End of training
Browse files- README.md +5 -5
- all_results.json +16 -0
- eval_results.json +11 -0
- train_results.json +8 -0
- trainer_state.json +913 -0
README.md
CHANGED
|
@@ -23,11 +23,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 23 |
|
| 24 |
This model is a fine-tuned version of [microsoft/swin-large-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-large-patch4-window7-224-in22k) on the medmnist-v2 dataset.
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
-
- Loss: 0.
|
| 27 |
-
- Accuracy: 0.
|
| 28 |
-
- Precision: 0.
|
| 29 |
-
- Recall: 0.
|
| 30 |
-
- F1: 0.
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
|
|
|
| 23 |
|
| 24 |
This model is a fine-tuned version of [microsoft/swin-large-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-large-patch4-window7-224-in22k) on the medmnist-v2 dataset.
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
+
- Loss: 0.5530
|
| 27 |
+
- Accuracy: 0.7875
|
| 28 |
+
- Precision: 0.6308
|
| 29 |
+
- Recall: 0.6101
|
| 30 |
+
- F1: 0.6150
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_accuracy": 0.7875311720698255,
|
| 4 |
+
"eval_f1": 0.6149515279179031,
|
| 5 |
+
"eval_loss": 0.5530101656913757,
|
| 6 |
+
"eval_precision": 0.6308444617762794,
|
| 7 |
+
"eval_recall": 0.6101381269194385,
|
| 8 |
+
"eval_runtime": 22.3695,
|
| 9 |
+
"eval_samples_per_second": 89.631,
|
| 10 |
+
"eval_steps_per_second": 5.633,
|
| 11 |
+
"total_flos": 1.2358226368562393e+19,
|
| 12 |
+
"train_loss": 0.6880708248243419,
|
| 13 |
+
"train_runtime": 1791.2534,
|
| 14 |
+
"train_samples_per_second": 39.118,
|
| 15 |
+
"train_steps_per_second": 0.609
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_accuracy": 0.7875311720698255,
|
| 4 |
+
"eval_f1": 0.6149515279179031,
|
| 5 |
+
"eval_loss": 0.5530101656913757,
|
| 6 |
+
"eval_precision": 0.6308444617762794,
|
| 7 |
+
"eval_recall": 0.6101381269194385,
|
| 8 |
+
"eval_runtime": 22.3695,
|
| 9 |
+
"eval_samples_per_second": 89.631,
|
| 10 |
+
"eval_steps_per_second": 5.633
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"total_flos": 1.2358226368562393e+19,
|
| 4 |
+
"train_loss": 0.6880708248243419,
|
| 5 |
+
"train_runtime": 1791.2534,
|
| 6 |
+
"train_samples_per_second": 39.118,
|
| 7 |
+
"train_steps_per_second": 0.609
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,913 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7926221335992024,
|
| 3 |
+
"best_model_checkpoint": "swin-large-patch4-window7-224-in22k-finetuned-lora-medmnistv2/checkpoint-985",
|
| 4 |
+
"epoch": 9.954337899543379,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1090,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.09,
|
| 13 |
+
"grad_norm": 3.442108154296875,
|
| 14 |
+
"learning_rate": 0.004954128440366973,
|
| 15 |
+
"loss": 1.1496,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.18,
|
| 20 |
+
"grad_norm": 1.3745663166046143,
|
| 21 |
+
"learning_rate": 0.004908256880733945,
|
| 22 |
+
"loss": 0.9658,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.27,
|
| 27 |
+
"grad_norm": 1.6516236066818237,
|
| 28 |
+
"learning_rate": 0.004862385321100918,
|
| 29 |
+
"loss": 0.9225,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.37,
|
| 34 |
+
"grad_norm": 1.4029324054718018,
|
| 35 |
+
"learning_rate": 0.00481651376146789,
|
| 36 |
+
"loss": 0.8968,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.46,
|
| 41 |
+
"grad_norm": 1.2768313884735107,
|
| 42 |
+
"learning_rate": 0.0047706422018348625,
|
| 43 |
+
"loss": 0.8582,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.55,
|
| 48 |
+
"grad_norm": 1.250333309173584,
|
| 49 |
+
"learning_rate": 0.004724770642201835,
|
| 50 |
+
