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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) 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 [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the medmnist-v2 dataset.
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
+
- Loss: 0.6135
|
| 27 |
+
- Accuracy: 0.7681
|
| 28 |
+
- Precision: 0.6821
|
| 29 |
+
- Recall: 0.5061
|
| 30 |
+
- F1: 0.5497
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_accuracy": 0.7680798004987531,
|
| 4 |
+
"eval_f1": 0.5497441003867443,
|
| 5 |
+
"eval_loss": 0.6134716868400574,
|
| 6 |
+
"eval_precision": 0.6820690374636337,
|
| 7 |
+
"eval_recall": 0.5060769324556328,
|
| 8 |
+
"eval_runtime": 10.818,
|
| 9 |
+
"eval_samples_per_second": 185.339,
|
| 10 |
+
"eval_steps_per_second": 11.647,
|
| 11 |
+
"total_flos": 5.442882169274339e+18,
|
| 12 |
+
"train_loss": 0.7558124568484245,
|
| 13 |
+
"train_runtime": 825.7866,
|
| 14 |
+
"train_samples_per_second": 84.852,
|
| 15 |
+
"train_steps_per_second": 1.32
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"eval_accuracy": 0.7680798004987531,
|
| 4 |
+
"eval_f1": 0.5497441003867443,
|
| 5 |
+
"eval_loss": 0.6134716868400574,
|
| 6 |
+
"eval_precision": 0.6820690374636337,
|
| 7 |
+
"eval_recall": 0.5060769324556328,
|
| 8 |
+
"eval_runtime": 10.818,
|
| 9 |
+
"eval_samples_per_second": 185.339,
|
| 10 |
+
"eval_steps_per_second": 11.647
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.95,
|
| 3 |
+
"total_flos": 5.442882169274339e+18,
|
| 4 |
+
"train_loss": 0.7558124568484245,
|
| 5 |
+
"train_runtime": 825.7866,
|
| 6 |
+
"train_samples_per_second": 84.852,
|
| 7 |
+
"train_steps_per_second": 1.32
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,913 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7676969092721835,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-lora-medmnistv2/checkpoint-1090",
|
| 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": 1.0251981019973755,
|
| 14 |
+
"learning_rate": 0.004954128440366973,
|
| 15 |
+
"loss": 1.2326,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.18,
|
| 20 |
+
"grad_norm": 0.6157691478729248,
|
| 21 |
+
"learning_rate": 0.004908256880733945,
|
| 22 |
+
"loss": 0.9567,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.27,
|
| 27 |
+
"grad_norm": 1.101731538772583,
|
| 28 |
+
"learning_rate": 0.004862385321100918,
|
| 29 |
+
"loss": 0.905,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.37,
|
| 34 |
+
"grad_norm": 0.8456130623817444,
|
| 35 |
+
"learning_rate": 0.00481651376146789,
|
| 36 |
+
"loss": 0.9132,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.46,
|
| 41 |
+
"grad_norm": 0.6145790219306946,
|
| 42 |
+
"learning_rate": 0.0047706422018348625,
|
| 43 |
+
"loss": 0.8798,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.55,
|
| 48 |
+
"grad_norm": 1.1609084606170654,
|
| 49 |
+
"learning_rate": 0.004724770642201835,
|
| 50 |
+
"loss": 0.7849,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.64,
|
| 55 |
+
"grad_norm": 0.7732934355735779,
|
| 56 |
+
"learning_rate": 0.004678899082568808,
|
| 57 |
+
"loss": 0.7901,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.73,
|
| 62 |
+
"grad_norm": 0.7393956780433655,
|
| 63 |
+
"learning_rate": 0.00463302752293578,
|
| 64 |
+
"loss": 0.7932,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.82,
|
| 69 |
+
"grad_norm": 0.667579174041748,
|
| 70 |
+
