🍻 cheers
Browse files- README.md +2 -1
- all_results.json +14 -0
- eval_results.json +9 -0
- runs/Aug12_00-26-12_ba2a0abc8a06/events.out.tfevents.1754959725.ba2a0abc8a06.475.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +645 -0
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -17,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
# vit-ena24-clase
|
| 19 |
|
| 20 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.3132
|
| 23 |
- Accuracy: 0.9321
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224-in21k
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# vit-ena24-clase
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the ena24_MD dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
- Loss: 0.3132
|
| 24 |
- Accuracy: 0.9321
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.9320610687022901,
|
| 4 |
+
"eval_f1": 0.878929195485808,
|
| 5 |
+
"eval_loss": 0.31322020292282104,
|
| 6 |
+
"eval_runtime": 60.8031,
|
| 7 |
+
"eval_samples_per_second": 21.545,
|
| 8 |
+
"eval_steps_per_second": 2.697,
|
| 9 |
+
"total_flos": 4.7604111703864934e+17,
|
| 10 |
+
"train_loss": 0.9329218550895652,
|
| 11 |
+
"train_runtime": 1211.4391,
|
| 12 |
+
"train_samples_per_second": 5.07,
|
| 13 |
+
"train_steps_per_second": 0.634
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.9320610687022901,
|
| 4 |
+
"eval_f1": 0.878929195485808,
|
| 5 |
+
"eval_loss": 0.31322020292282104,
|
| 6 |
+
"eval_runtime": 60.8031,
|
| 7 |
+
"eval_samples_per_second": 21.545,
|
| 8 |
+
"eval_steps_per_second": 2.697
|
| 9 |
+
}
|
runs/Aug12_00-26-12_ba2a0abc8a06/events.out.tfevents.1754959725.ba2a0abc8a06.475.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4db8c93b7bddd321da7b03f5cf1fbb1b05ff6c65e82d5a037fd0cccc887fae3b
|
| 3 |
+
size 457
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 4.7604111703864934e+17,
|
| 4 |
+
"train_loss": 0.9329218550895652,
|
| 5 |
+
"train_runtime": 1211.4391,
|
| 6 |
+
"train_samples_per_second": 5.07,
|
| 7 |
+
"train_steps_per_second": 0.634
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,645 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 700,
|
| 3 |
+
"best_metric": 0.31322020292282104,
|
| 4 |
+
"best_model_checkpoint": "./vit-ena24-clase/checkpoint-700",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 768,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.013020833333333334,
|
| 14 |
+
"grad_norm": 2.9136159420013428,
|
| 15 |
+
"learning_rate": 0.00019765625,
|
| 16 |
+
"loss": 3.0193,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.026041666666666668,
|
| 21 |
+
"grad_norm": 2.3808610439300537,
|
| 22 |
+
"learning_rate": 0.00019505208333333335,
|
| 23 |
+
"loss": 2.7993,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0390625,
|
| 28 |
+
"grad_norm": 2.8466873168945312,
|
| 29 |
+
"learning_rate": 0.0001924479166666667,
|
| 30 |
+
"loss": 2.7477,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.052083333333333336,
|
| 35 |
+
"grad_norm": 2.4645440578460693,
|
| 36 |
+
"learning_rate": 0.00018984375000000002,
|
| 37 |
+
"loss": 2.6105,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06510416666666667,
|
| 42 |
+
"grad_norm": 2.5182878971099854,
|
| 43 |
+
"learning_rate": 0.00018723958333333334,
|
| 44 |
+
"loss": 2.3659,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.078125,
|
| 49 |
+
"grad_norm": 2.2973132133483887,
|
| 50 |
+
"learning_rate": 0.0001846354166666667,
|
| 51 |
+
"loss": 2.1802,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09114583333333333,
|
| 56 |
+
"grad_norm": 2.9924745559692383,
|
| 57 |
+
"learning_rate": 0.00018203125,
|
| 58 |
+
