End of training
Browse files- README.md +5 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec19_13-02-49_algo-1/events.out.tfevents.1766150078.algo-1.68.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +760 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: Logiroad/Validated_cracks_raw_dataset_359_convnext_model
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# validated_cracks_dataset_relabeled_convnext_finetune_model
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [Logiroad/Validated_cracks_raw_dataset_359_convnext_model](https://huggingface.co/Logiroad/Validated_cracks_raw_dataset_359_convnext_model) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss:
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: Logiroad/Validated_cracks_raw_dataset_359_convnext_model
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# validated_cracks_dataset_relabeled_convnext_finetune_model
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [Logiroad/Validated_cracks_raw_dataset_359_convnext_model](https://huggingface.co/Logiroad/Validated_cracks_raw_dataset_359_convnext_model) on the Logiroad/validated_cracks_dataset_relabeled dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.9997
|
| 24 |
+
- Accuracy: 0.6139
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.6138996138996139,
|
| 4 |
+
"eval_loss": 0.9997061491012573,
|
| 5 |
+
"eval_runtime": 1.754,
|
| 6 |
+
"eval_samples_per_second": 147.664,
|
| 7 |
+
"eval_steps_per_second": 18.814,
|
| 8 |
+
"total_flos": 2.0405460235862016e+18,
|
| 9 |
+
"train_loss": 0.7678916913491708,
|
| 10 |
+
"train_runtime": 680.1267,
|
| 11 |
+
"train_samples_per_second": 37.934,
|
| 12 |
+
"train_steps_per_second": 4.764
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.6138996138996139,
|
| 4 |
+
"eval_loss": 0.9997061491012573,
|
| 5 |
+
"eval_runtime": 1.754,
|
| 6 |
+
"eval_samples_per_second": 147.664,
|
| 7 |
+
"eval_steps_per_second": 18.814
|
| 8 |
+
}
|
runs/Dec19_13-02-49_algo-1/events.out.tfevents.1766150078.algo-1.68.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04acb5b7b7fadebb1bd508e8e8a5a334f026af96d44c51f8caa1f8924f841c6e
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 2.0405460235862016e+18,
|
| 4 |
+
"train_loss": 0.7678916913491708,
|
| 5 |
+
"train_runtime": 680.1267,
|
| 6 |
+
"train_samples_per_second": 37.934,
|
| 7 |
+
"train_steps_per_second": 4.764
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6138996138996139,
|
| 3 |
+
"best_model_checkpoint": "./Validated_cracks_raw_dataset_359_relabeled2_outputs/checkpoint-432",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.46296296296296297,
|
| 13 |
+
"grad_norm": 11.11386489868164,
|
| 14 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 15 |
+
"loss": 0.9464,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9259259259259259,
|
| 20 |
+
"grad_norm": 22.558269500732422,
|
| 21 |
+
"learning_rate": 5.864197530864199e-06,
|
| 22 |
+
"loss": 0.9468,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.6023166023166023,
|
| 28 |
+
"eval_loss": 0.9880222678184509,
|
| 29 |
+
"eval_runtime": 1.7676,
|
| 30 |
+
"eval_samples_per_second": 146.524,
|
| 31 |
+
"eval_steps_per_second": 18.669,
|
| 32 |
+
"step": 108
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.3888888888888888,
|
| 36 |
+
"grad_norm": 23.05500030517578,
|
