78.37,
|
| 33 |
+
"eval_steps_per_second": 2.666,
|
| 34 |
+
"step": 13
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.92,
|
| 38 |
+
"eval_accuracy": 0.6462585034013606,
|
| 39 |
+
"eval_loss": 0.9370094537734985,
|
| 40 |
+
"eval_runtime": 1.9321,
|
| 41 |
+
"eval_samples_per_second": 76.085,
|
| 42 |
+
"eval_steps_per_second": 2.588,
|
| 43 |
+
"step": 19
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.69,
|
| 47 |
+
"grad_norm": 2.5531413555145264,
|
| 48 |
+
"learning_rate": 5e-05,
|
| 49 |
+
"loss": 0.93,
|
| 50 |
+
"step": 24
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.7687074829931972,
|
| 55 |
+
"eval_loss": 0.6869749426841736,
|
| 56 |
+
"eval_runtime": 1.98,
|
| 57 |
+
"eval_samples_per_second": 74.243,
|
| 58 |
+
"eval_steps_per_second": 2.525,
|
| 59 |
+
"step": 26
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.92,
|
| 63 |
+
"eval_accuracy": 0.8163265306122449,
|
| 64 |
+
"eval_loss": 0.47430145740509033,
|
| 65 |
+
"eval_runtime": 1.9979,
|
| 66 |
+
"eval_samples_per_second": 73.576,
|
| 67 |
+
"eval_steps_per_second": 2.503,
|
| 68 |
+
"step": 32
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.54,
|
| 72 |
+
"grad_norm": 3.2196173667907715,
|
| 73 |
+
"learning_rate": 4.722222222222222e-05,
|
| 74 |
+
"loss": 0.5368,
|
| 75 |
+
"step": 36
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 6.0,
|
| 79 |
+
"eval_accuracy": 0.9183673469387755,
|
| 80 |
+
"eval_loss": 0.2825295627117157,
|
| 81 |
+
"eval_runtime": 1.9929,
|
| 82 |
+
"eval_samples_per_second": 73.762,
|
| 83 |
+
"eval_steps_per_second": 2.509,
|
| 84 |
+
"step": 39
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.92,
|
| 88 |
+
"eval_accuracy": 0.9523809523809523,
|
| 89 |
+
"eval_loss": 0.20659707486629486,
|
| 90 |
+
"eval_runtime": 1.9939,
|
| 91 |
+
"eval_samples_per_second": 73.724,
|
| 92 |
+
"eval_steps_per_second": 2.508,
|
| 93 |
+
"step": 45
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 7.38,
|
| 97 |
+
"grad_norm": 2.4960756301879883,
|
| 98 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 99 |
+
"loss": 0.2989,
|
| 100 |
+
"step": 48
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 8.0,
|
| 104 |
+
"eval_accuracy": 0.9727891156462585,
|
| 105 |
+
"eval_loss": 0.12244854122400284,
|
| 106 |
+
"eval_runtime": 1.9835,
|
| 107 |
+
"eval_samples_per_second": 74.113,
|
| 108 |
+
"eval_steps_per_second": 2.521,
|
| 109 |
+
"step": 52
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 8.92,
|
| 113 |
+
"eval_accuracy": 0.9591836734693877,
|
| 114 |
+
"eval_loss": 0.1452772617340088,
|
| 115 |
+
"eval_runtime": 2.011,
|
| 116 |
+
"eval_samples_per_second": 73.097,
|
| 117 |
+
"eval_steps_per_second": 2.486,
|
| 118 |
+
"step": 58
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 9.23,
|
| 122 |
+
"grad_norm": 2.9144787788391113,
|
| 123 |
+
"learning_rate": 4.166666666666667e-05,
|
| 124 |
+
"loss": 0.1746,
|
| 125 |
+
"step": 60
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 10.0,
|
| 129 |
+
"eval_accuracy": 1.0,
|
| 130 |
+
"eval_loss": 0.03671738877892494,
|
| 131 |
+
"eval_runtime": 2.0115,
|
| 132 |
+
"eval_samples_per_second": 73.081,
|
| 133 |
+
"eval_steps_per_second": 2.486,
|
| 134 |
+
"step": 65
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.92,
|
| 138 |
+
"eval_accuracy": 0.9863945578231292,
|
| 139 |
+
"eval_loss": 0.06163691729307175,
|
| 140 |
+
"eval_runtime": 1.9765,
|
| 141 |
+
"eval_samples_per_second": 74.373,
|
| 142 |
+
"eval_steps_per_second": 2.53,
|
| 143 |
+
"step": 71
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.08,
|
| 147 |
+
"grad_norm": 1.5245277881622314,
|
| 148 |
+
"learning_rate": 3.888888888888889e-05,
|
| 149 |
+
"loss": 0.1596,
|
| 150 |
+
"step": 72