"loss": 2.1703,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.10416666666666667,
|
| 63 |
+
"grad_norm": 2.4488980770111084,
|
| 64 |
+
"learning_rate": 0.00017942708333333333,
|
| 65 |
+
"loss": 1.8086,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.1171875,
|
| 70 |
+
"grad_norm": 2.935206890106201,
|
| 71 |
+
"learning_rate": 0.00017682291666666668,
|
| 72 |
+
"loss": 1.8366,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.13020833333333334,
|
| 77 |
+
"grad_norm": 2.841130018234253,
|
| 78 |
+
"learning_rate": 0.00017421875,
|
| 79 |
+
"loss": 1.9153,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.13020833333333334,
|
| 84 |
+
"eval_accuracy": 0.5725190839694656,
|
| 85 |
+
"eval_f1": 0.4645958852584343,
|
| 86 |
+
"eval_loss": 1.7646938562393188,
|
| 87 |
+
"eval_runtime": 80.3983,
|
| 88 |
+
"eval_samples_per_second": 16.294,
|
| 89 |
+
"eval_steps_per_second": 2.04,
|
| 90 |
+
"step": 100
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.14322916666666666,
|
| 94 |
+
"grad_norm": 1.978049397468567,
|
| 95 |
+
"learning_rate": 0.00017161458333333332,
|
| 96 |
+
"loss": 1.7021,
|
| 97 |
+
"step": 110
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.15625,
|
| 101 |
+
"grad_norm": 3.838554859161377,
|
| 102 |
+
"learning_rate": 0.00016901041666666667,
|
| 103 |
+
"loss": 1.606,
|
| 104 |
+
"step": 120
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.16927083333333334,
|
| 108 |
+
"grad_norm": 3.2451305389404297,
|
| 109 |
+
"learning_rate": 0.00016640625,
|
| 110 |
+
"loss": 1.5631,
|
| 111 |
+
"step": 130
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.18229166666666666,
|
| 115 |
+
"grad_norm": 2.3414347171783447,
|
| 116 |
+
"learning_rate": 0.00016380208333333335,
|
| 117 |
+
"loss": 1.4515,
|
| 118 |
+
"step": 140
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.1953125,
|
| 122 |
+
"grad_norm": 3.25227952003479,
|
| 123 |
+
"learning_rate": 0.0001611979166666667,
|
| 124 |
+
"loss": 1.3315,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.20833333333333334,
|
| 129 |
+
"grad_norm": 4.009838104248047,
|
| 130 |
+
"learning_rate": 0.00015859375000000002,
|
| 131 |
+
"loss": 1.3692,
|
| 132 |
+
"step": 160
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.22135416666666666,
|
| 136 |
+
"grad_norm": 3.831077814102173,
|
| 137 |
+
"learning_rate": 0.00015598958333333334,
|
| 138 |
+
"loss": 1.1402,
|
| 139 |
+
"step": 170
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.234375,
|
| 143 |
+
"grad_norm": 4.556130886077881,
|
| 144 |
+
"learning_rate": 0.0001533854166666667,
|
| 145 |
+
"loss": 1.4215,
|
| 146 |
+
"step": 180
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.24739583333333334,
|
| 150 |
+
"grad_norm": 1.8576217889785767,
|
| 151 |
+
"learning_rate": 0.00015078125,
|
| 152 |
+
"loss": 1.2708,
|
| 153 |
+
"step": 190
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.2604166666666667,
|
| 157 |
+
"grad_norm": 2.6255905628204346,
|
| 158 |
+
"learning_rate": 0.00014817708333333333,
|
| 159 |
+
"loss": 1.2463,
|
| 160 |
+
"step": 200
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.2604166666666667,
|
| 164 |
+
"eval_accuracy": 0.7641221374045801,
|
| 165 |
+
"eval_f1": 0.6932650150404468,
|
| 166 |
+
"eval_loss": 1.1007627248764038,
|
| 167 |
+
"eval_runtime": 80.2425,
|
| 168 |
+
"eval_samples_per_second": 16.326,
|
| 169 |
+
"eval_steps_per_second": 2.044,
|
| 170 |
+
"step": 200
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2734375,
|
| 174 |
+
"grad_norm": 1.541453242301941,
|
| 175 |
+
"learning_rate": 0.00014557291666666668,
|
| 176 |
+
"loss": 0.9071,
|
| 177 |
+
"step": 210
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2864583333333333,