"loss": 0.8362,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.64,
|
| 55 |
+
"grad_norm": 1.2696894407272339,
|
| 56 |
+
"learning_rate": 0.004678899082568808,
|
| 57 |
+
"loss": 0.8948,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.73,
|
| 62 |
+
"grad_norm": 1.02017080783844,
|
| 63 |
+
"learning_rate": 0.00463302752293578,
|
| 64 |
+
"loss": 0.8448,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.82,
|
| 69 |
+
"grad_norm": 1.1102752685546875,
|
| 70 |
+
"learning_rate": 0.0045871559633027525,
|
| 71 |
+
"loss": 0.8288,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.91,
|
| 76 |
+
"grad_norm": 0.8990870714187622,
|
| 77 |
+
"learning_rate": 0.004541284403669725,
|
| 78 |
+
"loss": 0.8516,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.0,
|
| 83 |
+
"eval_accuracy": 0.7547357926221336,
|
| 84 |
+
"eval_f1": 0.41878801809381894,
|
| 85 |
+
"eval_loss": 0.7470152974128723,
|
| 86 |
+
"eval_precision": 0.5455908887866956,
|
| 87 |
+
"eval_recall": 0.3913939658670338,
|
| 88 |
+
"eval_runtime": 11.5403,
|
| 89 |
+
"eval_samples_per_second": 86.913,
|
| 90 |
+
"eval_steps_per_second": 5.459,
|
| 91 |
+
"step": 109
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.0,
|
| 95 |
+
"grad_norm": 1.1006461381912231,
|
| 96 |
+
"learning_rate": 0.004495412844036698,
|
| 97 |
+
"loss": 0.8742,
|
| 98 |
+
"step": 110
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.1,
|
| 102 |
+
"grad_norm": 1.271594524383545,
|
| 103 |
+
"learning_rate": 0.0044495412844036695,
|
| 104 |
+
"loss": 0.8544,
|
| 105 |
+
"step": 120
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.19,
|
| 109 |
+
"grad_norm": 1.24767005443573,
|
| 110 |
+
"learning_rate": 0.004403669724770643,
|
| 111 |
+
"loss": 0.792,
|
| 112 |
+
"step": 130
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.28,
|
| 116 |
+
"grad_norm": 1.5129131078720093,
|
| 117 |
+
"learning_rate": 0.004357798165137615,
|
| 118 |
+
"loss": 0.8107,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.37,
|
| 123 |
+
"grad_norm": 0.728676974773407,
|
| 124 |
+
"learning_rate": 0.004311926605504587,
|
| 125 |
+
"loss": 0.7921,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.46,
|
| 130 |
+
"grad_norm": 1.073712944984436,
|
| 131 |
+
"learning_rate": 0.0042660550458715595,
|
| 132 |
+
"loss": 0.7453,
|
| 133 |
+
"step": 160
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.55,
|
| 137 |
+
"grad_norm": 1.1818609237670898,
|
| 138 |
+
"learning_rate": 0.004220183486238533,
|
| 139 |
+
"loss": 0.8023,
|
| 140 |
+
"step": 170
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.64,
|
| 144 |
+
"grad_norm": 0.6278601884841919,
|
| 145 |
+
"learning_rate": 0.004174311926605505,
|
| 146 |
+
"loss": 0.8196,
|
| 147 |
+
"step": 180
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.74,
|
| 151 |
+
"grad_norm": 0.796804666519165,
|
| 152 |
+
"learning_rate": 0.004128440366972477,
|
| 153 |
+
"loss": 0.8063,
|
| 154 |
+
"step": 190
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.83,
|
| 158 |
+
"grad_norm": 1.4323725700378418,
|
| 159 |
+
"learning_rate": 0.00408256880733945,
|
| 160 |
+
"loss": 0.7128,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.92,
|
| 165 |
+
"grad_norm": 0.5659187436103821,
|
| 166 |
+
"learning_rate": 0.004036697247706422,
|
| 167 |
+
"loss": 0.7738,
|
| 168 |
+
"step": 210
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.0,
|
| 172 |
+
"eval_accuracy": 0.7168494516450648,
|
| 173 |
+
"eval_f1": 0.3577314775338443,
|
| 174 |
+
"eval_loss": 0.8952543139457703,
|
| 175 |
+
"eval_precision": 0.42246935817657555,
|
| 176 |
+
"eval_recall": 0.4458814441250039,
|
| 177 |
+
"eval_runtime": 11.2901,
|
| 178 |
+
"eval_samples_per_second": 88.839,
|
| 179 |