"learning_rate": 0.0045871559633027525,
|
| 71 |
+
"loss": 0.7811,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.91,
|
| 76 |
+
"grad_norm": 0.8475561141967773,
|
| 77 |
+
"learning_rate": 0.004541284403669725,
|
| 78 |
+
"loss": 0.7579,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.0,
|
| 83 |
+
"eval_accuracy": 0.7427716849451645,
|
| 84 |
+
"eval_f1": 0.3926656517433595,
|
| 85 |
+
"eval_loss": 0.7045032978057861,
|
| 86 |
+
"eval_precision": 0.5204175676877641,
|
| 87 |
+
"eval_recall": 0.3710290953850673,
|
| 88 |
+
"eval_runtime": 5.2715,
|
| 89 |
+
"eval_samples_per_second": 190.268,
|
| 90 |
+
"eval_steps_per_second": 11.951,
|
| 91 |
+
"step": 109
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.0,
|
| 95 |
+
"grad_norm": 0.7959792613983154,
|
| 96 |
+
"learning_rate": 0.004495412844036698,
|
| 97 |
+
"loss": 0.8206,
|
| 98 |
+
"step": 110
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.1,
|
| 102 |
+
"grad_norm": 0.823697566986084,
|
| 103 |
+
"learning_rate": 0.0044495412844036695,
|
| 104 |
+
"loss": 0.7631,
|
| 105 |
+
"step": 120
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.19,
|
| 109 |
+
"grad_norm": 0.7514823079109192,
|
| 110 |
+
"learning_rate": 0.004403669724770643,
|
| 111 |
+
"loss": 0.7404,
|
| 112 |
+
"step": 130
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.28,
|
| 116 |
+
"grad_norm": 1.4910705089569092,
|
| 117 |
+
"learning_rate": 0.004357798165137615,
|
| 118 |
+
"loss": 0.7533,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.37,
|
| 123 |
+
"grad_norm": 1.1586723327636719,
|
| 124 |
+
"learning_rate": 0.004311926605504587,
|
| 125 |
+
"loss": 0.7781,
|
| 126 |
+
"step": 150
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.46,
|
| 130 |
+
"grad_norm": 1.106838583946228,
|
| 131 |
+
"learning_rate": 0.0042660550458715595,
|
| 132 |
+
"loss": 0.7337,
|
| 133 |
+
"step": 160
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.55,
|
| 137 |
+
"grad_norm": 1.4145853519439697,
|
| 138 |
+
"learning_rate": 0.004220183486238533,
|
| 139 |
+
"loss": 0.7871,
|
| 140 |
+
"step": 170
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.64,
|
| 144 |
+
"grad_norm": 1.0210392475128174,
|
| 145 |
+
"learning_rate": 0.004174311926605505,
|
| 146 |
+
"loss": 0.827,
|
| 147 |
+
"step": 180
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.74,
|
| 151 |
+
"grad_norm": 1.4612568616867065,
|
| 152 |
+
"learning_rate": 0.004128440366972477,
|
| 153 |
+
"loss": 0.8952,
|
| 154 |
+
"step": 190
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.83,
|
| 158 |
+
"grad_norm": 2.333279848098755,
|
| 159 |
+
"learning_rate": 0.00408256880733945,
|
| 160 |
+
"loss": 0.8019,
|
| 161 |
+
"step": 200
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.92,
|
| 165 |
+
"grad_norm": 1.9485087394714355,
|
| 166 |
+
"learning_rate": 0.004036697247706422,
|
| 167 |
+
"loss": 0.7689,
|
| 168 |
+
"step": 210
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.0,
|
| 172 |
+
"eval_accuracy": 0.7278165503489531,
|
| 173 |
+
"eval_f1": 0.3573330762987477,
|
| 174 |
+
"eval_loss": 0.75123530626297,
|
| 175 |
+
"eval_precision": 0.3964031496153109,
|
| 176 |
+
"eval_recall": 0.35266989798605725,
|
| 177 |
+
"eval_runtime": 5.3783,
|
| 178 |
+
"eval_samples_per_second": 186.49,
|
| 179 |
+
"eval_steps_per_second": 11.714,
|
| 180 |
+
"step": 219
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.01,
|
| 184 |
+
"grad_norm": 1.6179887056350708,
