Commit
•
509071a
1
Parent(s):
0053395
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Mar11_00-37-16_89e2f3456c0c/events.out.tfevents.1710118958.89e2f3456c0c.2146.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +551 -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: 0.0213
|
| 21 |
- Accuracy: 1.0
|
| 22 |
|
| 23 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.021263618022203445,
|
| 5 |
+
"eval_runtime": 1.3658,
|
| 6 |
+
"eval_samples_per_second": 72.485,
|
| 7 |
+
"eval_steps_per_second": 2.929,
|
| 8 |
+
"total_flos": 2.7494650758139085e+18,
|
| 9 |
+
"train_loss": 0.2073148890797581,
|
| 10 |
+
"train_runtime": 1492.3317,
|
| 11 |
+
"train_samples_per_second": 23.775,
|
| 12 |
+
"train_steps_per_second": 0.188
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.021263618022203445,
|
| 5 |
+
"eval_runtime": 1.3658,
|
| 6 |
+
"eval_samples_per_second": 72.485,
|
| 7 |
+
"eval_steps_per_second": 2.929
|
| 8 |
+
}
|
runs/Mar11_00-37-16_89e2f3456c0c/events.out.tfevents.1710118958.89e2f3456c0c.2146.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24841351ae2575516e908f008d1bff94ea851e2e32c4bcb641dadcde731696af
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 2.7494650758139085e+18,
|
| 4 |
+
"train_loss": 0.2073148890797581,
|
| 5 |
+
"train_runtime": 1492.3317,
|
| 6 |
+
"train_samples_per_second": 23.775,
|
| 7 |
+
"train_steps_per_second": 0.188
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,551 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-dmae-va-U3-40A/checkpoint-119",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.37373737373737376,
|
| 14 |
+
"eval_loss": 1.329177737236023,
|
| 15 |
+
"eval_runtime": 1.2424,
|
| 16 |
+
"eval_samples_per_second": 79.683,
|
| 17 |
+
"eval_steps_per_second": 3.22,
|
| 18 |
+
"step": 7
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.71,
|
| 22 |
+
"grad_norm": 3.085270404815674,
|
| 23 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 24 |
+
"loss": 1.3407,
|
| 25 |
+
"step": 12
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_accuracy": 0.5151515151515151,
|
| 30 |
+
"eval_loss": 1.107913851737976,
|
| 31 |
+
"eval_runtime": 1.2457,
|
| 32 |
+
"eval_samples_per_second": 79.472,
|
| 33 |
+
"eval_steps_per_second": 3.211,
|
| 34 |
+
"step": 14
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 3.0,
|
| 38 |
+
"eval_accuracy": 0.6262626262626263,
|
| 39 |
+
"eval_loss": 0.8917983770370483,
|
| 40 |
+
"eval_runtime": 1.2574,
|
| 41 |
+
"eval_samples_per_second": 78.737,
|
| 42 |
+
"eval_steps_per_second": 3.181,
|
| 43 |
+
"step": 21
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.43,
|
| 47 |
+
"grad_norm": 3.550651788711548,
|
| 48 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 49 |
+
"loss": 0.9919,
|
| 50 |
+
"step": 24
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.7878787878787878,
|
| 55 |
+
"eval_loss": 0.6446935534477234,
|
| 56 |
+
"eval_runtime": 1.2679,
|
| 57 |
+
"eval_samples_per_second": 78.083,
|
| 58 |
+
"eval_steps_per_second": 3.155,
|
| 59 |
+
"step": 28
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 5.0,
|
| 63 |
+
"eval_accuracy": 0.8282828282828283,
|