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 12.0,
|
| 154 |
+
"eval_accuracy": 1.0,
|
| 155 |
+
"eval_loss": 0.02339293248951435,
|
| 156 |
+
"eval_runtime": 1.9968,
|
| 157 |
+
"eval_samples_per_second": 73.619,
|
| 158 |
+
"eval_steps_per_second": 2.504,
|
| 159 |
+
"step": 78
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 12.92,
|
| 163 |
+
"grad_norm": 2.6396472454071045,
|
| 164 |
+
"learning_rate": 3.611111111111111e-05,
|
| 165 |
+
"loss": 0.1094,
|
| 166 |
+
"step": 84
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.92,
|
| 170 |
+
"eval_accuracy": 1.0,
|
| 171 |
+
"eval_loss": 0.029785707592964172,
|
| 172 |
+
"eval_runtime": 1.9947,
|
| 173 |
+
"eval_samples_per_second": 73.695,
|
| 174 |
+
"eval_steps_per_second": 2.507,
|
| 175 |
+
"step": 84
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 14.0,
|
| 179 |
+
"eval_accuracy": 0.9931972789115646,
|
| 180 |
+
"eval_loss": 0.044449541717767715,
|
| 181 |
+
"eval_runtime": 1.9933,
|
| 182 |
+
"eval_samples_per_second": 73.747,
|
| 183 |
+
"eval_steps_per_second": 2.508,
|
| 184 |
+
"step": 91
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 14.77,
|
| 188 |
+
"grad_norm": 1.8975120782852173,
|
| 189 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 190 |
+
"loss": 0.1123,
|
| 191 |
+
"step": 96
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 14.92,
|
| 195 |
+
"eval_accuracy": 1.0,
|
| 196 |
+
"eval_loss": 0.025111636146903038,
|
| 197 |
+
"eval_runtime": 2.0344,
|
| 198 |
+
"eval_samples_per_second": 72.256,
|
| 199 |
+
"eval_steps_per_second": 2.458,
|
| 200 |
+
"step": 97
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 16.0,
|
| 204 |
+
"eval_accuracy": 1.0,
|
| 205 |
+
"eval_loss": 0.018478814512491226,
|
| 206 |
+
"eval_runtime": 1.9976,
|
| 207 |
+
"eval_samples_per_second": 73.589,
|
| 208 |
+
"eval_steps_per_second": 2.503,
|
| 209 |
+
"step": 104
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 16.62,
|
| 213 |
+
"grad_norm": 2.431682825088501,
|
| 214 |
+
"learning_rate": 3.055555555555556e-05,
|
| 215 |
+
"loss": 0.0761,
|
| 216 |
+
"step": 108
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 16.92,
|
| 220 |
+
"eval_accuracy": 1.0,
|
| 221 |
+
"eval_loss": 0.015857575461268425,
|
| 222 |
+
"eval_runtime": 2.0057,
|
| 223 |
+
"eval_samples_per_second": 73.291,
|
| 224 |
+
"eval_steps_per_second": 2.493,
|
| 225 |
+
"step": 110
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 18.0,
|
| 229 |
+
"eval_accuracy": 1.0,
|
| 230 |
+
"eval_loss": 0.01796538569033146,
|
| 231 |
+
"eval_runtime": 2.0303,
|
| 232 |
+
"eval_samples_per_second": 72.403,
|
| 233 |
+
"eval_steps_per_second": 2.463,
|
| 234 |
+
"step": 117
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 18.46,
|
| 238 |
+
"grad_norm": 2.1142959594726562,
|
| 239 |
+
"learning_rate": 2.777777777777778e-05,
|
| 240 |
+
"loss": 0.0743,
|
| 241 |
+
"step": 120
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 18.92,
|
| 245 |
+
"eval_accuracy": 1.0,
|
| 246 |
+
"eval_loss": 0.011095724068582058,
|
| 247 |
+
"eval_runtime": 1.9932,
|
| 248 |
+
"eval_samples_per_second": 73.752,
|
| 249 |
+
"eval_steps_per_second": 2.509,
|
| 250 |
+
"step": 123
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 20.0,
|
| 254 |
+
"eval_accuracy": 1.0,
|
| 255 |
+
"eval_loss": 0.013449358753859997,
|
| 256 |
+
"eval_runtime": 1.9948,
|
| 257 |
+
"eval_samples_per_second": 73.693,
|
| 258 |
+
"eval_steps_per_second": 2.507,
|
| 259 |
+
"step": 130
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 20.31,
|
| 263 |
+
"grad_norm": 1.569281816482544,
|
| 264 |
+
"learning_rate": 2.5e-05,
|
| 265 |
+
"loss": 0.072,
|
| 266 |
+
"step": 132
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 20.92,
|