|
| 181 |
+
"grad_norm": 2.118795871734619,
|
| 182 |
+
"learning_rate": 0.00014296875,
|
| 183 |
+
"loss": 1.227,
|
| 184 |
+
"step": 220
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2994791666666667,
|
| 188 |
+
"grad_norm": 4.64823579788208,
|
| 189 |
+
"learning_rate": 0.00014036458333333332,
|
| 190 |
+
"loss": 1.1662,
|
| 191 |
+
"step": 230
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3125,
|
| 195 |
+
"grad_norm": 4.002511024475098,
|
| 196 |
+
"learning_rate": 0.00013776041666666667,
|
| 197 |
+
"loss": 1.1199,
|
| 198 |
+
"step": 240
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3255208333333333,
|
| 202 |
+
"grad_norm": 2.231586456298828,
|
| 203 |
+
"learning_rate": 0.00013515625,
|
| 204 |
+
"loss": 0.9278,
|
| 205 |
+
"step": 250
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3385416666666667,
|
| 209 |
+
"grad_norm": 4.344457626342773,
|
| 210 |
+
"learning_rate": 0.00013255208333333334,
|
| 211 |
+
"loss": 1.0149,
|
| 212 |
+
"step": 260
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3515625,
|
| 216 |
+
"grad_norm": 3.3675150871276855,
|
| 217 |
+
"learning_rate": 0.0001299479166666667,
|
| 218 |
+
"loss": 0.8122,
|
| 219 |
+
"step": 270
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3645833333333333,
|
| 223 |
+
"grad_norm": 4.569275379180908,
|
| 224 |
+
"learning_rate": 0.00012734375000000002,
|
| 225 |
+
"loss": 1.1547,
|
| 226 |
+
"step": 280
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3776041666666667,
|
| 230 |
+
"grad_norm": 7.2577056884765625,
|
| 231 |
+
"learning_rate": 0.00012473958333333334,
|
| 232 |
+
"loss": 0.8584,
|
| 233 |
+
"step": 290
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.390625,
|
| 237 |
+
"grad_norm": 1.9357669353485107,
|
| 238 |
+
"learning_rate": 0.0001221354166666667,
|
| 239 |
+
"loss": 0.884,
|
| 240 |
+
"step": 300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.390625,
|
| 244 |
+
"eval_accuracy": 0.783206106870229,
|
| 245 |
+
"eval_f1": 0.7112739549768266,
|
| 246 |
+
"eval_loss": 0.9143364429473877,
|
| 247 |
+
"eval_runtime": 76.8791,
|
| 248 |
+
"eval_samples_per_second": 17.04,
|
| 249 |
+
"eval_steps_per_second": 2.133,
|
| 250 |
+
"step": 300
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.4036458333333333,
|
| 254 |
+
"grad_norm": 1.330380916595459,
|
| 255 |
+
"learning_rate": 0.00011953125000000001,
|
| 256 |
+
"loss": 0.9432,
|
| 257 |
+
"step": 310
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.4166666666666667,
|
| 261 |
+
"grad_norm": 2.399350643157959,
|
| 262 |
+
"learning_rate": 0.00011692708333333333,
|
| 263 |
+
"loss": 0.8459,
|
| 264 |
+
"step": 320
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.4296875,
|
| 268 |
+
"grad_norm": 2.1869707107543945,
|
| 269 |
+
"learning_rate": 0.00011432291666666668,
|
| 270 |
+
"loss": 0.7181,
|
| 271 |
+
"step": 330
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.4427083333333333,
|
| 275 |
+
"grad_norm": 1.096358060836792,
|
| 276 |
+
"learning_rate": 0.00011171875,
|
| 277 |
+
"loss": 0.5569,
|
| 278 |
+
"step": 340
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.4557291666666667,
|
| 282 |
+
"grad_norm": 2.5247671604156494,
|
| 283 |
+
"learning_rate": 0.00010911458333333334,
|
| 284 |
+
"loss": 0.5736,
|
| 285 |
+
"step": 350
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.46875,
|
| 289 |
+
"grad_norm": 2.0583877563476562,
|
| 290 |
+
"learning_rate": 0.00010651041666666667,
|
| 291 |
+
"loss": 0.5871,
|
| 292 |
+
"step": 360
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.4817708333333333,
|
| 296 |
+
"grad_norm": 1.8924134969711304,
|