| 64 |
+
"eval_loss": 0.45023515820503235,
|
| 65 |
+
"eval_runtime": 1.4064,
|
| 66 |
+
"eval_samples_per_second": 70.393,
|
| 67 |
+
"eval_steps_per_second": 2.844,
|
| 68 |
+
"step": 35
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.14,
|
| 72 |
+
"grad_norm": 3.4338462352752686,
|
| 73 |
+
"learning_rate": 4.841269841269841e-05,
|
| 74 |
+
"loss": 0.5761,
|
| 75 |
+
"step": 36
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 6.0,
|
| 79 |
+
"eval_accuracy": 0.9191919191919192,
|
| 80 |
+
"eval_loss": 0.27204275131225586,
|
| 81 |
+
"eval_runtime": 1.4909,
|
| 82 |
+
"eval_samples_per_second": 66.403,
|
| 83 |
+
"eval_steps_per_second": 2.683,
|
| 84 |
+
"step": 42
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.86,
|
| 88 |
+
"grad_norm": 1.6468698978424072,
|
| 89 |
+
"learning_rate": 4.603174603174603e-05,
|
| 90 |
+
"loss": 0.3111,
|
| 91 |
+
"step": 48
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 7.0,
|
| 95 |
+
"eval_accuracy": 0.9292929292929293,
|
| 96 |
+
"eval_loss": 0.23024092614650726,
|
| 97 |
+
"eval_runtime": 1.3066,
|
| 98 |
+
"eval_samples_per_second": 75.766,
|
| 99 |
+
"eval_steps_per_second": 3.061,
|
| 100 |
+
"step": 49
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 8.0,
|
| 104 |
+
"eval_accuracy": 0.9494949494949495,
|
| 105 |
+
"eval_loss": 0.16501076519489288,
|
| 106 |
+
"eval_runtime": 1.3125,
|
| 107 |
+
"eval_samples_per_second": 75.428,
|
| 108 |
+
"eval_steps_per_second": 3.048,
|
| 109 |
+
"step": 56
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 8.57,
|
| 113 |
+
"grad_norm": 1.7793350219726562,
|
| 114 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 115 |
+
"loss": 0.204,
|
| 116 |
+
"step": 60
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 9.0,
|
| 120 |
+
"eval_accuracy": 0.9494949494949495,
|
| 121 |
+
"eval_loss": 0.1503186672925949,
|
| 122 |
+
"eval_runtime": 1.316,
|
| 123 |
+
"eval_samples_per_second": 75.227,
|
| 124 |
+
"eval_steps_per_second": 3.039,
|
| 125 |
+
"step": 63
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 10.0,
|
| 129 |
+
"eval_accuracy": 0.9797979797979798,
|
| 130 |
+
"eval_loss": 0.08136877417564392,
|
| 131 |
+
"eval_runtime": 1.3015,
|
| 132 |
+
"eval_samples_per_second": 76.065,
|
| 133 |
+
"eval_steps_per_second": 3.073,
|
| 134 |
+
"step": 70
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.29,
|
| 138 |
+
"grad_norm": 1.3909555673599243,
|
| 139 |
+
"learning_rate": 4.126984126984127e-05,
|
| 140 |
+
"loss": 0.1518,
|
| 141 |
+
"step": 72
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 11.0,
|
| 145 |
+
"eval_accuracy": 0.9797979797979798,
|
| 146 |
+
"eval_loss": 0.06037978082895279,
|
| 147 |
+
"eval_runtime": 1.2829,
|
| 148 |
+
"eval_samples_per_second": 77.171,
|
| 149 |
+
"eval_steps_per_second": 3.118,
|
| 150 |
+
"step": 77
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 12.0,
|
| 154 |
+
"grad_norm": 1.9186725616455078,
|
| 155 |
+
"learning_rate": 3.888888888888889e-05,
|
| 156 |
+
"loss": 0.1272,
|
| 157 |
+
"step": 84
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 12.0,
|
| 161 |
+
"eval_accuracy": 0.9494949494949495,
|
| 162 |
+