+
"eval_steps_per_second": 5.58,
|
| 180 |
+
"step": 219
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.01,
|
| 184 |
+
"grad_norm": 2.2657077312469482,
|
| 185 |
+
"learning_rate": 0.003990825688073394,
|
| 186 |
+
"loss": 0.8065,
|
| 187 |
+
"step": 220
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.1,
|
| 191 |
+
"grad_norm": 0.6983966827392578,
|
| 192 |
+
"learning_rate": 0.003944954128440367,
|
| 193 |
+
"loss": 0.8244,
|
| 194 |
+
"step": 230
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.19,
|
| 198 |
+
"grad_norm": 0.9509605765342712,
|
| 199 |
+
"learning_rate": 0.0038990825688073397,
|
| 200 |
+
"loss": 0.7429,
|
| 201 |
+
"step": 240
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.28,
|
| 205 |
+
"grad_norm": 0.8736411333084106,
|
| 206 |
+
"learning_rate": 0.0038532110091743124,
|
| 207 |
+
"loss": 0.7479,
|
| 208 |
+
"step": 250
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.37,
|
| 212 |
+
"grad_norm": 1.0205260515213013,
|
| 213 |
+
"learning_rate": 0.0038073394495412843,
|
| 214 |
+
"loss": 0.7733,
|
| 215 |
+
"step": 260
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.47,
|
| 219 |
+
"grad_norm": 0.9213468432426453,
|
| 220 |
+
"learning_rate": 0.003761467889908257,
|
| 221 |
+
"loss": 0.6993,
|
| 222 |
+
"step": 270
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.56,
|
| 226 |
+
"grad_norm": 0.7223290801048279,
|
| 227 |
+
"learning_rate": 0.0037155963302752293,
|
| 228 |
+
"loss": 0.7582,
|
| 229 |
+
"step": 280
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.65,
|
| 233 |
+
"grad_norm": 0.8736669421195984,
|
| 234 |
+
"learning_rate": 0.003669724770642202,
|
| 235 |
+
"loss": 0.7483,
|
| 236 |
+
"step": 290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.74,
|
| 240 |
+
"grad_norm": 1.2014636993408203,
|
| 241 |
+
"learning_rate": 0.0036238532110091743,
|
| 242 |
+
"loss": 0.7592,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.83,
|
| 247 |
+
"grad_norm": 1.290935754776001,
|
| 248 |
+
"learning_rate": 0.003577981651376147,
|
| 249 |
+
"loss": 0.819,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.92,
|
| 254 |
+
"grad_norm": 0.751524806022644,
|
| 255 |
+
"learning_rate": 0.0035321100917431194,
|
| 256 |
+
"loss": 0.6994,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"eval_accuracy": 0.7607178464606181,
|
| 262 |
+
"eval_f1": 0.5104809166351773,
|
| 263 |
+
"eval_loss": 0.6592820286750793,
|
| 264 |
+
"eval_precision": 0.6256931769842607,
|
| 265 |
+
"eval_recall": 0.5058522052667251,
|
| 266 |
+
"eval_runtime": 11.3494,
|
| 267 |
+
"eval_samples_per_second": 88.375,
|
| 268 |
+
"eval_steps_per_second": 5.551,
|
| 269 |
+
"step": 328
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.01,
|
| 273 |
+
"grad_norm": 1.0127859115600586,
|
| 274 |
+
"learning_rate": 0.003486238532110092,
|
| 275 |
+
"loss": 0.723,
|
| 276 |
+
"step": 330
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 3.11,
|
| 280 |
+
"grad_norm": 0.8557056784629822,
|
| 281 |
+
"learning_rate": 0.0034403669724770644,
|
| 282 |
+
"loss": 0.762,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 3.2,
|
| 287 |
+
"grad_norm": 0.7101117968559265,
|
| 288 |
+
"learning_rate": 0.003394495412844037,
|
| 289 |
+
"loss": 0.746,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 3.29,
|
| 294 |
+
"grad_norm": 0.732359766960144,
|
| 295 |
+
"learning_rate": 0.003348623853211009,
|
| 296 |
+
"loss": 0.7152,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.38,
|
| 301 |
+
"grad_norm": 0.7687917947769165,
|
| 302 |
+
"learning_rate": 0.0033027522935779817,
|
| 303 |
+
"loss": 0.6849,
|
| 304 |
+
"step": 370
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.47,
|
| 308 |