|
| 185 |
+
"learning_rate": 0.003990825688073394,
|
| 186 |
+
"loss": 0.8527,
|
| 187 |
+
"step": 220
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.1,
|
| 191 |
+
"grad_norm": 1.0684826374053955,
|
| 192 |
+
"learning_rate": 0.00394954128440367,
|
| 193 |
+
"loss": 0.8174,
|
| 194 |
+
"step": 230
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.19,
|
| 198 |
+
"grad_norm": 0.6670346260070801,
|
| 199 |
+
"learning_rate": 0.003903669724770642,
|
| 200 |
+
"loss": 0.7894,
|
| 201 |
+
"step": 240
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.28,
|
| 205 |
+
"grad_norm": 1.4650906324386597,
|
| 206 |
+
"learning_rate": 0.003857798165137615,
|
| 207 |
+
"loss": 0.7712,
|
| 208 |
+
"step": 250
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.37,
|
| 212 |
+
"grad_norm": 1.9125275611877441,
|
| 213 |
+
"learning_rate": 0.003811926605504587,
|
| 214 |
+
"loss": 0.8228,
|
| 215 |
+
"step": 260
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.47,
|
| 219 |
+
"grad_norm": 2.175058603286743,
|
| 220 |
+
"learning_rate": 0.0037706422018348625,
|
| 221 |
+
"loss": 0.7592,
|
| 222 |
+
"step": 270
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.56,
|
| 226 |
+
"grad_norm": 1.2160431146621704,
|
| 227 |
+
"learning_rate": 0.003724770642201835,
|
| 228 |
+
"loss": 0.7986,
|
| 229 |
+
"step": 280
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.65,
|
| 233 |
+
"grad_norm": 0.9580293893814087,
|
| 234 |
+
"learning_rate": 0.0036788990825688075,
|
| 235 |
+
"loss": 0.7767,
|
| 236 |
+
"step": 290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.74,
|
| 240 |
+
"grad_norm": 1.544980525970459,
|
| 241 |
+
"learning_rate": 0.0036330275229357802,
|
| 242 |
+
"loss": 0.7882,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.83,
|
| 247 |
+
"grad_norm": 2.8008100986480713,
|
| 248 |
+
"learning_rate": 0.003587155963302752,
|
| 249 |
+
"loss": 0.8818,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.92,
|
| 254 |
+
"grad_norm": 3.061676025390625,
|
| 255 |
+
"learning_rate": 0.003541284403669725,
|
| 256 |
+
"loss": 0.7353,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"eval_accuracy": 0.7357926221335992,
|
| 262 |
+
"eval_f1": 0.40020878247545494,
|
| 263 |
+
"eval_loss": 0.7191066145896912,
|
| 264 |
+
"eval_precision": 0.4630387632736769,
|
| 265 |
+
"eval_recall": 0.4201909974743698,
|
| 266 |
+
"eval_runtime": 5.4433,
|
| 267 |
+
"eval_samples_per_second": 184.263,
|
| 268 |
+
"eval_steps_per_second": 11.574,
|
| 269 |
+
"step": 328
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.01,
|
| 273 |
+
"grad_norm": 6.1841206550598145,
|
| 274 |
+
"learning_rate": 0.003495412844036697,
|
| 275 |
+
"loss": 0.762,
|
| 276 |
+
"step": 330
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 3.11,
|
| 280 |
+
"grad_norm": 2.103243589401245,
|
| 281 |
+
"learning_rate": 0.00344954128440367,
|
| 282 |
+
"loss": 0.798,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 3.2,
|
| 287 |
+
"grad_norm": 2.337041139602661,
|
| 288 |
+
"learning_rate": 0.0034082568807339447,
|
| 289 |
+
"loss": 0.7817,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 3.29,
|
| 294 |
+
"grad_norm": 3.2973079681396484,
|
| 295 |
+
"learning_rate": 0.0033669724770642204,
|
| 296 |
+
"loss": 0.7669,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.38,
|
| 301 |
+
"grad_norm": 1.7525138854980469,
|
| 302 |
+
"learning_rate": 0.003321100917431193,
|
| 303 |
+