| 37 |
+
"learning_rate": 8.950617283950618e-06,
|
| 38 |
+
"loss": 0.9061,
|
| 39 |
+
"step": 150
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.8518518518518519,
|
| 43 |
+
"grad_norm": 20.498937606811523,
|
| 44 |
+
"learning_rate": 9.997164110237345e-06,
|
| 45 |
+
"loss": 0.9275,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_accuracy": 0.583011583011583,
|
| 51 |
+
"eval_loss": 1.0344593524932861,
|
| 52 |
+
"eval_runtime": 1.7521,
|
| 53 |
+
"eval_samples_per_second": 147.825,
|
| 54 |
+
"eval_steps_per_second": 18.835,
|
| 55 |
+
"step": 216
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.314814814814815,
|
| 59 |
+
"grad_norm": 10.24534797668457,
|
| 60 |
+
"learning_rate": 9.98206922906478e-06,
|
| 61 |
+
"loss": 0.9077,
|
| 62 |
+
"step": 250
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.7777777777777777,
|
| 66 |
+
"grad_norm": 62.0069580078125,
|
| 67 |
+
"learning_rate": 9.954002016824226e-06,
|
| 68 |
+
"loss": 0.9377,
|
| 69 |
+
"step": 300
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_accuracy": 0.5675675675675675,
|
| 74 |
+
"eval_loss": 1.023147463798523,
|
| 75 |
+
"eval_runtime": 1.7707,
|
| 76 |
+
"eval_samples_per_second": 146.27,
|
| 77 |
+
"eval_steps_per_second": 18.637,
|
| 78 |
+
"step": 324
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.240740740740741,
|
| 82 |
+
"grad_norm": 19.093095779418945,
|
| 83 |
+
"learning_rate": 9.91398066213228e-06,
|
| 84 |
+
"loss": 0.8912,
|
| 85 |
+
"step": 350
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.7037037037037037,
|
| 89 |
+
"grad_norm": 14.229964256286621,
|
| 90 |
+
"learning_rate": 9.860476209695013e-06,
|
| 91 |
+
"loss": 0.8461,
|
| 92 |
+
"step": 400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.0,
|
| 96 |
+
"eval_accuracy": 0.6138996138996139,
|
| 97 |
+
"eval_loss": 0.9997061491012573,
|
| 98 |
+
"eval_runtime": 1.7439,
|
| 99 |
+
"eval_samples_per_second": 148.519,
|
| 100 |
+
"eval_steps_per_second": 18.923,
|
| 101 |
+
"step": 432
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.166666666666667,
|
| 105 |
+
"grad_norm": 25.24408721923828,
|
| 106 |
+
"learning_rate": 9.794316030562418e-06,
|
| 107 |
+
"loss": 0.8663,
|
| 108 |
+
"step": 450
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.62962962962963,
|
| 112 |
+
"grad_norm": 27.978118896484375,
|
| 113 |
+
"learning_rate": 9.715672392864017e-06,
|
| 114 |
+
"loss": 0.8429,
|
| 115 |
+
"step": 500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 5.0,
|
| 119 |
+
"eval_accuracy": 0.5868725868725869,
|
| 120 |
+
"eval_loss": 0.9872773885726929,
|
| 121 |
+
"eval_runtime": 1.7504,
|
| 122 |
+
"eval_samples_per_second": 147.962,
|
| 123 |
+
"eval_steps_per_second": 18.852,
|
| 124 |
+
"step": 540
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.092592592592593,
|
| 128 |
+
"grad_norm": 22.3253173828125,
|
| 129 |
+
"learning_rate": 9.62475006920687e-06,
|
| 130 |
+
"loss": 0.8246,
|
| 131 |
+
"step": 550
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.555555555555555,
|
| 135 |
+
"grad_norm": 29.907615661621094,
|
| 136 |
+
"learning_rate": 9.521785803487888e-06,
|
| 137 |
+
"loss": 0.781,
|
| 138 |
+
"step": 600
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 6.0,
|
| 142 |
+
"eval_accuracy": 0.6061776061776062,
|
| 143 |
+
"eval_loss": 0.9970757365226746,
|