"eval_loss": 0.1265028864145279,
|
| 163 |
+
"eval_runtime": 1.2863,
|
| 164 |
+
"eval_samples_per_second": 76.965,
|
| 165 |
+
"eval_steps_per_second": 3.11,
|
| 166 |
+
"step": 84
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 13.0,
|
| 170 |
+
"eval_accuracy": 0.9797979797979798,
|
| 171 |
+
"eval_loss": 0.05176503211259842,
|
| 172 |
+
"eval_runtime": 1.3002,
|
| 173 |
+
"eval_samples_per_second": 76.143,
|
| 174 |
+
"eval_steps_per_second": 3.076,
|
| 175 |
+
"step": 91
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 13.71,
|
| 179 |
+
"grad_norm": 1.4530500173568726,
|
| 180 |
+
"learning_rate": 3.650793650793651e-05,
|
| 181 |
+
"loss": 0.1379,
|
| 182 |
+
"step": 96
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 14.0,
|
| 186 |
+
"eval_accuracy": 0.98989898989899,
|
| 187 |
+
"eval_loss": 0.044787079095840454,
|
| 188 |
+
"eval_runtime": 1.303,
|
| 189 |
+
"eval_samples_per_second": 75.979,
|
| 190 |
+
"eval_steps_per_second": 3.07,
|
| 191 |
+
"step": 98
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 15.0,
|
| 195 |
+
"eval_accuracy": 0.98989898989899,
|
| 196 |
+
"eval_loss": 0.03611420467495918,
|
| 197 |
+
"eval_runtime": 1.309,
|
| 198 |
+
"eval_samples_per_second": 75.628,
|
| 199 |
+
"eval_steps_per_second": 3.056,
|
| 200 |
+
"step": 105
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 15.43,
|
| 204 |
+
"grad_norm": 1.3324140310287476,
|
| 205 |
+
"learning_rate": 3.412698412698413e-05,
|
| 206 |
+
"loss": 0.092,
|
| 207 |
+
"step": 108
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 16.0,
|
| 211 |
+
"eval_accuracy": 0.98989898989899,
|
| 212 |
+
"eval_loss": 0.032215822488069534,
|
| 213 |
+
"eval_runtime": 1.3002,
|
| 214 |
+
"eval_samples_per_second": 76.141,
|
| 215 |
+
"eval_steps_per_second": 3.076,
|
| 216 |
+
"step": 112
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 17.0,
|
| 220 |
+
"eval_accuracy": 1.0,
|
| 221 |
+
"eval_loss": 0.021263618022203445,
|
| 222 |
+
"eval_runtime": 1.307,
|
| 223 |
+
"eval_samples_per_second": 75.748,
|
| 224 |
+
"eval_steps_per_second": 3.061,
|
| 225 |
+
"step": 119
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 17.14,
|
| 229 |
+
"grad_norm": 2.2976629734039307,
|
| 230 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 231 |
+
"loss": 0.0762,
|
| 232 |
+
"step": 120
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 18.0,
|
| 236 |
+
"eval_accuracy": 0.98989898989899,
|
| 237 |
+
"eval_loss": 0.04688708856701851,
|
| 238 |
+
"eval_runtime": 1.315,
|
| 239 |
+
"eval_samples_per_second": 75.288,
|
| 240 |
+
"eval_steps_per_second": 3.042,
|
| 241 |
+
"step": 126
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 18.86,
|
| 245 |
+
"grad_norm": 2.308321714401245,
|
| 246 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 247 |
+
"loss": 0.0954,
|
| 248 |
+
"step": 132
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 19.0,
|
| 252 |
+
"eval_accuracy": 0.98989898989899,
|
| 253 |
+
"eval_loss": 0.061472997069358826,
|
| 254 |
+
"eval_runtime": 1.3031,
|
| 255 |
+
"eval_samples_per_second": 75.971,
|
| 256 |
+
"eval_steps_per_second": 3.07,
|
| 257 |
+
"step": 133
