Commit
•
4489b50
1
Parent(s):
7dadbb2
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Sep19_04-00-17_47da270bdea6/events.out.tfevents.1726722924.47da270bdea6.486.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +576 -0
README.md
CHANGED
|
@@ -33,7 +33,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [WinKawaks/vit-tiny-patch16-224](https://huggingface.co/WinKawaks/vit-tiny-patch16-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
- Accuracy: 0.9338
|
| 38 |
|
| 39 |
## Model description
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [WinKawaks/vit-tiny-patch16-224](https://huggingface.co/WinKawaks/vit-tiny-patch16-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2121
|
| 37 |
- Accuracy: 0.9338
|
| 38 |
|
| 39 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.901960784313726,
|
| 3 |
+
"eval_accuracy": 0.9338235294117647,
|
| 4 |
+
"eval_loss": 0.21208395063877106,
|
| 5 |
+
"eval_runtime": 33.0507,
|
| 6 |
+
"eval_samples_per_second": 4.115,
|
| 7 |
+
"eval_steps_per_second": 0.514,
|
| 8 |
+
"total_flos": 9.102711432019968e+16,
|
| 9 |
+
"train_loss": 0.39481510582723117,
|
| 10 |
+
"train_runtime": 4386.4067,
|
| 11 |
+
"train_samples_per_second": 4.186,
|
| 12 |
+
"train_steps_per_second": 0.13
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.901960784313726,
|
| 3 |
+
"eval_accuracy": 0.9338235294117647,
|
| 4 |
+
"eval_loss": 0.21208395063877106,
|
| 5 |
+
"eval_runtime": 33.0507,
|
| 6 |
+
"eval_samples_per_second": 4.115,
|
| 7 |
+
"eval_steps_per_second": 0.514
|
| 8 |
+
}
|
runs/Sep19_04-00-17_47da270bdea6/events.out.tfevents.1726722924.47da270bdea6.486.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6faa2c31d46b30c0ad8dcbcd78cec3b45469e4deea1188c6e9385f62b4cae9d8
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.901960784313726,
|
| 3 |
+
"total_flos": 9.102711432019968e+16,
|
| 4 |
+
"train_loss": 0.39481510582723117,
|
| 5 |
+
"train_runtime": 4386.4067,
|
| 6 |
+
"train_samples_per_second": 4.186,
|
| 7 |
+
"train_steps_per_second": 0.13
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,576 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9338235294117647,
|
| 3 |
+
"best_model_checkpoint": "vit-tiny-patch16-224-finetuned-papsmear/checkpoint-382",
|
| 4 |
+
"epoch": 14.901960784313726,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 570,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26143790849673204,
|
| 13 |
+
"grad_norm": 22.882160186767578,
|
| 14 |
+
"learning_rate": 8.771929824561403e-06,
|
| 15 |
+
"loss": 1.9804,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5228758169934641,
|
| 20 |
+
"grad_norm": 17.451801300048828,
|
| 21 |
+
"learning_rate": 1.7543859649122806e-05,
|
| 22 |
+
"loss": 1.6798,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7843137254901961,
|
| 27 |
+
"grad_norm": 14.190028190612793,
|
| 28 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 29 |
+
"loss": 1.4005,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9934640522875817,
|
| 34 |
+
"eval_accuracy": 0.5294117647058824,
|
| 35 |
+
"eval_loss": 1.2214075326919556,
|
| 36 |
+
"eval_runtime": 30.1025,
|
| 37 |
+
"eval_samples_per_second": 4.518,
|
| 38 |
+
"eval_steps_per_second": 0.565,
|
| 39 |
+
"step": 38
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0457516339869282,
|
| 43 |
+
"grad_norm": 17.032344818115234,