"loss": 0.8788,
|
| 304 |
+
"step": 370
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.47,
|
| 308 |
+
"grad_norm": 3.467615842819214,
|
| 309 |
+
"learning_rate": 0.003275229357798165,
|
| 310 |
+
"loss": 0.9385,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.56,
|
| 315 |
+
"grad_norm": 2.560269355773926,
|
| 316 |
+
"learning_rate": 0.0032293577981651377,
|
| 317 |
+
"loss": 0.9773,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.65,
|
| 322 |
+
"grad_norm": 0.9047974348068237,
|
| 323 |
+
"learning_rate": 0.00318348623853211,
|
| 324 |
+
"loss": 0.948,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.74,
|
| 329 |
+
"grad_norm": 0.7493894100189209,
|
| 330 |
+
"learning_rate": 0.0031376146788990827,
|
| 331 |
+
"loss": 0.9429,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.84,
|
| 336 |
+
"grad_norm": 0.787327229976654,
|
| 337 |
+
"learning_rate": 0.003091743119266055,
|
| 338 |
+
"loss": 0.9193,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.93,
|
| 343 |
+
"grad_norm": 0.5133536458015442,
|
| 344 |
+
"learning_rate": 0.0030458715596330278,
|
| 345 |
+
"loss": 0.8429,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 4.0,
|
| 350 |
+
"eval_accuracy": 0.6809571286141576,
|
| 351 |
+
"eval_f1": 0.18506746606597574,
|
| 352 |
+
"eval_loss": 0.7857978940010071,
|
| 353 |
+
"eval_precision": 0.4280081257540274,
|
| 354 |
+
"eval_recall": 0.1812976208760752,
|
| 355 |
+
"eval_runtime": 5.4671,
|
| 356 |
+
"eval_samples_per_second": 183.46,
|
| 357 |
+
"eval_steps_per_second": 11.523,
|
| 358 |
+
"step": 438
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 4.02,
|
| 362 |
+
"grad_norm": 1.170052170753479,
|
| 363 |
+
"learning_rate": 0.003,
|
| 364 |
+
"loss": 0.7773,
|
| 365 |
+
"step": 440
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.11,
|
| 369 |
+
"grad_norm": 1.006057858467102,
|
| 370 |
+
"learning_rate": 0.002954128440366973,
|
| 371 |
+
"loss": 0.8049,
|
| 372 |
+
"step": 450
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 4.2,
|
| 376 |
+
"grad_norm": 0.5809192657470703,
|
| 377 |
+
"learning_rate": 0.002908256880733945,
|
| 378 |
+
"loss": 0.81,
|
| 379 |
+
"step": 460
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 4.29,
|
| 383 |
+
"grad_norm": 0.7377546429634094,
|
| 384 |
+
"learning_rate": 0.002862385321100918,
|
| 385 |
+
"loss": 0.7943,
|
| 386 |
+
"step": 470
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.38,
|
| 390 |
+
"grad_norm": 0.5214153528213501,
|
| 391 |
+
"learning_rate": 0.0028165137614678897,
|
| 392 |
+
"loss": 0.7664,
|
| 393 |
+
"step": 480
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.47,
|
| 397 |
+
"grad_norm": 0.615777313709259,
|
| 398 |
+
"learning_rate": 0.0027706422018348624,
|
| 399 |
+
"loss": 0.7769,
|
| 400 |
+
"step": 490
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.57,
|
| 404 |
+
"grad_norm": 0.5365859866142273,
|
| 405 |
+
"learning_rate": 0.0027247706422018347,
|
| 406 |
+
"loss": 0.8232,
|
| 407 |
+
"step": 500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.66,
|
| 411 |
+
"grad_norm": 0.7110781669616699,
|
| 412 |
+
"learning_rate": 0.0026788990825688075,
|
| 413 |
+
"loss": 0.7755,
|
| 414 |
+
"step": 510
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.75,
|
| 418 |
+
"grad_norm": 0.8415727615356445,
|
| 419 |
+
"learning_rate": 0.0026330275229357798,
|
| 420 |
+
"loss": 0.8154,
|
| 421 |
+
"step": 520
