Commit
•
4126ca4
1
Parent(s):
e4016d4
End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr02_02-48-53_d3cc9417b200/events.out.tfevents.1712027896.d3cc9417b200.2135.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +575 -0
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 1.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.0039
|
| 21 |
+
- Accuracy: 0.8167
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.94,
|
| 3 |
+
"eval_accuracy": 0.8166666666666667,
|
| 4 |
+
"eval_loss": 1.003859043121338,
|
| 5 |
+
"eval_runtime": 1.6285,
|
| 6 |
+
"eval_samples_per_second": 36.843,
|
| 7 |
+
"eval_steps_per_second": 1.228,
|
| 8 |
+
"total_flos": 2.8633958865108173e+18,
|
| 9 |
+
"train_loss": 0.23703576421656578,
|
| 10 |
+
"train_runtime": 1686.3425,
|
| 11 |
+
"train_samples_per_second": 24.258,
|
| 12 |
+
"train_steps_per_second": 0.174
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.94,
|
| 3 |
+
"eval_accuracy": 0.8166666666666667,
|
| 4 |
+
"eval_loss": 1.003859043121338,
|
| 5 |
+
"eval_runtime": 1.6285,
|
| 6 |
+
"eval_samples_per_second": 36.843,
|
| 7 |
+
"eval_steps_per_second": 1.228
|
| 8 |
+
}
|
runs/Apr02_02-48-53_d3cc9417b200/events.out.tfevents.1712027896.d3cc9417b200.2135.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87521e4f88f7a2c347fa568c83fd1fca3546d5f845a0b33865423b4323b7a47f
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.94,
|
| 3 |
+
"total_flos": 2.8633958865108173e+18,
|
| 4 |
+
"train_loss": 0.23703576421656578,
|
| 5 |
+
"train_runtime": 1686.3425,
|
| 6 |
+
"train_samples_per_second": 24.258,
|
| 7 |
+
"train_steps_per_second": 0.174
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,575 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8166666666666667,
|
| 3 |
+
"best_model_checkpoint": "beit-base-patch16-224-dmae-va-U5-42/checkpoint-224",
|
| 4 |
+
"epoch": 37.935483870967744,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 294,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9,
|
| 13 |
+
"eval_accuracy": 0.4666666666666667,
|
| 14 |
+
"eval_loss": 1.347098708152771,
|
| 15 |
+
"eval_runtime": 1.5986,
|
| 16 |
+
"eval_samples_per_second": 37.533,
|
| 17 |
+
"eval_steps_per_second": 1.251,
|
| 18 |
+
"step": 7
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.29,
|
| 22 |
+
"grad_norm": 10.756444931030273,
|
| 23 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 24 |
+
"loss": 1.6023,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.94,
|
| 29 |
+
"eval_accuracy": 0.5833333333333334,
|
| 30 |
+
"eval_loss": 1.0873388051986694,
|
| 31 |
+
"eval_runtime": 1.6091,
|
| 32 |
+
"eval_samples_per_second": 37.288,
|
| 33 |
+
"eval_steps_per_second": 1.243,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.58,
|
| 38 |
+
"grad_norm": 9.189943313598633,
|
| 39 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 40 |
+
"loss": 1.1509,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.97,
|
| 45 |
+
"eval_accuracy": 0.5833333333333334,
|
| 46 |
+
"eval_loss": 0.9947898983955383,
|
| 47 |
+
"eval_runtime": 2.0991,
|
| 48 |
+
"eval_samples_per_second": 28.584,
|
| 49 |
+
"eval_steps_per_second": 0.953,
|
| 50 |
+
"step": 23
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.87,
|
| 54 |
+
"grad_norm": 11.76810073852539,
|
| 55 |
+
"learning_rate": 5e-05,
|
| 56 |
+