| 144 |
+
"eval_runtime": 1.7424,
|
| 145 |
+
"eval_samples_per_second": 148.643,
|
| 146 |
+
"eval_steps_per_second": 18.939,
|
| 147 |
+
"step": 648
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 6.018518518518518,
|
| 151 |
+
"grad_norm": 23.293292999267578,
|
| 152 |
+
"learning_rate": 9.407047694459149e-06,
|
| 153 |
+
"loss": 0.8409,
|
| 154 |
+
"step": 650
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.481481481481482,
|
| 158 |
+
"grad_norm": 21.52410888671875,
|
| 159 |
+
"learning_rate": 9.280834497651334e-06,
|
| 160 |
+
"loss": 0.8137,
|
| 161 |
+
"step": 700
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.944444444444445,
|
| 165 |
+
"grad_norm": 34.57286071777344,
|
| 166 |
+
"learning_rate": 9.143474847472932e-06,
|
| 167 |
+
"loss": 0.7621,
|
| 168 |
+
"step": 750
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"eval_accuracy": 0.583011583011583,
|
| 173 |
+
"eval_loss": 1.0092536211013794,
|
| 174 |
+
"eval_runtime": 1.7382,
|
| 175 |
+
"eval_samples_per_second": 149.004,
|
| 176 |
+
"eval_steps_per_second": 18.985,
|
| 177 |
+
"step": 756
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.407407407407407,
|
| 181 |
+
"grad_norm": 15.154962539672852,
|
| 182 |
+
"learning_rate": 8.9953264015107e-06,
|
| 183 |
+
"loss": 0.7191,
|
| 184 |
+
"step": 800
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.87037037037037,
|
| 188 |
+
"grad_norm": 40.10356140136719,
|
| 189 |
+
"learning_rate": 8.836774909259442e-06,
|
| 190 |
+
"loss": 0.7649,
|
| 191 |
+
"step": 850
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 8.0,
|
| 195 |
+
"eval_accuracy": 0.5675675675675675,
|
| 196 |
+
"eval_loss": 1.0584689378738403,
|
| 197 |
+
"eval_runtime": 1.7294,
|
| 198 |
+
"eval_samples_per_second": 149.767,
|
| 199 |
+
"eval_steps_per_second": 19.082,
|
| 200 |
+
"step": 864
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.333333333333334,
|
| 204 |
+
"grad_norm": 65.0152587890625,
|
| 205 |
+
"learning_rate": 8.668233207706014e-06,
|
| 206 |
+
"loss": 0.7822,
|
| 207 |
+
"step": 900
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 8.796296296296296,
|
| 211 |
+
"grad_norm": 27.71040916442871,
|
| 212 |
+
"learning_rate": 8.490140146382767e-06,
|
| 213 |
+
"loss": 0.7912,
|
| 214 |
+
"step": 950
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 9.0,
|
| 218 |
+
"eval_accuracy": 0.5752895752895753,
|
| 219 |
+
"eval_loss": 1.0433698892593384,
|
| 220 |
+
"eval_runtime": 1.7507,
|
| 221 |
+
"eval_samples_per_second": 147.94,
|
| 222 |
+
"eval_steps_per_second": 18.85,
|
| 223 |
+
"step": 972
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 9.25925925925926,
|
| 227 |
+
"grad_norm": 35.2958984375,
|
| 228 |
+
"learning_rate": 8.302959444689442e-06,
|
| 229 |
+
"loss": 0.7821,
|
| 230 |
+
"step": 1000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.722222222222221,
|
| 234 |
+
"grad_norm": 47.18446731567383,
|
| 235 |
+
"learning_rate": 8.107178484458825e-06,
|
| 236 |
+
"loss": 0.7529,
|
| 237 |
+
"step": 1050
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 10.0,
|
| 241 |
+
"eval_accuracy": 0.61003861003861,
|
| 242 |
+
"eval_loss": 1.022578477859497,
|
| 243 |
+
"eval_runtime": 1.747,
|
| 244 |
+
"eval_samples_per_second": 148.256,
|
| 245 |
+
"eval_steps_per_second": 18.89,
|
| 246 |
+
"step": 1080