| 270 |
+
"eval_accuracy": 1.0,
|
| 271 |
+
"eval_loss": 0.012339694425463676,
|
| 272 |
+
"eval_runtime": 1.9955,
|
| 273 |
+
"eval_samples_per_second": 73.666,
|
| 274 |
+
"eval_steps_per_second": 2.506,
|
| 275 |
+
"step": 136
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 22.0,
|
| 279 |
+
"eval_accuracy": 1.0,
|
| 280 |
+
"eval_loss": 0.010039811953902245,
|
| 281 |
+
"eval_runtime": 1.9958,
|
| 282 |
+
"eval_samples_per_second": 73.653,
|
| 283 |
+
"eval_steps_per_second": 2.505,
|
| 284 |
+
"step": 143
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 22.15,
|
| 288 |
+
"grad_norm": 2.3920440673828125,
|
| 289 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 290 |
+
"loss": 0.0744,
|
| 291 |
+
"step": 144
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 22.92,
|
| 295 |
+
"eval_accuracy": 1.0,
|
| 296 |
+
"eval_loss": 0.007376910652965307,
|
| 297 |
+
"eval_runtime": 1.9896,
|
| 298 |
+
"eval_samples_per_second": 73.884,
|
| 299 |
+
"eval_steps_per_second": 2.513,
|
| 300 |
+
"step": 149
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 24.0,
|
| 304 |
+
"grad_norm": 0.7604403495788574,
|
| 305 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 306 |
+
"loss": 0.0688,
|
| 307 |
+
"step": 156
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 24.0,
|
| 311 |
+
"eval_accuracy": 1.0,
|
| 312 |
+
"eval_loss": 0.006399971432983875,
|
| 313 |
+
"eval_runtime": 1.9944,
|
| 314 |
+
"eval_samples_per_second": 73.705,
|
| 315 |
+
"eval_steps_per_second": 2.507,
|
| 316 |
+
"step": 156
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 24.92,
|
| 320 |
+
"eval_accuracy": 1.0,
|
| 321 |
+
"eval_loss": 0.006952380761504173,
|
| 322 |
+
"eval_runtime": 2.0094,
|
| 323 |
+
"eval_samples_per_second": 73.157,
|
| 324 |
+
"eval_steps_per_second": 2.488,
|
| 325 |
+
"step": 162
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 25.85,
|
| 329 |
+
"grad_norm": 1.1607767343521118,
|
| 330 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 331 |
+
"loss": 0.0737,
|
| 332 |
+
"step": 168
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 26.0,
|
| 336 |
+
"eval_accuracy": 1.0,
|
| 337 |
+
"eval_loss": 0.006380416918545961,
|
| 338 |
+
"eval_runtime": 2.0125,
|
| 339 |
+
"eval_samples_per_second": 73.044,
|
| 340 |
+
"eval_steps_per_second": 2.484,
|
| 341 |
+
"step": 169
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 26.92,
|
| 345 |
+
"eval_accuracy": 1.0,
|
| 346 |
+
"eval_loss": 0.00552676897495985,
|
| 347 |
+
"eval_runtime": 2.2042,
|
| 348 |
+
"eval_samples_per_second": 66.69,
|
| 349 |
+
"eval_steps_per_second": 2.268,
|
| 350 |
+
"step": 175
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 27.69,
|
| 354 |
+
"grad_norm": 2.0216870307922363,
|
| 355 |
+
"learning_rate": 1.388888888888889e-05,
|
| 356 |
+
"loss": 0.053,
|
| 357 |
+
"step": 180
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 28.0,
|
| 361 |
+
"eval_accuracy": 1.0,
|
| 362 |
+
"eval_loss": 0.007457624655216932,
|
| 363 |
+
"eval_runtime": 2.2171,
|
| 364 |
+
"eval_samples_per_second": 66.304,
|
| 365 |
+
"eval_steps_per_second": 2.255,
|
| 366 |
+
"step": 182
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 28.92,
|
| 370 |
+
"eval_accuracy": 1.0,
|
| 371 |
+
"eval_loss": 0.004635449033230543,
|
| 372 |
+
"eval_runtime": 2.1386,
|
| 373 |
+
"eval_samples_per_second": 68.735,
|
| 374 |
+
"eval_steps_per_second": 2.338,
|
| 375 |
+
"step": 188
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 29.54,
|
| 379 |
+
"grad_norm": 1.227410912513733,
|
| 380 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 381 |
+
"loss": 0.0677,
|
| 382 |
+
"step": 192
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 30.0,
|
| 386 |
+
"eval_accuracy": 1.0,
|
| 387 |