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 20.0,
|
| 261 |
+
"eval_accuracy": 0.98989898989899,
|
| 262 |
+
"eval_loss": 0.03133073449134827,
|
| 263 |
+
"eval_runtime": 1.2993,
|
| 264 |
+
"eval_samples_per_second": 76.194,
|
| 265 |
+
"eval_steps_per_second": 3.079,
|
| 266 |
+
"step": 140
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 20.57,
|
| 270 |
+
"grad_norm": 1.274895191192627,
|
| 271 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 272 |
+
"loss": 0.0795,
|
| 273 |
+
"step": 144
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 21.0,
|
| 277 |
+
"eval_accuracy": 0.98989898989899,
|
| 278 |
+
"eval_loss": 0.0380566343665123,
|
| 279 |
+
"eval_runtime": 1.3136,
|
| 280 |
+
"eval_samples_per_second": 75.368,
|
| 281 |
+
"eval_steps_per_second": 3.045,
|
| 282 |
+
"step": 147
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 22.0,
|
| 286 |
+
"eval_accuracy": 1.0,
|
| 287 |
+
"eval_loss": 0.013774486258625984,
|
| 288 |
+
"eval_runtime": 1.309,
|
| 289 |
+
"eval_samples_per_second": 75.628,
|
| 290 |
+
"eval_steps_per_second": 3.056,
|
| 291 |
+
"step": 154
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 22.29,
|
| 295 |
+
"grad_norm": 1.425993800163269,
|
| 296 |
+
"learning_rate": 2.4603174603174602e-05,
|
| 297 |
+
"loss": 0.077,
|
| 298 |
+
"step": 156
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 23.0,
|
| 302 |
+
"eval_accuracy": 1.0,
|
| 303 |
+
"eval_loss": 0.01703532226383686,
|
| 304 |
+
"eval_runtime": 1.3122,
|
| 305 |
+
"eval_samples_per_second": 75.446,
|
| 306 |
+
"eval_steps_per_second": 3.048,
|
| 307 |
+
"step": 161
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 24.0,
|
| 311 |
+
"grad_norm": 0.986053466796875,
|
| 312 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 313 |
+
"loss": 0.0675,
|
| 314 |
+
"step": 168
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 24.0,
|
| 318 |
+
"eval_accuracy": 1.0,
|
| 319 |
+
"eval_loss": 0.010675261728465557,
|
| 320 |
+
"eval_runtime": 1.3159,
|
| 321 |
+
"eval_samples_per_second": 75.231,
|
| 322 |
+
"eval_steps_per_second": 3.04,
|
| 323 |
+
"step": 168
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 25.0,
|
| 327 |
+
"eval_accuracy": 0.98989898989899,
|
| 328 |
+
"eval_loss": 0.019328022375702858,
|
| 329 |
+
"eval_runtime": 1.3069,
|
| 330 |
+
"eval_samples_per_second": 75.752,
|
| 331 |
+
"eval_steps_per_second": 3.061,
|
| 332 |
+
"step": 175
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 25.71,
|
| 336 |
+
"grad_norm": 2.7184066772460938,
|
| 337 |
+
"learning_rate": 1.984126984126984e-05,
|
| 338 |
+
"loss": 0.0659,
|
| 339 |
+
"step": 180
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 26.0,
|
| 343 |
+
"eval_accuracy": 0.98989898989899,
|
| 344 |
+
"eval_loss": 0.025486120954155922,
|
| 345 |
+
"eval_runtime": 1.3084,
|
| 346 |
+
"eval_samples_per_second": 75.667,
|
| 347 |
+
"eval_steps_per_second": 3.057,
|
| 348 |
+
"step": 182
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 27.0,
|
| 352 |
+
"eval_accuracy": 0.98989898989899,
|
| 353 |
+
"eval_loss": 0.020136240869760513,
|
| 354 |
+
"eval_runtime": 1.3096,
|
| 355 |
+
"eval_samples_per_second": 75.596,
|
| 356 |
+