+
"grad_norm": 0.9754200577735901,
|
| 309 |
+
"learning_rate": 0.003256880733944954,
|
| 310 |
+
"loss": 0.721,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.56,
|
| 315 |
+
"grad_norm": 0.9926655888557434,
|
| 316 |
+
"learning_rate": 0.003211009174311927,
|
| 317 |
+
"loss": 0.7457,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.65,
|
| 322 |
+
"grad_norm": 1.4650967121124268,
|
| 323 |
+
"learning_rate": 0.003165137614678899,
|
| 324 |
+
"loss": 0.7209,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.74,
|
| 329 |
+
"grad_norm": 0.7933741211891174,
|
| 330 |
+
"learning_rate": 0.003119266055045872,
|
| 331 |
+
"loss": 0.7785,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.84,
|
| 336 |
+
"grad_norm": 0.7204951047897339,
|
| 337 |
+
"learning_rate": 0.003073394495412844,
|
| 338 |
+
"loss": 0.7276,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.93,
|
| 343 |
+
"grad_norm": 1.0182371139526367,
|
| 344 |
+
"learning_rate": 0.003027522935779817,
|
| 345 |
+
"loss": 0.6731,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 4.0,
|
| 350 |
+
"eval_accuracy": 0.7716849451645065,
|
| 351 |
+
"eval_f1": 0.5382789492033663,
|
| 352 |
+
"eval_loss": 0.6144729852676392,
|
| 353 |
+
"eval_precision": 0.6321625904242083,
|
| 354 |
+
"eval_recall": 0.500050058598068,
|
| 355 |
+
"eval_runtime": 11.305,
|
| 356 |
+
"eval_samples_per_second": 88.722,
|
| 357 |
+
"eval_steps_per_second": 5.573,
|
| 358 |
+
"step": 438
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 4.02,
|
| 362 |
+
"grad_norm": 0.5864942669868469,
|
| 363 |
+
"learning_rate": 0.002981651376146789,
|
| 364 |
+
"loss": 0.685,
|
| 365 |
+
"step": 440
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.11,
|
| 369 |
+
"grad_norm": 0.816683292388916,
|
| 370 |
+
"learning_rate": 0.002935779816513762,
|
| 371 |
+
"loss": 0.6859,
|
| 372 |
+
"step": 450
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 4.2,
|
| 376 |
+
"grad_norm": 0.6105664968490601,
|
| 377 |
+
"learning_rate": 0.0028899082568807338,
|
| 378 |
+
"loss": 0.6785,
|
| 379 |
+
"step": 460
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 4.29,
|
| 383 |
+
"grad_norm": 1.0114610195159912,
|
| 384 |
+
"learning_rate": 0.0028440366972477065,
|
| 385 |
+
"loss": 0.6661,
|
| 386 |
+
"step": 470
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.38,
|
| 390 |
+
"grad_norm": 1.1204463243484497,
|
| 391 |
+
"learning_rate": 0.002798165137614679,
|
| 392 |
+
"loss": 0.664,
|
| 393 |
+
"step": 480
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.47,
|
| 397 |
+
"grad_norm": 1.0051954984664917,
|
| 398 |
+
"learning_rate": 0.0027522935779816515,
|
| 399 |
+
"loss": 0.7172,
|
| 400 |
+
"step": 490
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.57,
|
| 404 |
+
"grad_norm": 0.6869407892227173,
|
| 405 |
+
"learning_rate": 0.002706422018348624,
|
| 406 |
+
"loss": 0.7341,
|
| 407 |
+
"step": 500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.66,
|
| 411 |
+
"grad_norm": 0.720870316028595,
|
| 412 |
+
"learning_rate": 0.0026605504587155966,
|
| 413 |
+
"loss": 0.7028,
|
| 414 |
+
"step": 510
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.75,
|
| 418 |
+
"grad_norm": 0.9448994994163513,
|
| 419 |
+
"learning_rate": 0.002614678899082569,
|
| 420 |
+
"loss": 0.7383,
|
| 421 |
+
"step": 520
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 4.84,
|
| 425 |
+
"grad_norm": 0.9630647897720337,
|
| 426 |
+
"learning_rate": 0.0025688073394495416,
|
| 427 |
+
"loss": 0.6727,
|
| 428 |
+
"step": 530
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.93,
|
| 432 |
+
"grad_norm": 0.8725408315658569,
|
| 433 |
+
"learning_rate": 0.0025229357798165135,
|
| 434 |
+