"loss": 0.826,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"eval_accuracy": 0.7166666666666667,
|
| 62 |
+
"eval_loss": 0.7244278192520142,
|
| 63 |
+
"eval_runtime": 1.6205,
|
| 64 |
+
"eval_samples_per_second": 37.025,
|
| 65 |
+
"eval_steps_per_second": 1.234,
|
| 66 |
+
"step": 31
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.9,
|
| 70 |
+
"eval_accuracy": 0.7333333333333333,
|
| 71 |
+
"eval_loss": 0.5740881562232971,
|
| 72 |
+
"eval_runtime": 2.1053,
|
| 73 |
+
"eval_samples_per_second": 28.5,
|
| 74 |
+
"eval_steps_per_second": 0.95,
|
| 75 |
+
"step": 38
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.16,
|
| 79 |
+
"grad_norm": 10.483253479003906,
|
| 80 |
+
"learning_rate": 4.810606060606061e-05,
|
| 81 |
+
"loss": 0.5551,
|
| 82 |
+
"step": 40
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.94,
|
| 86 |
+
"eval_accuracy": 0.75,
|
| 87 |
+
"eval_loss": 0.6568958759307861,
|
| 88 |
+
"eval_runtime": 2.0199,
|
| 89 |
+
"eval_samples_per_second": 29.704,
|
| 90 |
+
"eval_steps_per_second": 0.99,
|
| 91 |
+
"step": 46
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.45,
|
| 95 |
+
"grad_norm": 6.183708667755127,
|
| 96 |
+
"learning_rate": 4.621212121212121e-05,
|
| 97 |
+
"loss": 0.3649,
|
| 98 |
+
"step": 50
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.97,
|
| 102 |
+
"eval_accuracy": 0.7166666666666667,
|
| 103 |
+
"eval_loss": 0.6321703195571899,
|
| 104 |
+
"eval_runtime": 1.6185,
|
| 105 |
+
"eval_samples_per_second": 37.07,
|
| 106 |
+
"eval_steps_per_second": 1.236,
|
| 107 |
+
"step": 54
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 7.74,
|
| 111 |
+
"grad_norm": 8.211569786071777,
|
| 112 |
+
"learning_rate": 4.431818181818182e-05,
|
| 113 |
+
"loss": 0.2592,
|
| 114 |
+
"step": 60
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 8.0,
|
| 118 |
+
"eval_accuracy": 0.7333333333333333,
|
| 119 |
+
"eval_loss": 0.6993545889854431,
|
| 120 |
+
"eval_runtime": 1.7914,
|
| 121 |
+
"eval_samples_per_second": 33.494,
|
| 122 |
+
"eval_steps_per_second": 1.116,
|
| 123 |
+
"step": 62
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 8.9,
|
| 127 |
+
"eval_accuracy": 0.7333333333333333,
|
| 128 |
+
"eval_loss": 0.6589908599853516,
|
| 129 |
+
"eval_runtime": 1.6239,
|
| 130 |
+
"eval_samples_per_second": 36.949,
|
| 131 |
+
"eval_steps_per_second": 1.232,
|
| 132 |
+
"step": 69
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.03,
|
| 136 |
+
"grad_norm": 6.782934665679932,
|
| 137 |
+
"learning_rate": 4.242424242424243e-05,
|
| 138 |
+
"loss": 0.1958,
|
| 139 |
+
"step": 70
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 9.94,
|
| 143 |
+
"eval_accuracy": 0.7666666666666667,
|
| 144 |
+
"eval_loss": 0.6845510601997375,
|
| 145 |
+
"eval_runtime": 1.8061,
|
| 146 |
+
"eval_samples_per_second": 33.22,
|
| 147 |
+
"eval_steps_per_second": 1.107,
|
| 148 |
+
"step": 77
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.32,
|
| 152 |
+
"grad_norm": 6.670173645019531,
|
| 153 |
+
"learning_rate": 4.053030303030303e-05,
|
| 154 |
+
"loss": 0.1664,
|
| 155 |
+
"step": 80
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 10.97,
|
| 159 |
+
"eval_accuracy": 0.7666666666666667,
|
| 160 |
+
"eval_loss": 0.7165997624397278,
|
| 161 |
+
"eval_runtime": 1.6803,
|
| 162 |
+
"eval_samples_per_second": 35.708,