"eval_steps_per_second": 3.054,
|
| 357 |
+
"step": 189
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 27.43,
|
| 361 |
+
"grad_norm": 2.4670774936676025,
|
| 362 |
+
"learning_rate": 1.746031746031746e-05,
|
| 363 |
+
"loss": 0.0758,
|
| 364 |
+
"step": 192
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 28.0,
|
| 368 |
+
"eval_accuracy": 0.98989898989899,
|
| 369 |
+
"eval_loss": 0.03251149132847786,
|
| 370 |
+
"eval_runtime": 1.2895,
|
| 371 |
+
"eval_samples_per_second": 76.775,
|
| 372 |
+
"eval_steps_per_second": 3.102,
|
| 373 |
+
"step": 196
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 29.0,
|
| 377 |
+
"eval_accuracy": 1.0,
|
| 378 |
+
"eval_loss": 0.011012612842023373,
|
| 379 |
+
"eval_runtime": 1.3105,
|
| 380 |
+
"eval_samples_per_second": 75.542,
|
| 381 |
+
"eval_steps_per_second": 3.052,
|
| 382 |
+
"step": 203
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 29.14,
|
| 386 |
+
"grad_norm": 1.395484209060669,
|
| 387 |
+
"learning_rate": 1.5079365079365079e-05,
|
| 388 |
+
"loss": 0.0589,
|
| 389 |
+
"step": 204
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 30.0,
|
| 393 |
+
"eval_accuracy": 1.0,
|
| 394 |
+
"eval_loss": 0.0159281175583601,
|
| 395 |
+
"eval_runtime": 1.3316,
|
| 396 |
+
"eval_samples_per_second": 74.346,
|
| 397 |
+
"eval_steps_per_second": 3.004,
|
| 398 |
+
"step": 210
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 30.86,
|
| 402 |
+
"grad_norm": 1.4966486692428589,
|
| 403 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 404 |
+
"loss": 0.0521,
|
| 405 |
+
"step": 216
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 31.0,
|
| 409 |
+
"eval_accuracy": 0.98989898989899,
|
| 410 |
+
"eval_loss": 0.031857237219810486,
|
| 411 |
+
"eval_runtime": 1.2919,
|
| 412 |
+
"eval_samples_per_second": 76.629,
|
| 413 |
+
"eval_steps_per_second": 3.096,
|
| 414 |
+
"step": 217
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 32.0,
|
| 418 |
+
"eval_accuracy": 0.9797979797979798,
|
| 419 |
+
"eval_loss": 0.029438314959406853,
|
| 420 |
+
"eval_runtime": 1.3011,
|
| 421 |
+
"eval_samples_per_second": 76.091,
|
| 422 |
+
"eval_steps_per_second": 3.074,
|
| 423 |
+
"step": 224
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 32.57,
|
| 427 |
+
"grad_norm": 1.3882635831832886,
|
| 428 |
+
"learning_rate": 1.0317460317460318e-05,
|
| 429 |
+
"loss": 0.0618,
|
| 430 |
+
"step": 228
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 33.0,
|
| 434 |
+
"eval_accuracy": 0.9797979797979798,
|
| 435 |
+
"eval_loss": 0.039191748946905136,
|
| 436 |
+
"eval_runtime": 1.3063,
|
| 437 |
+
"eval_samples_per_second": 75.788,
|
| 438 |
+
"eval_steps_per_second": 3.062,
|
| 439 |
+
"step": 231
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 34.0,
|
| 443 |
+
"eval_accuracy": 0.98989898989899,
|
| 444 |
+
"eval_loss": 0.026867415755987167,
|
| 445 |
+
"eval_runtime": 1.2944,
|
| 446 |
+
"eval_samples_per_second": 76.481,
|
| 447 |
+
"eval_steps_per_second": 3.09,
|
| 448 |
+
"step": 238
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 34.29,
|
| 452 |
+
"grad_norm": 1.203762412071228,
|
| 453 |
+
"learning_rate": 7.936507936507936e-06,