"loss": 0.7266,
|
| 435 |
+
"step": 540
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 5.0,
|
| 439 |
+
"eval_accuracy": 0.7397806580259222,
|
| 440 |
+
"eval_f1": 0.4934894169090878,
|
| 441 |
+
"eval_loss": 0.6838864088058472,
|
| 442 |
+
"eval_precision": 0.5519819307791096,
|
| 443 |
+
"eval_recall": 0.5344054167239178,
|
| 444 |
+
"eval_runtime": 11.2776,
|
| 445 |
+
"eval_samples_per_second": 88.937,
|
| 446 |
+
"eval_steps_per_second": 5.586,
|
| 447 |
+
"step": 547
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 5.02,
|
| 451 |
+
"grad_norm": 1.1792114973068237,
|
| 452 |
+
"learning_rate": 0.0024770642201834866,
|
| 453 |
+
"loss": 0.6676,
|
| 454 |
+
"step": 550
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 5.11,
|
| 458 |
+
"grad_norm": 0.7483982443809509,
|
| 459 |
+
"learning_rate": 0.002431192660550459,
|
| 460 |
+
"loss": 0.7229,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 5.21,
|
| 465 |
+
"grad_norm": 0.8732199668884277,
|
| 466 |
+
"learning_rate": 0.0023853211009174312,
|
| 467 |
+
"loss": 0.6366,
|
| 468 |
+
"step": 570
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 5.3,
|
| 472 |
+
"grad_norm": 1.1174126863479614,
|
| 473 |
+
"learning_rate": 0.002339449541284404,
|
| 474 |
+
"loss": 0.6855,
|
| 475 |
+
"step": 580
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 5.39,
|
| 479 |
+
"grad_norm": 0.7935824990272522,
|
| 480 |
+
"learning_rate": 0.0022935779816513763,
|
| 481 |
+
"loss": 0.7225,
|
| 482 |
+
"step": 590
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 5.48,
|
| 486 |
+
"grad_norm": 0.9188850522041321,
|
| 487 |
+
"learning_rate": 0.002247706422018349,
|
| 488 |
+
"loss": 0.6657,
|
| 489 |
+
"step": 600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 5.57,
|
| 493 |
+
"grad_norm": 0.7955396771430969,
|
| 494 |
+
"learning_rate": 0.0022018348623853213,
|
| 495 |
+
"loss": 0.6797,
|
| 496 |
+
"step": 610
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 5.66,
|
| 500 |
+
"grad_norm": 1.0317906141281128,
|
| 501 |
+
"learning_rate": 0.0021559633027522936,
|
| 502 |
+
"loss": 0.6589,
|
| 503 |
+
"step": 620
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.75,
|
| 507 |
+
"grad_norm": 0.6561917662620544,
|
| 508 |
+
"learning_rate": 0.0021100917431192663,
|
| 509 |
+
"loss": 0.6435,
|
| 510 |
+
"step": 630
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 5.84,
|
| 514 |
+
"grad_norm": 0.8070980906486511,
|
| 515 |
+
"learning_rate": 0.0020642201834862386,
|
| 516 |
+
"loss": 0.6303,
|
| 517 |
+
"step": 640
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 5.94,
|
| 521 |
+
"grad_norm": 0.6900261044502258,
|
| 522 |
+
"learning_rate": 0.002018348623853211,
|
| 523 |
+
"loss": 0.6388,
|
| 524 |
+
"step": 650
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 6.0,
|
| 528 |
+
"eval_accuracy": 0.7666999002991027,
|
| 529 |
+
"eval_f1": 0.5338252978902394,
|
| 530 |
+
"eval_loss": 0.6242878437042236,
|
| 531 |
+
"eval_precision": 0.6116740127093266,
|
| 532 |
+
"eval_recall": 0.5062826891866704,
|
| 533 |
+
"eval_runtime": 11.256,
|
| 534 |
+
"eval_samples_per_second": 89.108,
|
| 535 |
+
"eval_steps_per_second": 5.597,
|
| 536 |
+
"step": 657
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 6.03,
|
| 540 |
+
"grad_norm": 0.8859825134277344,
|
| 541 |
+
"learning_rate": 0.0019724770642201837,
|
| 542 |
+
"loss": 0.6252,
|
| 543 |
+
"step": 660
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 6.12,
|
| 547 |
+
"grad_norm": 0.7855720520019531,
|
| 548 |
+
"learning_rate": 0.0019266055045871562,
|
| 549 |
+
"loss": 0.6026,
|
| 550 |
+
"step": 670
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 6.21,
|
| 554 |
+