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 10.185185185185185,
|
| 250 |
+
"grad_norm": 18.027130126953125,
|
| 251 |
+
"learning_rate": 7.903307040910035e-06,
|
| 252 |
+
"loss": 0.7478,
|
| 253 |
+
"step": 1100
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 10.648148148148149,
|
| 257 |
+
"grad_norm": 17.87278175354004,
|
| 258 |
+
"learning_rate": 7.691875955293851e-06,
|
| 259 |
+
"loss": 0.6924,
|
| 260 |
+
"step": 1150
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 11.0,
|
| 264 |
+
"eval_accuracy": 0.5868725868725869,
|
| 265 |
+
"eval_loss": 1.0567034482955933,
|
| 266 |
+
"eval_runtime": 1.7505,
|
| 267 |
+
"eval_samples_per_second": 147.961,
|
| 268 |
+
"eval_steps_per_second": 18.852,
|
| 269 |
+
"step": 1188
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 11.11111111111111,
|
| 273 |
+
"grad_norm": 30.35451316833496,
|
| 274 |
+
"learning_rate": 7.473435752686213e-06,
|
| 275 |
+
"loss": 0.7193,
|
| 276 |
+
"step": 1200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 11.574074074074074,
|
| 280 |
+
"grad_norm": 27.861392974853516,
|
| 281 |
+
"learning_rate": 7.248555208528913e-06,
|
| 282 |
+
"loss": 0.7015,
|
| 283 |
+
"step": 1250
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 12.0,
|
| 287 |
+
"eval_accuracy": 0.583011583011583,
|
| 288 |
+
"eval_loss": 1.0600991249084473,
|
| 289 |
+
"eval_runtime": 1.7432,
|
| 290 |
+
"eval_samples_per_second": 148.579,
|
| 291 |
+
"eval_steps_per_second": 18.931,
|
| 292 |
+
"step": 1296
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 12.037037037037036,
|
| 296 |
+
"grad_norm": 16.875736236572266,
|
| 297 |
+
"learning_rate": 7.017819867649898e-06,
|
| 298 |
+
"loss": 0.7261,
|
| 299 |
+
"step": 1300
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 12.5,
|
| 303 |
+
"grad_norm": 29.472204208374023,
|
| 304 |
+
"learning_rate": 6.781830519619392e-06,
|
| 305 |
+
"loss": 0.7286,
|
| 306 |
+
"step": 1350
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 12.962962962962964,
|
| 310 |
+
"grad_norm": 79.9528579711914,
|
| 311 |
+
"learning_rate": 6.541201634411704e-06,
|
| 312 |
+
"loss": 0.7094,
|
| 313 |
+
"step": 1400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 13.0,
|
| 317 |
+
"eval_accuracy": 0.5637065637065637,
|
| 318 |
+
"eval_loss": 1.0830413103103638,
|
| 319 |
+
"eval_runtime": 1.7485,
|
| 320 |
+
"eval_samples_per_second": 148.126,
|
| 321 |
+
"eval_steps_per_second": 18.873,
|
| 322 |
+
"step": 1404
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 13.425925925925926,
|
| 326 |
+
"grad_norm": 32.787437438964844,
|
| 327 |
+
"learning_rate": 6.296559762445928e-06,
|
| 328 |
+
"loss": 0.6587,
|
| 329 |
+
"step": 1450
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.88888888888889,
|
| 333 |
+
"grad_norm": 21.805397033691406,
|
| 334 |
+
"learning_rate": 6.048541903171552e-06,
|
| 335 |
+
"loss": 0.749,
|
| 336 |
+
"step": 1500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 14.0,
|
| 340 |
+
"eval_accuracy": 0.5907335907335908,
|
| 341 |
+
"eval_loss": 1.0383802652359009,
|
| 342 |
+
"eval_runtime": 1.7396,
|
| 343 |
+
"eval_samples_per_second": 148.885,
|
| 344 |
+
"eval_steps_per_second": 18.97,
|
| 345 |
+
"step": 1512
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 14.351851851851851,
|
| 349 |
+
"grad_norm": 14.21861457824707,
|
| 350 |
+