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 4.84,
|
| 425 |
+
"grad_norm": 0.7670438289642334,
|
| 426 |
+
"learning_rate": 0.0025871559633027525,
|
| 427 |
+
"loss": 0.7386,
|
| 428 |
+
"step": 530
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.93,
|
| 432 |
+
"grad_norm": 0.8495368957519531,
|
| 433 |
+
"learning_rate": 0.002541284403669725,
|
| 434 |
+
"loss": 0.7929,
|
| 435 |
+
"step": 540
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 5.0,
|
| 439 |
+
"eval_accuracy": 0.7218344965104686,
|
| 440 |
+
"eval_f1": 0.35225874228381837,
|
| 441 |
+
"eval_loss": 0.7013418674468994,
|
| 442 |
+
"eval_precision": 0.5157907633577941,
|
| 443 |
+
"eval_recall": 0.397101205649215,
|
| 444 |
+
"eval_runtime": 5.3335,
|
| 445 |
+
"eval_samples_per_second": 188.058,
|
| 446 |
+
"eval_steps_per_second": 11.812,
|
| 447 |
+
"step": 547
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 5.02,
|
| 451 |
+
"grad_norm": 0.9542437791824341,
|
| 452 |
+
"learning_rate": 0.0024954128440366975,
|
| 453 |
+
"loss": 0.75,
|
| 454 |
+
"step": 550
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 5.11,
|
| 458 |
+
"grad_norm": 0.7318850755691528,
|
| 459 |
+
"learning_rate": 0.00244954128440367,
|
| 460 |
+
"loss": 0.7665,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 5.21,
|
| 465 |
+
"grad_norm": 1.0321910381317139,
|
| 466 |
+
"learning_rate": 0.0024036697247706426,
|
| 467 |
+
"loss": 0.7015,
|
| 468 |
+
"step": 570
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 5.3,
|
| 472 |
+
"grad_norm": 0.786632239818573,
|
| 473 |
+
"learning_rate": 0.002357798165137615,
|
| 474 |
+
"loss": 0.7756,
|
| 475 |
+
"step": 580
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 5.39,
|
| 479 |
+
"grad_norm": 0.8995092511177063,
|
| 480 |
+
"learning_rate": 0.002311926605504587,
|
| 481 |
+
"loss": 0.79,
|
| 482 |
+
"step": 590
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 5.48,
|
| 486 |
+
"grad_norm": 0.7932329773902893,
|
| 487 |
+
"learning_rate": 0.00226605504587156,
|
| 488 |
+
"loss": 0.7069,
|
| 489 |
+
"step": 600
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 5.57,
|
| 493 |
+
"grad_norm": 0.7889924645423889,
|
| 494 |
+
"learning_rate": 0.002220183486238532,
|
| 495 |
+
"loss": 0.7507,
|
| 496 |
+
"step": 610
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 5.66,
|
| 500 |
+
"grad_norm": 0.8880289793014526,
|
| 501 |
+
"learning_rate": 0.002174311926605505,
|
| 502 |
+
"loss": 0.7341,
|
| 503 |
+
"step": 620
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.75,
|
| 507 |
+
"grad_norm": 0.5282242298126221,
|
| 508 |
+
"learning_rate": 0.0021284403669724773,
|
| 509 |
+
"loss": 0.685,
|
| 510 |
+
"step": 630
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 5.84,
|
| 514 |
+
"grad_norm": 1.16853666305542,
|
| 515 |
+
"learning_rate": 0.0020825688073394496,
|
| 516 |
+
"loss": 0.7149,
|
| 517 |
+
"step": 640
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 5.94,
|
| 521 |
+
"grad_norm": 0.6140002012252808,
|
| 522 |
+
"learning_rate": 0.0020366972477064223,
|
| 523 |
+
"loss": 0.6804,
|
| 524 |
+
"step": 650
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 6.0,
|
| 528 |
+
"eval_accuracy": 0.7607178464606181,
|
| 529 |
+
"eval_f1": 0.43914392933644575,
|
| 530 |
+
"eval_loss": 0.6822256445884705,
|
| 531 |
+
"eval_precision": 0.5010781993513719,
|
| 532 |
+
"eval_recall": 0.4239506721942319,
|
| 533 |
+
"eval_runtime": 5.2475,
|
| 534 |
+