|
| 44 |
+
"learning_rate": 3.508771929824561e-05,
|
| 45 |
+
"loss": 1.2065,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.3071895424836601,
|
| 50 |
+
"grad_norm": 13.36404800415039,
|
| 51 |
+
"learning_rate": 4.3859649122807014e-05,
|
| 52 |
+
"loss": 1.0837,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.5686274509803921,
|
| 57 |
+
"grad_norm": 19.17522430419922,
|
| 58 |
+
"learning_rate": 4.970760233918128e-05,
|
| 59 |
+
"loss": 0.8728,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.8300653594771243,
|
| 64 |
+
"grad_norm": 29.73708724975586,
|
| 65 |
+
"learning_rate": 4.8732943469785574e-05,
|
| 66 |
+
"loss": 0.8877,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.9869281045751634,
|
| 71 |
+
"eval_accuracy": 0.6691176470588235,
|
| 72 |
+
"eval_loss": 1.0726829767227173,
|
| 73 |
+
"eval_runtime": 30.3082,
|
| 74 |
+
"eval_samples_per_second": 4.487,
|
| 75 |
+
"eval_steps_per_second": 0.561,
|
| 76 |
+
"step": 76
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0915032679738563,
|
| 80 |
+
"grad_norm": 10.038992881774902,
|
| 81 |
+
"learning_rate": 4.7758284600389865e-05,
|
| 82 |
+
"loss": 0.8215,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.3529411764705883,
|
| 87 |
+
"grad_norm": 16.936914443969727,
|
| 88 |
+
"learning_rate": 4.678362573099415e-05,
|
| 89 |
+
"loss": 0.7058,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.6143790849673203,
|
| 94 |
+
"grad_norm": 16.51180648803711,
|
| 95 |
+
"learning_rate": 4.580896686159844e-05,
|
| 96 |
+
"loss": 0.668,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.8758169934640523,
|
| 101 |
+
"grad_norm": 14.309722900390625,
|
| 102 |
+
"learning_rate": 4.483430799220273e-05,
|
| 103 |
+
"loss": 0.603,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.980392156862745,
|
| 108 |
+
"eval_accuracy": 0.7573529411764706,
|
| 109 |
+
"eval_loss": 0.6806743144989014,
|
| 110 |
+
"eval_runtime": 30.2645,
|
| 111 |
+
"eval_samples_per_second": 4.494,
|
| 112 |
+
"eval_steps_per_second": 0.562,
|
| 113 |
+
"step": 114
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.1372549019607843,
|
| 117 |
+
"grad_norm": 11.795894622802734,
|
| 118 |
+
"learning_rate": 4.3859649122807014e-05,
|
| 119 |
+
"loss": 0.569,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.3986928104575163,
|
| 124 |
+
"grad_norm": 13.41748046875,
|
| 125 |
+
"learning_rate": 4.2884990253411305e-05,
|
| 126 |
+
"loss": 0.4949,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.6601307189542482,
|
| 131 |
+
"grad_norm": 18.057456970214844,
|
| 132 |
+
"learning_rate": 4.1910331384015596e-05,
|
| 133 |
+
"loss": 0.4527,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.9215686274509802,
|
| 138 |
+
"grad_norm": 9.27607536315918,
|
| 139 |
+
"learning_rate": 4.093567251461988e-05,
|
| 140 |
+
"loss": 0.465,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 4.0,
|
| 145 |
+
"eval_accuracy": 0.7573529411764706,
|
| 146 |
+
"eval_loss": 0.6485435366630554,
|
| 147 |
+
"eval_runtime": 30.9055,
|
| 148 |
+
"eval_samples_per_second": 4.401,
|
| 149 |
+
"eval_steps_per_second": 0.55,
|
| 150 |
+
"step": 153
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.183006535947713,
|
| 154 |
+
"grad_norm": 15.063582420349121,
|
| 155 |
+
"learning_rate": 3.996101364522417e-05,
|
| 156 |
+
"loss": 0.4146,
|
| 157 |
+