"learning_rate": 5.79779384644684e-06,
|
| 351 |
+
"loss": 0.6565,
|
| 352 |
+
"step": 1550
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 14.814814814814815,
|
| 356 |
+
"grad_norm": 111.92208862304688,
|
| 357 |
+
"learning_rate": 5.550041104970398e-06,
|
| 358 |
+
"loss": 0.7433,
|
| 359 |
+
"step": 1600
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 15.0,
|
| 363 |
+
"eval_accuracy": 0.5675675675675675,
|
| 364 |
+
"eval_loss": 1.0580743551254272,
|
| 365 |
+
"eval_runtime": 1.737,
|
| 366 |
+
"eval_samples_per_second": 149.108,
|
| 367 |
+
"eval_steps_per_second": 18.998,
|
| 368 |
+
"step": 1620
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 15.277777777777779,
|
| 372 |
+
"grad_norm": 27.13836669921875,
|
| 373 |
+
"learning_rate": 5.2958186601645375e-06,
|
| 374 |
+
"loss": 0.7729,
|
| 375 |
+
"step": 1650
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.74074074074074,
|
| 379 |
+
"grad_norm": 41.231163024902344,
|
| 380 |
+
"learning_rate": 5.040825961590755e-06,
|
| 381 |
+
"loss": 0.7595,
|
| 382 |
+
"step": 1700
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 16.0,
|
| 386 |
+
"eval_accuracy": 0.5714285714285714,
|
| 387 |
+
"eval_loss": 1.0641090869903564,
|
| 388 |
+
"eval_runtime": 1.7544,
|
| 389 |
+
"eval_samples_per_second": 147.629,
|
| 390 |
+
"eval_steps_per_second": 18.81,
|
| 391 |
+
"step": 1728
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 16.203703703703702,
|
| 395 |
+
"grad_norm": 18.411924362182617,
|
| 396 |
+
"learning_rate": 4.785726960220769e-06,
|
| 397 |
+
"loss": 0.7803,
|
| 398 |
+
"step": 1750
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 16.666666666666668,
|
| 402 |
+
"grad_norm": 34.5395393371582,
|
| 403 |
+
"learning_rate": 4.531185883817934e-06,
|
| 404 |
+
"loss": 0.8132,
|
| 405 |
+
"step": 1800
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 17.0,
|
| 409 |
+
"eval_accuracy": 0.5714285714285714,
|
| 410 |
+
"eval_loss": 1.0863285064697266,
|
| 411 |
+
"eval_runtime": 1.7323,
|
| 412 |
+
"eval_samples_per_second": 149.51,
|
| 413 |
+
"eval_steps_per_second": 19.05,
|
| 414 |
+
"step": 1836
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 17.12962962962963,
|
| 418 |
+
"grad_norm": 30.300302505493164,
|
| 419 |
+
"learning_rate": 4.277865507418416e-06,
|
| 420 |
+
"loss": 0.8507,
|
| 421 |
+
"step": 1850
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 17.59259259259259,
|
| 425 |
+
"grad_norm": 19.417051315307617,
|
| 426 |
+
"learning_rate": 4.026425427595017e-06,
|
| 427 |
+
"loss": 0.7951,
|
| 428 |
+
"step": 1900
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 18.0,
|
| 432 |
+
"eval_accuracy": 0.5868725868725869,
|
| 433 |
+
"eval_loss": 1.0517489910125732,
|
| 434 |
+
"eval_runtime": 1.7405,
|
| 435 |
+
"eval_samples_per_second": 148.809,
|
| 436 |
+
"eval_steps_per_second": 18.96,
|
| 437 |
+
"step": 1944
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 18.055555555555557,
|
| 441 |
+
"grad_norm": 20.044008255004883,
|
| 442 |
+
"learning_rate": 3.777520344997052e-06,
|
| 443 |
+
"loss": 0.8131,
|
| 444 |
+
"step": 1950
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 18.51851851851852,
|
| 448 |
+
"grad_norm": 29.19565200805664,
|
| 449 |
+
"learning_rate": 3.5317983596382976e-06,
|
| 450 |
+
"loss": 0.7714,
|
| 451 |
+