"eval_samples_per_second": 191.137,
|
| 535 |
+
"eval_steps_per_second": 12.006,
|
| 536 |
+
"step": 657
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 6.03,
|
| 540 |
+
"grad_norm": 0.7350448369979858,
|
| 541 |
+
"learning_rate": 0.0019908256880733946,
|
| 542 |
+
"loss": 0.7111,
|
| 543 |
+
"step": 660
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 6.12,
|
| 547 |
+
"grad_norm": 0.6007147431373596,
|
| 548 |
+
"learning_rate": 0.001944954128440367,
|
| 549 |
+
"loss": 0.6981,
|
| 550 |
+
"step": 670
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 6.21,
|
| 554 |
+
"grad_norm": 0.8717228770256042,
|
| 555 |
+
"learning_rate": 0.0018990825688073396,
|
| 556 |
+
"loss": 0.666,
|
| 557 |
+
"step": 680
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 6.3,
|
| 561 |
+
"grad_norm": 0.5795207023620605,
|
| 562 |
+
"learning_rate": 0.0018532110091743121,
|
| 563 |
+
"loss": 0.7206,
|
| 564 |
+
"step": 690
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 6.39,
|
| 568 |
+
"grad_norm": 0.9550521373748779,
|
| 569 |
+
"learning_rate": 0.0018073394495412844,
|
| 570 |
+
"loss": 0.7388,
|
| 571 |
+
"step": 700
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 6.48,
|
| 575 |
+
"grad_norm": 1.932844877243042,
|
| 576 |
+
"learning_rate": 0.001761467889908257,
|
| 577 |
+
"loss": 0.7198,
|
| 578 |
+
"step": 710
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 6.58,
|
| 582 |
+
"grad_norm": 0.8446419835090637,
|
| 583 |
+
"learning_rate": 0.0017155963302752295,
|
| 584 |
+
"loss": 0.6446,
|
| 585 |
+
"step": 720
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.67,
|
| 589 |
+
"grad_norm": 1.2699838876724243,
|
| 590 |
+
"learning_rate": 0.001669724770642202,
|
| 591 |
+
"loss": 0.683,
|
| 592 |
+
"step": 730
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 6.76,
|
| 596 |
+
"grad_norm": 0.8056779503822327,
|
| 597 |
+
"learning_rate": 0.0016238532110091745,
|
| 598 |
+
"loss": 0.7272,
|
| 599 |
+
"step": 740
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 6.85,
|
| 603 |
+
"grad_norm": 0.8416026830673218,
|
| 604 |
+
"learning_rate": 0.0015779816513761468,
|
| 605 |
+
"loss": 0.7436,
|
| 606 |
+
"step": 750
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 6.94,
|
| 610 |
+
"grad_norm": 0.8119596242904663,
|
| 611 |
+
"learning_rate": 0.0015321100917431193,
|
| 612 |
+
"loss": 0.6922,
|
| 613 |
+
"step": 760
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 7.0,
|
| 617 |
+
"eval_accuracy": 0.7666999002991027,
|
| 618 |
+
"eval_f1": 0.5226745508907684,
|
| 619 |
+
"eval_loss": 0.653312087059021,
|
| 620 |
+
"eval_precision": 0.6761986767002692,
|
| 621 |
+
"eval_recall": 0.5105511763591389,
|
| 622 |
+
"eval_runtime": 5.3357,
|
| 623 |
+
"eval_samples_per_second": 187.978,
|
| 624 |
+
"eval_steps_per_second": 11.807,
|
| 625 |
+
"step": 766
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.03,
|
| 629 |
+
"grad_norm": 1.218758225440979,
|
| 630 |
+
"learning_rate": 0.0014862385321100919,
|
| 631 |
+
"loss": 0.7674,
|
| 632 |
+
"step": 770
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.12,
|
| 636 |
+
"grad_norm": 1.0076816082000732,
|
| 637 |
+
"learning_rate": 0.0014403669724770644,
|
| 638 |
+
"loss": 0.6783,
|
| 639 |
+
"step": 780
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 7.21,
|
| 643 |
+
"grad_norm": 1.2119933366775513,
|
| 644 |
+
"learning_rate": 0.0013944954128440369,
|
| 645 |
+
"loss": 0.6743,
|