"step": 160
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.444444444444445,
|
| 161 |
+
"grad_norm": 9.970335960388184,
|
| 162 |
+
"learning_rate": 3.898635477582846e-05,
|
| 163 |
+
"loss": 0.3896,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.705882352941177,
|
| 168 |
+
"grad_norm": 12.845563888549805,
|
| 169 |
+
"learning_rate": 3.8011695906432746e-05,
|
| 170 |
+
"loss": 0.4451,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.967320261437909,
|
| 175 |
+
"grad_norm": 12.283594131469727,
|
| 176 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 177 |
+
"loss": 0.432,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.993464052287582,
|
| 182 |
+
"eval_accuracy": 0.8014705882352942,
|
| 183 |
+
"eval_loss": 0.5023567080497742,
|
| 184 |
+
"eval_runtime": 30.7544,
|
| 185 |
+
"eval_samples_per_second": 4.422,
|
| 186 |
+
"eval_steps_per_second": 0.553,
|
| 187 |
+
"step": 191
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.228758169934641,
|
| 191 |
+
"grad_norm": 10.823830604553223,
|
| 192 |
+
"learning_rate": 3.606237816764133e-05,
|
| 193 |
+
"loss": 0.3117,
|
| 194 |
+
"step": 200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.490196078431373,
|
| 198 |
+
"grad_norm": 10.013704299926758,
|
| 199 |
+
"learning_rate": 3.508771929824561e-05,
|
| 200 |
+
"loss": 0.3033,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.751633986928105,
|
| 205 |
+
"grad_norm": 4.759156227111816,
|
| 206 |
+
"learning_rate": 3.41130604288499e-05,
|
| 207 |
+
"loss": 0.2957,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.9869281045751634,
|
| 212 |
+
"eval_accuracy": 0.8161764705882353,
|
| 213 |
+
"eval_loss": 0.44854581356048584,
|
| 214 |
+
"eval_runtime": 29.9105,
|
| 215 |
+
"eval_samples_per_second": 4.547,
|
| 216 |
+
"eval_steps_per_second": 0.568,
|
| 217 |
+
"step": 229
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.0130718954248366,
|
| 221 |
+
"grad_norm": 13.833900451660156,
|
| 222 |
+
"learning_rate": 3.313840155945419e-05,
|
| 223 |
+
"loss": 0.3418,
|
| 224 |
+
"step": 230
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.2745098039215685,
|
| 228 |
+
"grad_norm": 10.218768119812012,
|
| 229 |
+
"learning_rate": 3.216374269005848e-05,
|
| 230 |
+
"loss": 0.2513,
|
| 231 |
+
"step": 240
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.5359477124183005,
|
| 235 |
+
"grad_norm": 8.130758285522461,
|
| 236 |
+
"learning_rate": 3.118908382066277e-05,
|
| 237 |
+
"loss": 0.2334,
|
| 238 |
+
"step": 250
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.7973856209150325,
|
| 242 |
+
"grad_norm": 11.395691871643066,
|
| 243 |
+
"learning_rate": 3.0214424951267055e-05,
|
| 244 |
+
"loss": 0.2203,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.980392156862745,
|
| 249 |
+
"eval_accuracy": 0.8529411764705882,
|
| 250 |
+
"eval_loss": 0.38499826192855835,
|
| 251 |
+
"eval_runtime": 30.3064,
|
| 252 |
+
"eval_samples_per_second": 4.488,
|
| 253 |
+
"eval_steps_per_second": 0.561,
|
| 254 |
+
"step": 267
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.0588235294117645,
|
| 258 |
+
"grad_norm": 27.90298080444336,
|
| 259 |
+
"learning_rate": 2.9239766081871346e-05,
|
| 260 |
+
"loss": 0.2809,
|
| 261 |
+
"step": 270
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.3202614379084965,
|
| 265 |
+
"grad_norm": 18.04613494873047,
|