"step": 2000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 18.98148148148148,
|
| 455 |
+
"grad_norm": 32.99634552001953,
|
| 456 |
+
"learning_rate": 3.289899283371657e-06,
|
| 457 |
+
"loss": 0.7861,
|
| 458 |
+
"step": 2050
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 19.0,
|
| 462 |
+
"eval_accuracy": 0.5907335907335908,
|
| 463 |
+
"eval_loss": 1.0631417036056519,
|
| 464 |
+
"eval_runtime": 1.739,
|
| 465 |
+
"eval_samples_per_second": 148.94,
|
| 466 |
+
"eval_steps_per_second": 18.977,
|
| 467 |
+
"step": 2052
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 19.444444444444443,
|
| 471 |
+
"grad_norm": 20.90323829650879,
|
| 472 |
+
"learning_rate": 3.052452973944603e-06,
|
| 473 |
+
"loss": 0.7129,
|
| 474 |
+
"step": 2100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 19.90740740740741,
|
| 478 |
+
"grad_norm": 40.24602508544922,
|
| 479 |
+
"learning_rate": 2.8200776949731435e-06,
|
| 480 |
+
"loss": 0.7334,
|
| 481 |
+
"step": 2150
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 20.0,
|
| 485 |
+
"eval_accuracy": 0.5907335907335908,
|
| 486 |
+
"eval_loss": 1.0507447719573975,
|
| 487 |
+
"eval_runtime": 1.893,
|
| 488 |
+
"eval_samples_per_second": 136.818,
|
| 489 |
+
"eval_steps_per_second": 17.432,
|
| 490 |
+
"step": 2160
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 20.37037037037037,
|
| 494 |
+
"grad_norm": 19.65435028076172,
|
| 495 |
+
"learning_rate": 2.5933785061046333e-06,
|
| 496 |
+
"loss": 0.7563,
|
| 497 |
+
"step": 2200
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 20.833333333333332,
|
| 501 |
+
"grad_norm": 96.42213439941406,
|
| 502 |
+
"learning_rate": 2.372945687561132e-06,
|
| 503 |
+
"loss": 0.6807,
|
| 504 |
+
"step": 2250
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 21.0,
|
| 508 |
+
"eval_accuracy": 0.5791505791505791,
|
| 509 |
+
"eval_loss": 1.0659432411193848,
|
| 510 |
+
"eval_runtime": 1.7425,
|
| 511 |
+
"eval_samples_per_second": 148.637,
|
| 512 |
+
"eval_steps_per_second": 18.938,
|
| 513 |
+
"step": 2268
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 21.296296296296298,
|
| 517 |
+
"grad_norm": 30.792505264282227,
|
| 518 |
+
"learning_rate": 2.159353203165484e-06,
|
| 519 |
+
"loss": 0.7706,
|
| 520 |
+
"step": 2300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 21.75925925925926,
|
| 524 |
+
"grad_norm": 62.397857666015625,
|
| 525 |
+
"learning_rate": 1.9531572058521074e-06,
|
| 526 |
+
"loss": 0.6905,
|
| 527 |
+
"step": 2350
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 22.0,
|
| 531 |
+
"eval_accuracy": 0.5752895752895753,
|
| 532 |
+
"eval_loss": 1.0680582523345947,
|
| 533 |
+
"eval_runtime": 1.7542,
|
| 534 |
+
"eval_samples_per_second": 147.645,
|
| 535 |
+
"eval_steps_per_second": 18.812,
|
| 536 |
+
"step": 2376
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 22.22222222222222,
|
| 540 |
+
"grad_norm": 17.79286766052246,
|
| 541 |
+
"learning_rate": 1.754894589553836e-06,
|
| 542 |
+
"loss": 0.763,
|
| 543 |
+
"step": 2400
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 22.685185185185187,
|
| 547 |
+
"grad_norm": 20.265037536621094,
|
| 548 |
+
"learning_rate": 1.5650815912354285e-06,
|
| 549 |
+
"loss": 0.7242,
|
| 550 |
+
"step": 2450
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 23.0,
|
| 554 |
+
"eval_accuracy": 0.5907335907335908,
|
| 555 |
+