"grad_norm": 0.7128086090087891,
|
| 555 |
+
"learning_rate": 0.0018807339449541285,
|
| 556 |
+
"loss": 0.5799,
|
| 557 |
+
"step": 680
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 6.3,
|
| 561 |
+
"grad_norm": 0.6269203424453735,
|
| 562 |
+
"learning_rate": 0.001834862385321101,
|
| 563 |
+
"loss": 0.6358,
|
| 564 |
+
"step": 690
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 6.39,
|
| 568 |
+
"grad_norm": 1.1196659803390503,
|
| 569 |
+
"learning_rate": 0.0017889908256880735,
|
| 570 |
+
"loss": 0.6282,
|
| 571 |
+
"step": 700
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 6.48,
|
| 575 |
+
"grad_norm": 1.253045678138733,
|
| 576 |
+
"learning_rate": 0.001743119266055046,
|
| 577 |
+
"loss": 0.6109,
|
| 578 |
+
"step": 710
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 6.58,
|
| 582 |
+
"grad_norm": 0.946592390537262,
|
| 583 |
+
"learning_rate": 0.0016972477064220186,
|
| 584 |
+
"loss": 0.58,
|
| 585 |
+
"step": 720
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.67,
|
| 589 |
+
"grad_norm": 0.9241579174995422,
|
| 590 |
+
"learning_rate": 0.0016513761467889909,
|
| 591 |
+
"loss": 0.6492,
|
| 592 |
+
"step": 730
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 6.76,
|
| 596 |
+
"grad_norm": 0.9794093370437622,
|
| 597 |
+
"learning_rate": 0.0016055045871559634,
|
| 598 |
+
"loss": 0.6333,
|
| 599 |
+
"step": 740
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 6.85,
|
| 603 |
+
"grad_norm": 0.9750506281852722,
|
| 604 |
+
"learning_rate": 0.001559633027522936,
|
| 605 |
+
"loss": 0.6883,
|
| 606 |
+
"step": 750
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 6.94,
|
| 610 |
+
"grad_norm": 1.1783692836761475,
|
| 611 |
+
"learning_rate": 0.0015137614678899084,
|
| 612 |
+
"loss": 0.6495,
|
| 613 |
+
"step": 760
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 7.0,
|
| 617 |
+
"eval_accuracy": 0.7826520438683948,
|
| 618 |
+
"eval_f1": 0.6162957297860807,
|
| 619 |
+
"eval_loss": 0.6161015033721924,
|
| 620 |
+
"eval_precision": 0.6357243111624864,
|
| 621 |
+
"eval_recall": 0.6152698884549003,
|
| 622 |
+
"eval_runtime": 11.1928,
|
| 623 |
+
"eval_samples_per_second": 89.611,
|
| 624 |
+
"eval_steps_per_second": 5.629,
|
| 625 |
+
"step": 766
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.03,
|
| 629 |
+
"grad_norm": 1.0667985677719116,
|
| 630 |
+
"learning_rate": 0.001467889908256881,
|
| 631 |
+
"loss": 0.6615,
|
| 632 |
+
"step": 770
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.12,
|
| 636 |
+
"grad_norm": 1.2581771612167358,
|
| 637 |
+
"learning_rate": 0.0014220183486238532,
|
| 638 |
+
"loss": 0.5515,
|
| 639 |
+
"step": 780
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 7.21,
|
| 643 |
+
"grad_norm": 1.012791395187378,
|
| 644 |
+
"learning_rate": 0.0013761467889908258,
|
| 645 |
+
"loss": 0.607,
|
| 646 |
+
"step": 790
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.31,
|
| 650 |
+
"grad_norm": 0.9969860911369324,
|
| 651 |
+
"learning_rate": 0.0013302752293577983,
|
| 652 |
+
"loss": 0.6396,
|
| 653 |
+
"step": 800
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.4,
|
| 657 |
+
"grad_norm": 0.9849869608879089,
|
| 658 |
+
"learning_rate": 0.0012844036697247708,
|
| 659 |
+
"loss": 0.6563,
|
| 660 |
+
"step": 810
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.49,
|
| 664 |
+
"grad_norm": 1.0733628273010254,
|
| 665 |
+
"learning_rate": 0.0012385321100917433,
|
| 666 |
+
"loss": 0.5833,
|
| 667 |
+
"step": 820
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.58,
|
| 671 |
+
"grad_norm": 1.044959306716919,
|
| 672 |
+
"learning_rate": 0.0011926605504587156,
|
| 673 |
+
"loss": 0.5745,
|
| 674 |
+
"step": 830