| 266 |
+
"learning_rate": 2.8265107212475634e-05,
|
| 267 |
+
"loss": 0.3022,
|
| 268 |
+
"step": 280
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.5816993464052285,
|
| 272 |
+
"grad_norm": 13.312712669372559,
|
| 273 |
+
"learning_rate": 2.729044834307992e-05,
|
| 274 |
+
"loss": 0.2407,
|
| 275 |
+
"step": 290
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.8431372549019605,
|
| 279 |
+
"grad_norm": 6.36769962310791,
|
| 280 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 281 |
+
"loss": 0.236,
|
| 282 |
+
"step": 300
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 8.0,
|
| 286 |
+
"eval_accuracy": 0.8455882352941176,
|
| 287 |
+
"eval_loss": 0.36281630396842957,
|
| 288 |
+
"eval_runtime": 29.9792,
|
| 289 |
+
"eval_samples_per_second": 4.536,
|
| 290 |
+
"eval_steps_per_second": 0.567,
|
| 291 |
+
"step": 306
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.104575163398692,
|
| 295 |
+
"grad_norm": 8.459989547729492,
|
| 296 |
+
"learning_rate": 2.53411306042885e-05,
|
| 297 |
+
"loss": 0.2048,
|
| 298 |
+
"step": 310
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 8.366013071895425,
|
| 302 |
+
"grad_norm": 10.514033317565918,
|
| 303 |
+
"learning_rate": 2.4366471734892787e-05,
|
| 304 |
+
"loss": 0.2115,
|
| 305 |
+
"step": 320
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.627450980392156,
|
| 309 |
+
"grad_norm": 5.906712055206299,
|
| 310 |
+
"learning_rate": 2.3391812865497074e-05,
|
| 311 |
+
"loss": 0.2652,
|
| 312 |
+
"step": 330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.88888888888889,
|
| 316 |
+
"grad_norm": 10.485462188720703,
|
| 317 |
+
"learning_rate": 2.2417153996101365e-05,
|
| 318 |
+
"loss": 0.1857,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.993464052287582,
|
| 323 |
+
"eval_accuracy": 0.8823529411764706,
|
| 324 |
+
"eval_loss": 0.29296010732650757,
|
| 325 |
+
"eval_runtime": 30.4276,
|
| 326 |
+
"eval_samples_per_second": 4.47,
|
| 327 |
+
"eval_steps_per_second": 0.559,
|
| 328 |
+
"step": 344
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 9.15032679738562,
|
| 332 |
+
"grad_norm": 6.746099948883057,
|
| 333 |
+
"learning_rate": 2.1442495126705653e-05,
|
| 334 |
+
"loss": 0.1834,
|
| 335 |
+
"step": 350
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 9.411764705882353,
|
| 339 |
+
"grad_norm": 13.859869003295898,
|
| 340 |
+
"learning_rate": 2.046783625730994e-05,
|
| 341 |
+
"loss": 0.1163,
|
| 342 |
+
"step": 360
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.673202614379084,
|
| 346 |
+
"grad_norm": 11.039109230041504,
|
| 347 |
+
"learning_rate": 1.949317738791423e-05,
|
| 348 |
+
"loss": 0.1548,
|
| 349 |
+
"step": 370
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.934640522875817,
|
| 353 |
+
"grad_norm": 10.195306777954102,
|
| 354 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 355 |
+
"loss": 0.1907,
|
| 356 |
+
"step": 380
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.986928104575163,
|
| 360 |
+
"eval_accuracy": 0.9338235294117647,
|
| 361 |
+
"eval_loss": 0.21208395063877106,
|
| 362 |
+
"eval_runtime": 32.7787,
|
| 363 |
+
"eval_samples_per_second": 4.149,
|
| 364 |
+
"eval_steps_per_second": 0.519,
|
| 365 |
+
"step": 382
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 10.196078431372548,
|
| 369 |
+
"grad_norm": 16.41983985900879,
|