| 297 |
+
"learning_rate": 0.00010390625000000001,
|
| 298 |
+
"loss": 0.6245,
|
| 299 |
+
"step": 370
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.4947916666666667,
|
| 303 |
+
"grad_norm": 2.7611899375915527,
|
| 304 |
+
"learning_rate": 0.00010130208333333333,
|
| 305 |
+
"loss": 1.0024,
|
| 306 |
+
"step": 380
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.5078125,
|
| 310 |
+
"grad_norm": 2.3983047008514404,
|
| 311 |
+
"learning_rate": 9.869791666666667e-05,
|
| 312 |
+
"loss": 0.6246,
|
| 313 |
+
"step": 390
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.5208333333333334,
|
| 317 |
+
"grad_norm": 4.21718168258667,
|
| 318 |
+
"learning_rate": 9.609375e-05,
|
| 319 |
+
"loss": 0.6852,
|
| 320 |
+
"step": 400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.5208333333333334,
|
| 324 |
+
"eval_accuracy": 0.8648854961832061,
|
| 325 |
+
"eval_f1": 0.8027433119068881,
|
| 326 |
+
"eval_loss": 0.6161403059959412,
|
| 327 |
+
"eval_runtime": 57.8706,
|
| 328 |
+
"eval_samples_per_second": 22.637,
|
| 329 |
+
"eval_steps_per_second": 2.834,
|
| 330 |
+
"step": 400
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.5338541666666666,
|
| 334 |
+
"grad_norm": 0.9279044270515442,
|
| 335 |
+
"learning_rate": 9.348958333333334e-05,
|
| 336 |
+
"loss": 0.6579,
|
| 337 |
+
"step": 410
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.546875,
|
| 341 |
+
"grad_norm": 1.944657802581787,
|
| 342 |
+
"learning_rate": 9.088541666666667e-05,
|
| 343 |
+
"loss": 0.6463,
|
| 344 |
+
"step": 420
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.5598958333333334,
|
| 348 |
+
"grad_norm": 1.5608443021774292,
|
| 349 |
+
"learning_rate": 8.828125000000001e-05,
|
| 350 |
+
"loss": 0.511,
|
| 351 |
+
"step": 430
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.5729166666666666,
|
| 355 |
+
"grad_norm": 2.153334617614746,
|
| 356 |
+
"learning_rate": 8.567708333333334e-05,
|
| 357 |
+
"loss": 0.4522,
|
| 358 |
+
"step": 440
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.5859375,
|
| 362 |
+
"grad_norm": 2.5961196422576904,
|
| 363 |
+
"learning_rate": 8.307291666666667e-05,
|
| 364 |
+
"loss": 0.6923,
|
| 365 |
+
"step": 450
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.5989583333333334,
|
| 369 |
+
"grad_norm": 8.20881462097168,
|
| 370 |
+
"learning_rate": 8.046875e-05,
|
| 371 |
+
"loss": 0.7458,
|
| 372 |
+
"step": 460
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.6119791666666666,
|
| 376 |
+
"grad_norm": 3.0623526573181152,
|
| 377 |
+
"learning_rate": 7.786458333333334e-05,
|
| 378 |
+
"loss": 0.6869,
|
| 379 |
+
"step": 470
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.625,
|
| 383 |
+
"grad_norm": 3.7377140522003174,
|
| 384 |
+
"learning_rate": 7.526041666666667e-05,
|
| 385 |
+
"loss": 0.5997,
|
| 386 |
+
"step": 480
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.6380208333333334,
|
| 390 |
+
"grad_norm": 1.7926647663116455,
|
| 391 |
+
"learning_rate": 7.265625000000001e-05,
|
| 392 |
+
"loss": 0.339,
|
| 393 |
+
"step": 490
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.6510416666666666,
|
| 397 |
+
"grad_norm": 0.831264317035675,
|
| 398 |
+
"learning_rate": 7.005208333333334e-05,
|
| 399 |
+
"loss": 0.5318,
|
| 400 |
+
"step": 500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.6510416666666666,
|
| 404 |
+
"eval_accuracy": 0.8946564885496183,
|
| 405 |
+
"eval_f1": 0.8375824690045511,
|
| 406 |
+
"eval_loss": 0.46913737058639526,
|
| 407 |
+
"eval_runtime": 58.7034,
|
| 408 |
+
"eval_samples_per_second": 22.316,
|
| 409 |
+
"eval_steps_per_second": 2.794,
|
| 410 |
+