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.67,
|
| 678 |
+
"grad_norm": 0.8886712193489075,
|
| 679 |
+
"learning_rate": 0.0011467889908256881,
|
| 680 |
+
"loss": 0.5437,
|
| 681 |
+
"step": 840
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.76,
|
| 685 |
+
"grad_norm": 1.2174370288848877,
|
| 686 |
+
"learning_rate": 0.0011009174311926607,
|
| 687 |
+
"loss": 0.5654,
|
| 688 |
+
"step": 850
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.85,
|
| 692 |
+
"grad_norm": 0.8696100115776062,
|
| 693 |
+
"learning_rate": 0.0010550458715596332,
|
| 694 |
+
"loss": 0.6165,
|
| 695 |
+
"step": 860
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.95,
|
| 699 |
+
"grad_norm": 1.2088637351989746,
|
| 700 |
+
"learning_rate": 0.0010091743119266055,
|
| 701 |
+
"loss": 0.5639,
|
| 702 |
+
"step": 870
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 8.0,
|
| 706 |
+
"eval_accuracy": 0.7836490528414756,
|
| 707 |
+
"eval_f1": 0.5930825856311631,
|
| 708 |
+
"eval_loss": 0.5751714706420898,
|
| 709 |
+
"eval_precision": 0.601839864113542,
|
| 710 |
+
"eval_recall": 0.59116309877434,
|
| 711 |
+
"eval_runtime": 11.2311,
|
| 712 |
+
"eval_samples_per_second": 89.305,
|
| 713 |
+
"eval_steps_per_second": 5.609,
|
| 714 |
+
"step": 876
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 8.04,
|
| 718 |
+
"grad_norm": 0.6919325590133667,
|
| 719 |
+
"learning_rate": 0.0009633027522935781,
|
| 720 |
+
"loss": 0.6158,
|
| 721 |
+
"step": 880
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 8.13,
|
| 725 |
+
"grad_norm": 1.1032708883285522,
|
| 726 |
+
"learning_rate": 0.0009174311926605505,
|
| 727 |
+
"loss": 0.4922,
|
| 728 |
+
"step": 890
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 8.22,
|
| 732 |
+
"grad_norm": 1.5484873056411743,
|
| 733 |
+
"learning_rate": 0.000871559633027523,
|
| 734 |
+
"loss": 0.5453,
|
| 735 |
+
"step": 900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 8.31,
|
| 739 |
+
"grad_norm": 1.1881980895996094,
|
| 740 |
+
"learning_rate": 0.0008256880733944954,
|
| 741 |
+
"loss": 0.6307,
|
| 742 |
+
"step": 910
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 8.4,
|
| 746 |
+
"grad_norm": 0.80064457654953,
|
| 747 |
+
"learning_rate": 0.000779816513761468,
|
| 748 |
+
"loss": 0.5129,
|
| 749 |
+
"step": 920
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 8.49,
|
| 753 |
+
"grad_norm": 0.7097495794296265,
|
| 754 |
+
"learning_rate": 0.0007339449541284405,
|
| 755 |
+
"loss": 0.6111,
|
| 756 |
+
"step": 930
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 8.58,
|
| 760 |
+
"grad_norm": 1.0539182424545288,
|
| 761 |
+
"learning_rate": 0.0006880733944954129,
|
| 762 |
+
"loss": 0.5803,
|
| 763 |
+
"step": 940
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 8.68,
|
| 767 |
+
"grad_norm": 1.1059874296188354,
|
| 768 |
+
"learning_rate": 0.0006422018348623854,
|
| 769 |
+
"loss": 0.6122,
|
| 770 |
+
"step": 950
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 8.77,
|
| 774 |
+
"grad_norm": 0.969431459903717,
|
| 775 |
+
"learning_rate": 0.0005963302752293578,
|
| 776 |
+
"loss": 0.5662,
|
| 777 |
+
"step": 960
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 8.86,
|
| 781 |
+
"grad_norm": 0.9566834568977356,
|
| 782 |
+
"learning_rate": 0.0005504587155963303,
|
| 783 |
+
"loss": 0.5317,
|
| 784 |
+
"step": 970
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 8.95,
|
| 788 |
+
"grad_norm": 1.040589690208435,
|
| 789 |
+
"learning_rate": 0.0005045871559633027,
|
| 790 |
+
"loss": 0.6012,
|
| 791 |
+
"step": 980
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 9.0,
|
| 795 |
+
"eval_accuracy": 0.7926221335992024,
|
| 796 |
+
"eval_f1": 0.6175857153722378,
|
| 797 |
+