"eval_loss": 1.0529667139053345,
|
| 556 |
+
"eval_runtime": 1.7502,
|
| 557 |
+
"eval_samples_per_second": 147.984,
|
| 558 |
+
"eval_steps_per_second": 18.855,
|
| 559 |
+
"step": 2484
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 23.14814814814815,
|
| 563 |
+
"grad_norm": 43.985389709472656,
|
| 564 |
+
"learning_rate": 1.3842124467137664e-06,
|
| 565 |
+
"loss": 0.7003,
|
| 566 |
+
"step": 2500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 23.61111111111111,
|
| 570 |
+
"grad_norm": 58.851200103759766,
|
| 571 |
+
"learning_rate": 1.212758103764733e-06,
|
| 572 |
+
"loss": 0.7342,
|
| 573 |
+
"step": 2550
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 24.0,
|
| 577 |
+
"eval_accuracy": 0.5637065637065637,
|
| 578 |
+
"eval_loss": 1.0659518241882324,
|
| 579 |
+
"eval_runtime": 1.7466,
|
| 580 |
+
"eval_samples_per_second": 148.291,
|
| 581 |
+
"eval_steps_per_second": 18.894,
|
| 582 |
+
"step": 2592
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 24.074074074074073,
|
| 586 |
+
"grad_norm": 34.4920654296875,
|
| 587 |
+
"learning_rate": 1.0511649958675745e-06,
|
| 588 |
+
"loss": 0.7262,
|
| 589 |
+
"step": 2600
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 24.537037037037038,
|
| 593 |
+
"grad_norm": 29.049121856689453,
|
| 594 |
+
"learning_rate": 8.998538797796879e-07,
|
| 595 |
+
"loss": 0.6924,
|
| 596 |
+
"step": 2650
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 25.0,
|
| 600 |
+
"grad_norm": 54.39902877807617,
|
| 601 |
+
"learning_rate": 7.592187399685524e-07,
|
| 602 |
+
"loss": 0.7736,
|
| 603 |
+
"step": 2700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 25.0,
|
| 607 |
+
"eval_accuracy": 0.5791505791505791,
|
| 608 |
+
"eval_loss": 1.070913314819336,
|
| 609 |
+
"eval_runtime": 1.7438,
|
| 610 |
+
"eval_samples_per_second": 148.528,
|
| 611 |
+
"eval_steps_per_second": 18.924,
|
| 612 |
+
"step": 2700
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 25.462962962962962,
|
| 616 |
+
"grad_norm": 41.184539794921875,
|
| 617 |
+
"learning_rate": 6.296257627534552e-07,
|
| 618 |
+
"loss": 0.6561,
|
| 619 |
+
"step": 2750
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 25.925925925925927,
|
| 623 |
+
"grad_norm": 57.05064010620117,
|
| 624 |
+
"learning_rate": 5.114123828281398e-07,
|
| 625 |
+
"loss": 0.7798,
|
| 626 |
+
"step": 2800
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 26.0,
|
| 630 |
+
"eval_accuracy": 0.5714285714285714,
|
| 631 |
+
"eval_loss": 1.0627210140228271,
|
| 632 |
+
"eval_runtime": 1.737,
|
| 633 |
+
"eval_samples_per_second": 149.108,
|
| 634 |
+
"eval_steps_per_second": 18.998,
|
| 635 |
+
"step": 2808
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 26.38888888888889,
|
| 639 |
+
"grad_norm": 29.81525993347168,
|
| 640 |
+
"learning_rate": 4.048864046470502e-07,
|
| 641 |
+
"loss": 0.7423,
|
| 642 |
+
"step": 2850
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 26.85185185185185,
|
| 646 |
+
"grad_norm": 36.73592758178711,
|
| 647 |
+
"learning_rate": 3.1032520096290817e-07,
|
| 648 |
+
"loss": 0.71,
|
| 649 |
+
"step": 2900
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 27.0,
|
| 653 |
+
"eval_accuracy": 0.5868725868725869,
|
| 654 |
+
"eval_loss": 1.060496211051941,
|
| 655 |
+
"eval_runtime": 1.7352,
|
| 656 |
+
"eval_samples_per_second": 149.263,
|
| 657 |
+