"step": 500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.6640625,
|
| 414 |
+
"grad_norm": 2.2791216373443604,
|
| 415 |
+
"learning_rate": 6.744791666666667e-05,
|
| 416 |
+
"loss": 0.5303,
|
| 417 |
+
"step": 510
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.6770833333333334,
|
| 421 |
+
"grad_norm": 8.944572448730469,
|
| 422 |
+
"learning_rate": 6.484375e-05,
|
| 423 |
+
"loss": 0.7781,
|
| 424 |
+
"step": 520
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.6901041666666666,
|
| 428 |
+
"grad_norm": 2.116546392440796,
|
| 429 |
+
"learning_rate": 6.223958333333334e-05,
|
| 430 |
+
"loss": 0.6038,
|
| 431 |
+
"step": 530
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.703125,
|
| 435 |
+
"grad_norm": 2.7448010444641113,
|
| 436 |
+
"learning_rate": 5.9635416666666665e-05,
|
| 437 |
+
"loss": 0.3419,
|
| 438 |
+
"step": 540
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.7161458333333334,
|
| 442 |
+
"grad_norm": 1.1109596490859985,
|
| 443 |
+
"learning_rate": 5.703125e-05,
|
| 444 |
+
"loss": 0.5202,
|
| 445 |
+
"step": 550
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.7291666666666666,
|
| 449 |
+
"grad_norm": 0.46282243728637695,
|
| 450 |
+
"learning_rate": 5.442708333333334e-05,
|
| 451 |
+
"loss": 0.477,
|
| 452 |
+
"step": 560
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.7421875,
|
| 456 |
+
"grad_norm": 2.3069491386413574,
|
| 457 |
+
"learning_rate": 5.1822916666666665e-05,
|
| 458 |
+
"loss": 0.4041,
|
| 459 |
+
"step": 570
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.7552083333333334,
|
| 463 |
+
"grad_norm": 7.2379655838012695,
|
| 464 |
+
"learning_rate": 4.921875e-05,
|
| 465 |
+
"loss": 0.5449,
|
| 466 |
+
"step": 580
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.7682291666666666,
|
| 470 |
+
"grad_norm": 1.788082242012024,
|
| 471 |
+
"learning_rate": 4.6614583333333336e-05,
|
| 472 |
+
"loss": 0.2639,
|
| 473 |
+
"step": 590
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.78125,
|
| 477 |
+
"grad_norm": 2.0945539474487305,
|
| 478 |
+
"learning_rate": 4.401041666666667e-05,
|
| 479 |
+
"loss": 0.5544,
|
| 480 |
+
"step": 600
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.78125,
|
| 484 |
+
"eval_accuracy": 0.9152671755725191,
|
| 485 |
+
"eval_f1": 0.898366041912979,
|
| 486 |
+
"eval_loss": 0.3783359229564667,
|
| 487 |
+
"eval_runtime": 58.3514,
|
| 488 |
+
"eval_samples_per_second": 22.45,
|
| 489 |
+
"eval_steps_per_second": 2.811,
|
| 490 |
+
"step": 600
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.7942708333333334,
|
| 494 |
+
"grad_norm": 0.33394262194633484,
|
| 495 |
+
"learning_rate": 4.140625e-05,
|
| 496 |
+
"loss": 0.3429,
|
| 497 |
+
"step": 610
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.8072916666666666,
|
| 501 |
+
"grad_norm": 1.3352487087249756,
|
| 502 |
+
"learning_rate": 3.8802083333333336e-05,
|
| 503 |
+
"loss": 0.3842,
|
| 504 |
+
"step": 620
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.8203125,
|
| 508 |
+
"grad_norm": 1.9343860149383545,
|
| 509 |
+
"learning_rate": 3.619791666666667e-05,
|
| 510 |
+
"loss": 0.3926,
|
| 511 |
+
"step": 630
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.8333333333333334,
|
| 515 |
+
"grad_norm": 2.879375696182251,
|
| 516 |
+
"learning_rate": 3.359375e-05,
|
| 517 |
+
"loss": 0.3239,
|
| 518 |
+
"step": 640
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.8463541666666666,
|
| 522 |
+
"grad_norm": 1.077412724494934,
|
| 523 |
+
"learning_rate": 3.0989583333333336e-05,
|
| 524 |
+
"loss": 0.2097,
|
| 525 |
+
"step": 650
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.859375,
|
| 529 |
+
"grad_norm": 1.6046684980392456,