"eval_loss": 0.5508496165275574,
|
| 798 |
+
"eval_precision": 0.6303209720740276,
|
| 799 |
+
"eval_recall": 0.6194980716058,
|
| 800 |
+
"eval_runtime": 11.1597,
|
| 801 |
+
"eval_samples_per_second": 89.877,
|
| 802 |
+
"eval_steps_per_second": 5.645,
|
| 803 |
+
"step": 985
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 9.04,
|
| 807 |
+
"grad_norm": 0.8962224721908569,
|
| 808 |
+
"learning_rate": 0.00045871559633027525,
|
| 809 |
+
"loss": 0.5106,
|
| 810 |
+
"step": 990
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 9.13,
|
| 814 |
+
"grad_norm": 0.9521830677986145,
|
| 815 |
+
"learning_rate": 0.0004128440366972477,
|
| 816 |
+
"loss": 0.5768,
|
| 817 |
+
"step": 1000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 9.22,
|
| 821 |
+
"grad_norm": 1.102037787437439,
|
| 822 |
+
"learning_rate": 0.00036697247706422024,
|
| 823 |
+
"loss": 0.5709,
|
| 824 |
+
"step": 1010
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 9.32,
|
| 828 |
+
"grad_norm": 1.3637118339538574,
|
| 829 |
+
"learning_rate": 0.0003211009174311927,
|
| 830 |
+
"loss": 0.4719,
|
| 831 |
+
"step": 1020
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 9.41,
|
| 835 |
+
"grad_norm": 0.8370681405067444,
|
| 836 |
+
"learning_rate": 0.00027522935779816516,
|
| 837 |
+
"loss": 0.5717,
|
| 838 |
+
"step": 1030
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 9.5,
|
| 842 |
+
"grad_norm": 1.0408118963241577,
|
| 843 |
+
"learning_rate": 0.00022935779816513763,
|
| 844 |
+
"loss": 0.5216,
|
| 845 |
+
"step": 1040
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 9.59,
|
| 849 |
+
"grad_norm": 1.2250603437423706,
|
| 850 |
+
"learning_rate": 0.00018348623853211012,
|
| 851 |
+
"loss": 0.5648,
|
| 852 |
+
"step": 1050
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 9.68,
|
| 856 |
+
"grad_norm": 0.9561938047409058,
|
| 857 |
+
"learning_rate": 0.00013761467889908258,
|
| 858 |
+
"loss": 0.5316,
|
| 859 |
+
"step": 1060
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 9.77,
|
| 863 |
+
"grad_norm": 1.2182472944259644,
|
| 864 |
+
"learning_rate": 9.174311926605506e-05,
|
| 865 |
+
"loss": 0.5773,
|
| 866 |
+
"step": 1070
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 9.86,
|
| 870 |
+
"grad_norm": 1.1085007190704346,
|
| 871 |
+
"learning_rate": 4.587155963302753e-05,
|
| 872 |
+
"loss": 0.5777,
|
| 873 |
+
"step": 1080
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 9.95,
|
| 877 |
+
"grad_norm": 0.9443553686141968,
|
| 878 |
+
"learning_rate": 0.0,
|
| 879 |
+
"loss": 0.5468,
|
| 880 |
+
"step": 1090
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 9.95,
|
| 884 |
+
"eval_accuracy": 0.7856430707876371,
|
| 885 |
+
"eval_f1": 0.6355311291077588,
|
| 886 |
+
"eval_loss": 0.566506028175354,
|
| 887 |
+
"eval_precision": 0.6470430666433826,
|
| 888 |
+
"eval_recall": 0.6287911299620902,
|
| 889 |
+
"eval_runtime": 11.1527,
|
| 890 |
+
"eval_samples_per_second": 89.934,
|
| 891 |
+
"eval_steps_per_second": 5.649,
|
| 892 |
+
"step": 1090
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 9.95,
|
| 896 |
+
"step": 1090,
|
| 897 |
+
"total_flos": 1.2358226368562393e+19,
|
| 898 |
+
"train_loss": 0.6880708248243419,
|
| 899 |
+
"train_runtime": 1791.2534,
|
| 900 |
+
"train_samples_per_second": 39.118,
|
| 901 |
+
"train_steps_per_second": 0.609
|
| 902 |
+
}
|
| 903 |
+
],
|
| 904 |
+
"logging_steps": 10,
|
| 905 |
+
"max_steps": 1090,
|
| 906 |
+
"num_input_tokens_seen": 0,
|
| 907 |
+
"num_train_epochs": 10,
|
| 908 |
+
"save_steps": 500,
|
| 909 |
+
"total_flos": 1.2358226368562393e+19,
|
| 910 |
+
"train_batch_size": 16,
|
| 911 |
+
"trial_name": null,
|
| 912 |
+
"trial_params": null
|
| 913 |
+
}
|