| 370 |
+
"learning_rate": 1.7543859649122806e-05,
|
| 371 |
+
"loss": 0.1645,
|
| 372 |
+
"step": 390
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 10.457516339869281,
|
| 376 |
+
"grad_norm": 13.68142032623291,
|
| 377 |
+
"learning_rate": 1.6569200779727097e-05,
|
| 378 |
+
"loss": 0.166,
|
| 379 |
+
"step": 400
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 10.718954248366012,
|
| 383 |
+
"grad_norm": 9.597548484802246,
|
| 384 |
+
"learning_rate": 1.5594541910331384e-05,
|
| 385 |
+
"loss": 0.1736,
|
| 386 |
+
"step": 410
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 10.980392156862745,
|
| 390 |
+
"grad_norm": 4.147403240203857,
|
| 391 |
+
"learning_rate": 1.4619883040935673e-05,
|
| 392 |
+
"loss": 0.1546,
|
| 393 |
+
"step": 420
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 10.980392156862745,
|
| 397 |
+
"eval_accuracy": 0.9264705882352942,
|
| 398 |
+
"eval_loss": 0.2242465317249298,
|
| 399 |
+
"eval_runtime": 31.8717,
|
| 400 |
+
"eval_samples_per_second": 4.267,
|
| 401 |
+
"eval_steps_per_second": 0.533,
|
| 402 |
+
"step": 420
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 11.241830065359476,
|
| 406 |
+
"grad_norm": 8.938773155212402,
|
| 407 |
+
"learning_rate": 1.364522417153996e-05,
|
| 408 |
+
"loss": 0.1202,
|
| 409 |
+
"step": 430
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 11.50326797385621,
|
| 413 |
+
"grad_norm": 5.6534199714660645,
|
| 414 |
+
"learning_rate": 1.267056530214425e-05,
|
| 415 |
+
"loss": 0.1263,
|
| 416 |
+
"step": 440
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 11.764705882352942,
|
| 420 |
+
"grad_norm": 11.322346687316895,
|
| 421 |
+
"learning_rate": 1.1695906432748537e-05,
|
| 422 |
+
"loss": 0.1375,
|
| 423 |
+
"step": 450
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 12.0,
|
| 427 |
+
"eval_accuracy": 0.9191176470588235,
|
| 428 |
+
"eval_loss": 0.19175946712493896,
|
| 429 |
+
"eval_runtime": 30.8919,
|
| 430 |
+
"eval_samples_per_second": 4.402,
|
| 431 |
+
"eval_steps_per_second": 0.55,
|
| 432 |
+
"step": 459
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 12.026143790849673,
|
| 436 |
+
"grad_norm": 3.5839948654174805,
|
| 437 |
+
"learning_rate": 1.0721247563352826e-05,
|
| 438 |
+
"loss": 0.0921,
|
| 439 |
+
"step": 460
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 12.287581699346406,
|
| 443 |
+
"grad_norm": 9.058156967163086,
|
| 444 |
+
"learning_rate": 9.746588693957115e-06,
|
| 445 |
+
"loss": 0.1056,
|
| 446 |
+
"step": 470
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 12.549019607843137,
|
| 450 |
+
"grad_norm": 18.891481399536133,
|
| 451 |
+
"learning_rate": 8.771929824561403e-06,
|
| 452 |
+
"loss": 0.1095,
|
| 453 |
+
"step": 480
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 12.81045751633987,
|
| 457 |
+
"grad_norm": 12.342869758605957,
|
| 458 |
+
"learning_rate": 7.797270955165692e-06,
|
| 459 |
+
"loss": 0.1237,
|
| 460 |
+
"step": 490
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 12.993464052287582,
|
| 464 |
+
"eval_accuracy": 0.9338235294117647,
|
| 465 |
+
"eval_loss": 0.18091708421707153,
|
| 466 |
+
"eval_runtime": 31.5611,
|
| 467 |
+
"eval_samples_per_second": 4.309,
|
| 468 |
+
"eval_steps_per_second": 0.539,
|
| 469 |
+
"step": 497
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 13.071895424836601,
|
| 473 |