|
| 163 |
+
"eval_steps_per_second": 1.19,
|
| 164 |
+
"step": 85
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 11.61,
|
| 168 |
+
"grad_norm": 7.649813175201416,
|
| 169 |
+
"learning_rate": 3.8636363636363636e-05,
|
| 170 |
+
"loss": 0.1571,
|
| 171 |
+
"step": 90
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 12.0,
|
| 175 |
+
"eval_accuracy": 0.7833333333333333,
|
| 176 |
+
"eval_loss": 0.7842047214508057,
|
| 177 |
+
"eval_runtime": 1.6667,
|
| 178 |
+
"eval_samples_per_second": 35.999,
|
| 179 |
+
"eval_steps_per_second": 1.2,
|
| 180 |
+
"step": 93
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 12.9,
|
| 184 |
+
"grad_norm": 5.415360927581787,
|
| 185 |
+
"learning_rate": 3.6742424242424246e-05,
|
| 186 |
+
"loss": 0.1174,
|
| 187 |
+
"step": 100
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 12.9,
|
| 191 |
+
"eval_accuracy": 0.8,
|
| 192 |
+
"eval_loss": 0.8464832901954651,
|
| 193 |
+
"eval_runtime": 1.8073,
|
| 194 |
+
"eval_samples_per_second": 33.198,
|
| 195 |
+
"eval_steps_per_second": 1.107,
|
| 196 |
+
"step": 100
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 13.94,
|
| 200 |
+
"eval_accuracy": 0.7666666666666667,
|
| 201 |
+
"eval_loss": 0.9116414189338684,
|
| 202 |
+
"eval_runtime": 1.6421,
|
| 203 |
+
"eval_samples_per_second": 36.539,
|
| 204 |
+
"eval_steps_per_second": 1.218,
|
| 205 |
+
"step": 108
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 14.19,
|
| 209 |
+
"grad_norm": 3.7090325355529785,
|
| 210 |
+
"learning_rate": 3.484848484848485e-05,
|
| 211 |
+
"loss": 0.0956,
|
| 212 |
+
"step": 110
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 14.97,
|
| 216 |
+
"eval_accuracy": 0.75,
|
| 217 |
+
"eval_loss": 0.9740654230117798,
|
| 218 |
+
"eval_runtime": 1.6076,
|
| 219 |
+
"eval_samples_per_second": 37.322,
|
| 220 |
+
"eval_steps_per_second": 1.244,
|
| 221 |
+
"step": 116
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 15.48,
|
| 225 |
+
"grad_norm": 7.993117809295654,
|
| 226 |
+
"learning_rate": 3.295454545454545e-05,
|
| 227 |
+
"loss": 0.1252,
|
| 228 |
+
"step": 120
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 16.0,
|
| 232 |
+
"eval_accuracy": 0.8,
|
| 233 |
+
"eval_loss": 0.7759976983070374,
|
| 234 |
+
"eval_runtime": 1.6595,
|
| 235 |
+
"eval_samples_per_second": 36.154,
|
| 236 |
+
"eval_steps_per_second": 1.205,
|
| 237 |
+
"step": 124
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 16.77,
|
| 241 |
+
"grad_norm": 3.9146523475646973,
|
| 242 |
+
"learning_rate": 3.106060606060606e-05,
|
| 243 |
+
"loss": 0.0933,
|
| 244 |
+
"step": 130
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 16.9,
|
| 248 |
+
"eval_accuracy": 0.7833333333333333,
|
| 249 |
+
"eval_loss": 0.9424102902412415,
|
| 250 |
+
"eval_runtime": 2.1274,
|
| 251 |
+
"eval_samples_per_second": 28.203,
|
| 252 |
+
"eval_steps_per_second": 0.94,
|
| 253 |
+
"step": 131
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 17.94,
|
| 257 |
+
"eval_accuracy": 0.7333333333333333,
|
| 258 |
+
"eval_loss": 1.0444563627243042,
|
| 259 |
+
"eval_runtime": 1.6154,
|
| 260 |
+
"eval_samples_per_second": 37.143,
|
| 261 |
+
"eval_steps_per_second": 1.238,
|
| 262 |
+
"step": 139
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 18.06,
|
| 266 |
+
"grad_norm": 7.39138126373291,