"eval_steps_per_second": 19.018,
|
| 658 |
+
"step": 2916
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 27.314814814814813,
|
| 662 |
+
"grad_norm": 63.34288024902344,
|
| 663 |
+
"learning_rate": 2.2797499060246253e-07,
|
| 664 |
+
"loss": 0.6777,
|
| 665 |
+
"step": 2950
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 27.77777777777778,
|
| 669 |
+
"grad_norm": 25.37461280822754,
|
| 670 |
+
"learning_rate": 1.5805019736097105e-07,
|
| 671 |
+
"loss": 0.706,
|
| 672 |
+
"step": 3000
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 28.0,
|
| 676 |
+
"eval_accuracy": 0.5791505791505791,
|
| 677 |
+
"eval_loss": 1.0605989694595337,
|
| 678 |
+
"eval_runtime": 1.7369,
|
| 679 |
+
"eval_samples_per_second": 149.116,
|
| 680 |
+
"eval_steps_per_second": 18.999,
|
| 681 |
+
"step": 3024
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 28.24074074074074,
|
| 685 |
+
"grad_norm": 42.660560607910156,
|
| 686 |
+
"learning_rate": 1.007328916846817e-07,
|
| 687 |
+
"loss": 0.7275,
|
| 688 |
+
"step": 3050
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 28.703703703703702,
|
| 692 |
+
"grad_norm": 31.769351959228516,
|
| 693 |
+
"learning_rate": 5.617231659511446e-08,
|
| 694 |
+
"loss": 0.7421,
|
| 695 |
+
"step": 3100
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 29.0,
|
| 699 |
+
"eval_accuracy": 0.5752895752895753,
|
| 700 |
+
"eval_loss": 1.0606402158737183,
|
| 701 |
+
"eval_runtime": 1.7325,
|
| 702 |
+
"eval_samples_per_second": 149.494,
|
| 703 |
+
"eval_steps_per_second": 19.048,
|
| 704 |
+
"step": 3132
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 29.166666666666668,
|
| 708 |
+
"grad_norm": 23.639081954956055,
|
| 709 |
+
"learning_rate": 2.448449908949879e-08,
|
| 710 |
+
"loss": 0.6976,
|
| 711 |
+
"step": 3150
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 29.62962962962963,
|
| 715 |
+
"grad_norm": 30.702539443969727,
|
| 716 |
+
"learning_rate": 5.751948029234511e-09,
|
| 717 |
+
"loss": 0.654,
|
| 718 |
+
"step": 3200
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 30.0,
|
| 722 |
+
"eval_accuracy": 0.5752895752895753,
|
| 723 |
+
"eval_loss": 1.060823917388916,
|
| 724 |
+
"eval_runtime": 3.0177,
|
| 725 |
+
"eval_samples_per_second": 85.826,
|
| 726 |
+
"eval_steps_per_second": 10.935,
|
| 727 |
+
"step": 3240
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 30.0,
|
| 731 |
+
"step": 3240,
|
| 732 |
+
"total_flos": 2.0405460235862016e+18,
|
| 733 |
+
"train_loss": 0.7678916913491708,
|
| 734 |
+
"train_runtime": 680.1267,
|
| 735 |
+
"train_samples_per_second": 37.934,
|
| 736 |
+
"train_steps_per_second": 4.764
|
| 737 |
+
}
|
| 738 |
+
],
|
| 739 |
+
"logging_steps": 50,
|
| 740 |
+
"max_steps": 3240,
|
| 741 |
+
"num_input_tokens_seen": 0,
|
| 742 |
+
"num_train_epochs": 30,
|
| 743 |
+
"save_steps": 500,
|
| 744 |
+
"stateful_callbacks": {
|
| 745 |
+
"TrainerControl": {
|
| 746 |
+
"args": {
|
| 747 |
+
"should_epoch_stop": false,
|
| 748 |
+
"should_evaluate": false,
|
| 749 |
+
"should_log": false,
|
| 750 |
+
"should_save": true,
|
| 751 |
+
"should_training_stop": true
|
| 752 |
+
},
|
| 753 |
+
"attributes": {}
|
| 754 |
+
}
|
| 755 |
+
},
|
| 756 |
+
"total_flos": 2.0405460235862016e+18,
|
| 757 |
+
"train_batch_size": 8,
|
| 758 |
+
"trial_name": null,
|
| 759 |
+
"trial_params": null
|
| 760 |
+
}
|