+
"grad_norm": 8.306052207946777,
|
| 474 |
+
"learning_rate": 6.82261208576998e-06,
|
| 475 |
+
"loss": 0.1274,
|
| 476 |
+
"step": 500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 13.333333333333334,
|
| 480 |
+
"grad_norm": 9.888544082641602,
|
| 481 |
+
"learning_rate": 5.8479532163742686e-06,
|
| 482 |
+
"loss": 0.1125,
|
| 483 |
+
"step": 510
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 13.594771241830065,
|
| 487 |
+
"grad_norm": 12.163204193115234,
|
| 488 |
+
"learning_rate": 4.873294346978558e-06,
|
| 489 |
+
"loss": 0.0847,
|
| 490 |
+
"step": 520
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 13.856209150326798,
|
| 494 |
+
"grad_norm": 18.569528579711914,
|
| 495 |
+
"learning_rate": 3.898635477582846e-06,
|
| 496 |
+
"loss": 0.1637,
|
| 497 |
+
"step": 530
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 13.986928104575163,
|
| 501 |
+
"eval_accuracy": 0.9338235294117647,
|
| 502 |
+
"eval_loss": 0.17736296355724335,
|
| 503 |
+
"eval_runtime": 30.6036,
|
| 504 |
+
"eval_samples_per_second": 4.444,
|
| 505 |
+
"eval_steps_per_second": 0.555,
|
| 506 |
+
"step": 535
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 14.117647058823529,
|
| 510 |
+
"grad_norm": 9.332296371459961,
|
| 511 |
+
"learning_rate": 2.9239766081871343e-06,
|
| 512 |
+
"loss": 0.1376,
|
| 513 |
+
"step": 540
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 14.379084967320262,
|
| 517 |
+
"grad_norm": 8.473248481750488,
|
| 518 |
+
"learning_rate": 1.949317738791423e-06,
|
| 519 |
+
"loss": 0.1184,
|
| 520 |
+
"step": 550
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 14.640522875816993,
|
| 524 |
+
"grad_norm": 8.763771057128906,
|
| 525 |
+
"learning_rate": 9.746588693957115e-07,
|
| 526 |
+
"loss": 0.1036,
|
| 527 |
+
"step": 560
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 14.901960784313726,
|
| 531 |
+
"grad_norm": 8.387015342712402,
|
| 532 |
+
"learning_rate": 0.0,
|
| 533 |
+
"loss": 0.0803,
|
| 534 |
+
"step": 570
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 14.901960784313726,
|
| 538 |
+
"eval_accuracy": 0.9338235294117647,
|
| 539 |
+
"eval_loss": 0.18818040192127228,
|
| 540 |
+
"eval_runtime": 31.0653,
|
| 541 |
+
"eval_samples_per_second": 4.378,
|
| 542 |
+
"eval_steps_per_second": 0.547,
|
| 543 |
+
"step": 570
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 14.901960784313726,
|
| 547 |
+
"step": 570,
|
| 548 |
+
"total_flos": 9.102711432019968e+16,
|
| 549 |
+
"train_loss": 0.39481510582723117,
|
| 550 |
+
"train_runtime": 4386.4067,
|
| 551 |
+
"train_samples_per_second": 4.186,
|
| 552 |
+
"train_steps_per_second": 0.13
|
| 553 |
+
}
|
| 554 |
+
],
|
| 555 |
+
"logging_steps": 10,
|
| 556 |
+
"max_steps": 570,
|
| 557 |
+
"num_input_tokens_seen": 0,
|
| 558 |
+
"num_train_epochs": 15,
|
| 559 |
+
"save_steps": 500,
|
| 560 |
+
"stateful_callbacks": {
|
| 561 |
+
"TrainerControl": {
|
| 562 |
+
"args": {
|
| 563 |
+
"should_epoch_stop": false,
|
| 564 |
+
"should_evaluate": false,
|
| 565 |
+
"should_log": false,
|
| 566 |
+
"should_save": true,
|
| 567 |
+
"should_training_stop": true
|
| 568 |
+
},
|
| 569 |
+
"attributes": {}
|
| 570 |
+
}
|
| 571 |
+
},
|
| 572 |
+
"total_flos": 9.102711432019968e+16,
|
| 573 |
+
"train_batch_size": 8,
|
| 574 |
+
"trial_name": null,
|
| 575 |
+
"trial_params": null
|
| 576 |
+
}
|