|
| 267 |
+
"learning_rate": 2.916666666666667e-05,
|
| 268 |
+
"loss": 0.1455,
|
| 269 |
+
"step": 140
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 18.97,
|
| 273 |
+
"eval_accuracy": 0.7333333333333333,
|
| 274 |
+
"eval_loss": 0.8524726629257202,
|
| 275 |
+
"eval_runtime": 1.681,
|
| 276 |
+
"eval_samples_per_second": 35.693,
|
| 277 |
+
"eval_steps_per_second": 1.19,
|
| 278 |
+
"step": 147
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 19.35,
|
| 282 |
+
"grad_norm": 3.7276432514190674,
|
| 283 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 284 |
+
"loss": 0.1034,
|
| 285 |
+
"step": 150
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 20.0,
|
| 289 |
+
"eval_accuracy": 0.7666666666666667,
|
| 290 |
+
"eval_loss": 0.822151243686676,
|
| 291 |
+
"eval_runtime": 2.0162,
|
| 292 |
+
"eval_samples_per_second": 29.759,
|
| 293 |
+
"eval_steps_per_second": 0.992,
|
| 294 |
+
"step": 155
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 20.65,
|
| 298 |
+
"grad_norm": 4.372833728790283,
|
| 299 |
+
"learning_rate": 2.537878787878788e-05,
|
| 300 |
+
"loss": 0.0855,
|
| 301 |
+
"step": 160
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 20.9,
|
| 305 |
+
"eval_accuracy": 0.7833333333333333,
|
| 306 |
+
"eval_loss": 0.8990534543991089,
|
| 307 |
+
"eval_runtime": 2.1376,
|
| 308 |
+
"eval_samples_per_second": 28.069,
|
| 309 |
+
"eval_steps_per_second": 0.936,
|
| 310 |
+
"step": 162
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 21.94,
|
| 314 |
+
"grad_norm": 6.204973220825195,
|
| 315 |
+
"learning_rate": 2.3484848484848487e-05,
|
| 316 |
+
"loss": 0.0985,
|
| 317 |
+
"step": 170
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 21.94,
|
| 321 |
+
"eval_accuracy": 0.8,
|
| 322 |
+
"eval_loss": 0.8954617977142334,
|
| 323 |
+
"eval_runtime": 1.7159,
|
| 324 |
+
"eval_samples_per_second": 34.968,
|
| 325 |
+
"eval_steps_per_second": 1.166,
|
| 326 |
+
"step": 170
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 22.97,
|
| 330 |
+
"eval_accuracy": 0.7666666666666667,
|
| 331 |
+
"eval_loss": 0.9603295922279358,
|
| 332 |
+
"eval_runtime": 1.628,
|
| 333 |
+
"eval_samples_per_second": 36.854,
|
| 334 |
+
"eval_steps_per_second": 1.228,
|
| 335 |
+
"step": 178
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 23.23,
|
| 339 |
+
"grad_norm": 2.5669796466827393,
|
| 340 |
+
"learning_rate": 2.1590909090909093e-05,
|
| 341 |
+
"loss": 0.087,
|
| 342 |
+
"step": 180
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 24.0,
|
| 346 |
+
"eval_accuracy": 0.7833333333333333,
|
| 347 |
+
"eval_loss": 0.9932332634925842,
|
| 348 |
+
"eval_runtime": 1.6549,
|
| 349 |
+
"eval_samples_per_second": 36.255,
|
| 350 |
+
"eval_steps_per_second": 1.209,
|
| 351 |
+
"step": 186
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 24.52,
|
| 355 |
+
"grad_norm": 4.33466911315918,
|
| 356 |
+
"learning_rate": 1.9696969696969697e-05,
|
| 357 |
+
"loss": 0.0832,
|
| 358 |
+
"step": 190
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 24.9,
|
| 362 |
+
"eval_accuracy": 0.7833333333333333,
|
| 363 |
+
"eval_loss": 1.0099666118621826,
|
| 364 |
+
"eval_runtime": 1.6219,
|
| 365 |
+
"eval_samples_per_second": 36.994,
|
| 366 |
+
"eval_steps_per_second": 1.233,
|
| 367 |
+
"step": 193
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 25.81,