|
| 530 |
+
"learning_rate": 2.838541666666667e-05,
|
| 531 |
+
"loss": 0.2736,
|
| 532 |
+
"step": 660
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.8723958333333334,
|
| 536 |
+
"grad_norm": 0.9762991070747375,
|
| 537 |
+
"learning_rate": 2.578125e-05,
|
| 538 |
+
"loss": 0.3271,
|
| 539 |
+
"step": 670
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.8854166666666666,
|
| 543 |
+
"grad_norm": 3.991189479827881,
|
| 544 |
+
"learning_rate": 2.3177083333333335e-05,
|
| 545 |
+
"loss": 0.4693,
|
| 546 |
+
"step": 680
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.8984375,
|
| 550 |
+
"grad_norm": 0.9380645155906677,
|
| 551 |
+
"learning_rate": 2.0572916666666668e-05,
|
| 552 |
+
"loss": 0.4921,
|
| 553 |
+
"step": 690
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.9114583333333334,
|
| 557 |
+
"grad_norm": 0.30667611956596375,
|
| 558 |
+
"learning_rate": 1.796875e-05,
|
| 559 |
+
"loss": 0.2321,
|
| 560 |
+
"step": 700
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.9114583333333334,
|
| 564 |
+
"eval_accuracy": 0.9320610687022901,
|
| 565 |
+
"eval_f1": 0.878929195485808,
|
| 566 |
+
"eval_loss": 0.31322020292282104,
|
| 567 |
+
"eval_runtime": 59.3146,
|
| 568 |
+
"eval_samples_per_second": 22.086,
|
| 569 |
+
"eval_steps_per_second": 2.765,
|
| 570 |
+
"step": 700
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.9244791666666666,
|
| 574 |
+
"grad_norm": 4.246342182159424,
|
| 575 |
+
"learning_rate": 1.5364583333333335e-05,
|
| 576 |
+
"loss": 0.4772,
|
| 577 |
+
"step": 710
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.9375,
|
| 581 |
+
"grad_norm": 5.404312610626221,
|
| 582 |
+
"learning_rate": 1.2760416666666666e-05,
|
| 583 |
+
"loss": 0.4208,
|
| 584 |
+
"step": 720
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.9505208333333334,
|
| 588 |
+
"grad_norm": 4.308945655822754,
|
| 589 |
+
"learning_rate": 1.0156250000000001e-05,
|
| 590 |
+
"loss": 0.3819,
|
| 591 |
+
"step": 730
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.9635416666666666,
|
| 595 |
+
"grad_norm": 0.27945882081985474,
|
| 596 |
+
"learning_rate": 7.552083333333333e-06,
|
| 597 |
+
"loss": 0.3349,
|
| 598 |
+
"step": 740
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.9765625,
|
| 602 |
+
"grad_norm": 2.4253346920013428,
|
| 603 |
+
"learning_rate": 4.947916666666666e-06,
|
| 604 |
+
"loss": 0.3802,
|
| 605 |
+
"step": 750
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.9895833333333334,
|
| 609 |
+
"grad_norm": 0.25109484791755676,
|
| 610 |
+
"learning_rate": 2.3437500000000002e-06,
|
| 611 |
+
"loss": 0.2813,
|
| 612 |
+
"step": 760
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.0,
|
| 616 |
+
"step": 768,
|
| 617 |
+
"total_flos": 4.7604111703864934e+17,
|
| 618 |
+
"train_loss": 0.9329218550895652,
|
| 619 |
+
"train_runtime": 1211.4391,
|
| 620 |
+
"train_samples_per_second": 5.07,
|
| 621 |
+
"train_steps_per_second": 0.634
|
| 622 |
+
}
|
| 623 |
+
],
|
| 624 |
+
"logging_steps": 10,
|
| 625 |
+
"max_steps": 768,
|
| 626 |
+
"num_input_tokens_seen": 0,
|
| 627 |
+
"num_train_epochs": 1,
|
| 628 |
+
"save_steps": 100,
|
| 629 |
+
"stateful_callbacks": {
|
| 630 |
+
"TrainerControl": {
|
| 631 |
+
"args": {
|
| 632 |
+
"should_epoch_stop": false,
|
| 633 |
+
"should_evaluate": false,
|
| 634 |
+
"should_log": false,
|
| 635 |
+
"should_save": true,
|
| 636 |
+
"should_training_stop": true
|
| 637 |
+
},
|
| 638 |
+
"attributes": {}
|
| 639 |
+
}
|
| 640 |
+
},
|
| 641 |
+
"total_flos": 4.7604111703864934e+17,
|
| 642 |
+
"train_batch_size": 8,
|
| 643 |
+
"trial_name": null,
|
| 644 |
+
"trial_params": null
|
| 645 |
+
}
|