|
| 454 |
+
"loss": 0.0422,
|
| 455 |
+
"step": 240
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 35.0,
|
| 459 |
+
"eval_accuracy": 0.98989898989899,
|
| 460 |
+
"eval_loss": 0.021003253757953644,
|
| 461 |
+
"eval_runtime": 1.3003,
|
| 462 |
+
"eval_samples_per_second": 76.137,
|
| 463 |
+
"eval_steps_per_second": 3.076,
|
| 464 |
+
"step": 245
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 36.0,
|
| 468 |
+
"grad_norm": 1.9063193798065186,
|
| 469 |
+
"learning_rate": 5.555555555555556e-06,
|
| 470 |
+
"loss": 0.0551,
|
| 471 |
+
"step": 252
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 36.0,
|
| 475 |
+
"eval_accuracy": 0.98989898989899,
|
| 476 |
+
"eval_loss": 0.01777764968574047,
|
| 477 |
+
"eval_runtime": 1.456,
|
| 478 |
+
"eval_samples_per_second": 67.993,
|
| 479 |
+
"eval_steps_per_second": 2.747,
|
| 480 |
+
"step": 252
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 37.0,
|
| 484 |
+
"eval_accuracy": 0.98989898989899,
|
| 485 |
+
"eval_loss": 0.01593274623155594,
|
| 486 |
+
"eval_runtime": 1.3158,
|
| 487 |
+
"eval_samples_per_second": 75.242,
|
| 488 |
+
"eval_steps_per_second": 3.04,
|
| 489 |
+
"step": 259
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 37.71,
|
| 493 |
+
"grad_norm": 1.4439316987991333,
|
| 494 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 495 |
+
"loss": 0.0518,
|
| 496 |
+
"step": 264
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 38.0,
|
| 500 |
+
"eval_accuracy": 0.98989898989899,
|
| 501 |
+
"eval_loss": 0.012379774823784828,
|
| 502 |
+
"eval_runtime": 1.3454,
|
| 503 |
+
"eval_samples_per_second": 73.583,
|
| 504 |
+
"eval_steps_per_second": 2.973,
|
| 505 |
+
"step": 266
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 39.0,
|
| 509 |
+
"eval_accuracy": 1.0,
|
| 510 |
+
"eval_loss": 0.011175006628036499,
|
| 511 |
+
"eval_runtime": 1.4492,
|
| 512 |
+
"eval_samples_per_second": 68.315,
|
| 513 |
+
"eval_steps_per_second": 2.76,
|
| 514 |
+
"step": 273
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 39.43,
|
| 518 |
+
"grad_norm": 1.1414995193481445,
|
| 519 |
+
"learning_rate": 7.936507936507937e-07,
|
| 520 |
+
"loss": 0.0313,
|
| 521 |
+
"step": 276
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 40.0,
|
| 525 |
+
"eval_accuracy": 1.0,
|
| 526 |
+
"eval_loss": 0.01096320990473032,
|
| 527 |
+
"eval_runtime": 1.3095,
|
| 528 |
+
"eval_samples_per_second": 75.6,
|
| 529 |
+
"eval_steps_per_second": 3.055,
|
| 530 |
+
"step": 280
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 40.0,
|
| 534 |
+
"step": 280,
|
| 535 |
+
"total_flos": 2.7494650758139085e+18,
|
| 536 |
+
"train_loss": 0.2073148890797581,
|
| 537 |
+
"train_runtime": 1492.3317,
|
| 538 |
+
"train_samples_per_second": 23.775,
|
| 539 |
+
"train_steps_per_second": 0.188
|
| 540 |
+
}
|
| 541 |
+
],
|
| 542 |
+
"logging_steps": 12,
|
| 543 |
+
"max_steps": 280,
|
| 544 |
+
"num_input_tokens_seen": 0,
|
| 545 |
+
"num_train_epochs": 40,
|
| 546 |
+
"save_steps": 500,
|
| 547 |
+
"total_flos": 2.7494650758139085e+18,
|
| 548 |
+
"train_batch_size": 32,
|
| 549 |
+
"trial_name": null,
|
| 550 |
+
"trial_params": null
|
| 551 |
+
}
|