|
| 371 |
+
"grad_norm": 2.6737489700317383,
|
| 372 |
+
"learning_rate": 1.7803030303030303e-05,
|
| 373 |
+
"loss": 0.0632,
|
| 374 |
+
"step": 200
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 25.94,
|
| 378 |
+
"eval_accuracy": 0.7666666666666667,
|
| 379 |
+
"eval_loss": 0.9393168091773987,
|
| 380 |
+
"eval_runtime": 1.62,
|
| 381 |
+
"eval_samples_per_second": 37.037,
|
| 382 |
+
"eval_steps_per_second": 1.235,
|
| 383 |
+
"step": 201
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 26.97,
|
| 387 |
+
"eval_accuracy": 0.7833333333333333,
|
| 388 |
+
"eval_loss": 0.9061955213546753,
|
| 389 |
+
"eval_runtime": 2.0226,
|
| 390 |
+
"eval_samples_per_second": 29.665,
|
| 391 |
+
"eval_steps_per_second": 0.989,
|
| 392 |
+
"step": 209
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 27.1,
|
| 396 |
+
"grad_norm": 6.029117107391357,
|
| 397 |
+
"learning_rate": 1.590909090909091e-05,
|
| 398 |
+
"loss": 0.0778,
|
| 399 |
+
"step": 210
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 28.0,
|
| 403 |
+
"eval_accuracy": 0.8,
|
| 404 |
+
"eval_loss": 0.9339290857315063,
|
| 405 |
+
"eval_runtime": 2.1654,
|
| 406 |
+
"eval_samples_per_second": 27.708,
|
| 407 |
+
"eval_steps_per_second": 0.924,
|
| 408 |
+
"step": 217
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 28.39,
|
| 412 |
+
"grad_norm": 3.781505584716797,
|
| 413 |
+
"learning_rate": 1.4015151515151515e-05,
|
| 414 |
+
"loss": 0.0627,
|
| 415 |
+
"step": 220
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 28.9,
|
| 419 |
+
"eval_accuracy": 0.8166666666666667,
|
| 420 |
+
"eval_loss": 1.003859043121338,
|
| 421 |
+
"eval_runtime": 1.6395,
|
| 422 |
+
"eval_samples_per_second": 36.596,
|
| 423 |
+
"eval_steps_per_second": 1.22,
|
| 424 |
+
"step": 224
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 29.68,
|
| 428 |
+
"grad_norm": 8.317666053771973,
|
| 429 |
+
"learning_rate": 1.2121212121212122e-05,
|
| 430 |
+
"loss": 0.0837,
|
| 431 |
+
"step": 230
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 29.94,
|
| 435 |
+
"eval_accuracy": 0.7333333333333333,
|
| 436 |
+
"eval_loss": 1.0636054277420044,
|
| 437 |
+
"eval_runtime": 2.1368,
|
| 438 |
+
"eval_samples_per_second": 28.079,
|
| 439 |
+
"eval_steps_per_second": 0.936,
|
| 440 |
+
"step": 232
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 30.97,
|
| 444 |
+
"grad_norm": 4.580255508422852,
|
| 445 |
+
"learning_rate": 1.0227272727272729e-05,
|
| 446 |
+
"loss": 0.0595,
|
| 447 |
+
"step": 240
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 30.97,
|
| 451 |
+
"eval_accuracy": 0.75,
|
| 452 |
+
"eval_loss": 1.0424180030822754,
|
| 453 |
+
"eval_runtime": 1.9745,
|
| 454 |
+
"eval_samples_per_second": 30.387,
|
| 455 |
+
"eval_steps_per_second": 1.013,
|
| 456 |
+
"step": 240
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 32.0,
|
| 460 |
+
"eval_accuracy": 0.8,
|
| 461 |
+
"eval_loss": 1.0514304637908936,
|
| 462 |
+
"eval_runtime": 1.6234,
|
| 463 |
+
"eval_samples_per_second": 36.96,
|
| 464 |
+
"eval_steps_per_second": 1.232,
|
| 465 |
+
"step": 248
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 32.26,
|
| 469 |
+
"grad_norm": 3.7219626903533936,
|
| 470 |
+
"learning_rate": 8.333333333333334e-06,
|
| 471 |
+
"loss": 0.0706,
|
| 472 |
+
"step": 250