+
"eval_loss": 0.004566570743918419,
|
| 388 |
+
"eval_runtime": 2.0055,
|
| 389 |
+
"eval_samples_per_second": 73.299,
|
| 390 |
+
"eval_steps_per_second": 2.493,
|
| 391 |
+
"step": 195
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 30.92,
|
| 395 |
+
"eval_accuracy": 1.0,
|
| 396 |
+
"eval_loss": 0.00975565891712904,
|
| 397 |
+
"eval_runtime": 1.9971,
|
| 398 |
+
"eval_samples_per_second": 73.607,
|
| 399 |
+
"eval_steps_per_second": 2.504,
|
| 400 |
+
"step": 201
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 31.38,
|
| 404 |
+
"grad_norm": 1.6289031505584717,
|
| 405 |
+
"learning_rate": 8.333333333333334e-06,
|
| 406 |
+
"loss": 0.055,
|
| 407 |
+
"step": 204
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 32.0,
|
| 411 |
+
"eval_accuracy": 1.0,
|
| 412 |
+
"eval_loss": 0.008506382815539837,
|
| 413 |
+
"eval_runtime": 1.9954,
|
| 414 |
+
"eval_samples_per_second": 73.671,
|
| 415 |
+
"eval_steps_per_second": 2.506,
|
| 416 |
+
"step": 208
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 32.92,
|
| 420 |
+
"eval_accuracy": 1.0,
|
| 421 |
+
"eval_loss": 0.0055803339928388596,
|
| 422 |
+
"eval_runtime": 2.0106,
|
| 423 |
+
"eval_samples_per_second": 73.111,
|
| 424 |
+
"eval_steps_per_second": 2.487,
|
| 425 |
+
"step": 214
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 33.23,
|
| 429 |
+
"grad_norm": 1.2801518440246582,
|
| 430 |
+
"learning_rate": 5.555555555555556e-06,
|
| 431 |
+
"loss": 0.0576,
|
| 432 |
+
"step": 216
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 34.0,
|
| 436 |
+
"eval_accuracy": 1.0,
|
| 437 |
+
"eval_loss": 0.005879435688257217,
|
| 438 |
+
"eval_runtime": 1.9938,
|
| 439 |
+
"eval_samples_per_second": 73.727,
|
| 440 |
+
"eval_steps_per_second": 2.508,
|
| 441 |
+
"step": 221
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 34.92,
|
| 445 |
+
"eval_accuracy": 1.0,
|
| 446 |
+
"eval_loss": 0.005405152682214975,
|
| 447 |
+
"eval_runtime": 2.0162,
|
| 448 |
+
"eval_samples_per_second": 72.911,
|
| 449 |
+
"eval_steps_per_second": 2.48,
|
| 450 |
+
"step": 227
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 35.08,
|
| 454 |
+
"grad_norm": 1.9271787405014038,
|
| 455 |
+
"learning_rate": 2.777777777777778e-06,
|
| 456 |
+
"loss": 0.0697,
|
| 457 |
+
"step": 228
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 36.0,
|
| 461 |
+
"eval_accuracy": 1.0,
|
| 462 |
+
"eval_loss": 0.005514008458703756,
|
| 463 |
+
"eval_runtime": 1.9717,
|
| 464 |
+
"eval_samples_per_second": 74.555,
|
| 465 |
+
"eval_steps_per_second": 2.536,
|
| 466 |
+
"step": 234
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 36.92,
|
| 470 |
+
"grad_norm": 0.9594835638999939,
|
| 471 |
+
"learning_rate": 0.0,
|
| 472 |
+
"loss": 0.0415,
|
| 473 |
+
"step": 240
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 36.92,
|
| 477 |
+
"eval_accuracy": 1.0,
|
| 478 |
+
"eval_loss": 0.005595473572611809,
|
| 479 |
+
"eval_runtime": 2.0157,
|
| 480 |
+
"eval_samples_per_second": 72.927,
|
| 481 |
+
"eval_steps_per_second": 2.481,
|
| 482 |
+
"step": 240
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 36.92,
|
| 486 |
+
"step": 240,
|
| 487 |
+
"total_flos": 2.366647785100247e+18,
|
| 488 |
+
"train_loss": 0.2210820493598779,
|
| 489 |
+
"train_runtime": 1366.588,
|
| 490 |
+
"train_samples_per_second": 24.206,
|
| 491 |
+
"train_steps_per_second": 0.176
|
| 492 |
+
}
|
| 493 |
+
],
|
| 494 |
+
"logging_steps": 12,
|
| 495 |
+
"max_steps": 240,
|
| 496 |
+
"num_input_tokens_seen": 0,
|
| 497 |
+
"num_train_epochs": 40,
|
| 498 |
+
"save_steps": 500,
|
| 499 |
+
"total_flos": 2.366647785100247e+18,
|
| 500 |
+
"train_batch_size": 32,
|
| 501 |
+
"trial_name": null,
|
| 502 |
+
"trial_params": null
|
| 503 |
+
}
|