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 32.9,
|
| 476 |
+
"eval_accuracy": 0.7833333333333333,
|
| 477 |
+
"eval_loss": 1.0638715028762817,
|
| 478 |
+
"eval_runtime": 1.6117,
|
| 479 |
+
"eval_samples_per_second": 37.228,
|
| 480 |
+
"eval_steps_per_second": 1.241,
|
| 481 |
+
"step": 255
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 33.55,
|
| 485 |
+
"grad_norm": 2.273451566696167,
|
| 486 |
+
"learning_rate": 6.43939393939394e-06,
|
| 487 |
+
"loss": 0.0565,
|
| 488 |
+
"step": 260
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 33.94,
|
| 492 |
+
"eval_accuracy": 0.7666666666666667,
|
| 493 |
+
"eval_loss": 1.0494216680526733,
|
| 494 |
+
"eval_runtime": 1.9391,
|
| 495 |
+
"eval_samples_per_second": 30.942,
|
| 496 |
+
"eval_steps_per_second": 1.031,
|
| 497 |
+
"step": 263
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 34.84,
|
| 501 |
+
"grad_norm": 2.9928078651428223,
|
| 502 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 503 |
+
"loss": 0.0515,
|
| 504 |
+
"step": 270
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 34.97,
|
| 508 |
+
"eval_accuracy": 0.7666666666666667,
|
| 509 |
+
"eval_loss": 1.0627790689468384,
|
| 510 |
+
"eval_runtime": 1.5942,
|
| 511 |
+
"eval_samples_per_second": 37.636,
|
| 512 |
+
"eval_steps_per_second": 1.255,
|
| 513 |
+
"step": 271
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 36.0,
|
| 517 |
+
"eval_accuracy": 0.7666666666666667,
|
| 518 |
+
"eval_loss": 1.1089335680007935,
|
| 519 |
+
"eval_runtime": 1.74,
|
| 520 |
+
"eval_samples_per_second": 34.484,
|
| 521 |
+
"eval_steps_per_second": 1.149,
|
| 522 |
+
"step": 279
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 36.13,
|
| 526 |
+
"grad_norm": 6.05413818359375,
|
| 527 |
+
"learning_rate": 2.651515151515152e-06,
|
| 528 |
+
"loss": 0.0614,
|
| 529 |
+
"step": 280
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 36.9,
|
| 533 |
+
"eval_accuracy": 0.8,
|
| 534 |
+
"eval_loss": 1.086135983467102,
|
| 535 |
+
"eval_runtime": 1.5851,
|
| 536 |
+
"eval_samples_per_second": 37.853,
|
| 537 |
+
"eval_steps_per_second": 1.262,
|
| 538 |
+
"step": 286
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 37.42,
|
| 542 |
+
"grad_norm": 3.7136027812957764,
|
| 543 |
+
"learning_rate": 7.575757575757576e-07,
|
| 544 |
+
"loss": 0.0496,
|
| 545 |
+
"step": 290
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 37.94,
|
| 549 |
+
"eval_accuracy": 0.8,
|
| 550 |
+
"eval_loss": 1.0713495016098022,
|
| 551 |
+
"eval_runtime": 1.6733,
|
| 552 |
+
"eval_samples_per_second": 35.857,
|
| 553 |
+
"eval_steps_per_second": 1.195,
|
| 554 |
+
"step": 294
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 37.94,
|
| 558 |
+
"step": 294,
|
| 559 |
+
"total_flos": 2.8633958865108173e+18,
|
| 560 |
+
"train_loss": 0.23703576421656578,
|
| 561 |
+
"train_runtime": 1686.3425,
|
| 562 |
+
"train_samples_per_second": 24.258,
|
| 563 |
+
"train_steps_per_second": 0.174
|
| 564 |
+
}
|
| 565 |
+
],
|
| 566 |
+
"logging_steps": 10,
|
| 567 |
+
"max_steps": 294,
|
| 568 |
+
"num_input_tokens_seen": 0,
|
| 569 |
+
"num_train_epochs": 42,
|
| 570 |
+
"save_steps": 500,
|
| 571 |
+
"total_flos": 2.8633958865108173e+18,
|
| 572 |
+
"train_batch_size": 32,
|
| 573 |
+
"trial_name": null,
|
| 574 |
+
"trial_params": null
|
| 575 |
+
}
|