| 646 |
+
"step": 790
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.31,
|
| 650 |
+
"grad_norm": 0.7832847237586975,
|
| 651 |
+
"learning_rate": 0.0013486238532110092,
|
| 652 |
+
"loss": 0.751,
|
| 653 |
+
"step": 800
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.4,
|
| 657 |
+
"grad_norm": 0.7728474140167236,
|
| 658 |
+
"learning_rate": 0.0013027522935779817,
|
| 659 |
+
"loss": 0.7252,
|
| 660 |
+
"step": 810
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 7.49,
|
| 664 |
+
"grad_norm": 1.1566990613937378,
|
| 665 |
+
"learning_rate": 0.0012568807339449542,
|
| 666 |
+
"loss": 0.6176,
|
| 667 |
+
"step": 820
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 7.58,
|
| 671 |
+
"grad_norm": 0.8185287117958069,
|
| 672 |
+
"learning_rate": 0.0012110091743119267,
|
| 673 |
+
"loss": 0.6928,
|
| 674 |
+
"step": 830
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.67,
|
| 678 |
+
"grad_norm": 1.3969968557357788,
|
| 679 |
+
"learning_rate": 0.0011651376146788993,
|
| 680 |
+
"loss": 0.6475,
|
| 681 |
+
"step": 840
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.76,
|
| 685 |
+
"grad_norm": 1.7943429946899414,
|
| 686 |
+
"learning_rate": 0.0011192660550458716,
|
| 687 |
+
"loss": 0.6414,
|
| 688 |
+
"step": 850
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.85,
|
| 692 |
+
"grad_norm": 1.6549348831176758,
|
| 693 |
+
"learning_rate": 0.001073394495412844,
|
| 694 |
+
"loss": 0.717,
|
| 695 |
+
"step": 860
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.95,
|
| 699 |
+
"grad_norm": 1.016310691833496,
|
| 700 |
+
"learning_rate": 0.0010275229357798166,
|
| 701 |
+
"loss": 0.6563,
|
| 702 |
+
"step": 870
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 8.0,
|
| 706 |
+
"eval_accuracy": 0.7467597208374875,
|
| 707 |
+
"eval_f1": 0.4496255647360247,
|
| 708 |
+
"eval_loss": 0.6757794618606567,
|
| 709 |
+
"eval_precision": 0.4547903051227771,
|
| 710 |
+
"eval_recall": 0.4588581671719845,
|
| 711 |
+
"eval_runtime": 5.4168,
|
| 712 |
+
"eval_samples_per_second": 185.164,
|
| 713 |
+
"eval_steps_per_second": 11.63,
|
| 714 |
+
"step": 876
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 8.04,
|
| 718 |
+
"grad_norm": 1.1468579769134521,
|
| 719 |
+
"learning_rate": 0.0009816513761467891,
|
| 720 |
+
"loss": 0.7141,
|
| 721 |
+
"step": 880
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 8.13,
|
| 725 |
+
"grad_norm": 3.9916632175445557,
|
| 726 |
+
"learning_rate": 0.0009357798165137615,
|
| 727 |
+
"loss": 0.6384,
|
| 728 |
+
"step": 890
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 8.22,
|
| 732 |
+
"grad_norm": 1.3290263414382935,
|
| 733 |
+
"learning_rate": 0.000889908256880734,
|
| 734 |
+
"loss": 0.6598,
|
| 735 |
+
"step": 900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 8.31,
|
| 739 |
+
"grad_norm": 2.3329484462738037,
|
| 740 |
+
"learning_rate": 0.0008440366972477065,
|
| 741 |
+
"loss": 0.744,
|
| 742 |
+
"step": 910
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 8.4,
|
| 746 |
+
"grad_norm": 1.5532264709472656,
|
| 747 |
+
"learning_rate": 0.000798165137614679,
|
| 748 |
+
"loss": 0.64,
|
| 749 |
+
"step": 920
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 8.49,
|
| 753 |
+
"grad_norm": 0.8567171692848206,
|
| 754 |
+
"learning_rate": 0.0007522935779816515,
|
| 755 |
+
"loss": 0.6803,
|
| 756 |
+
"step": 930
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 8.58,
|
| 760 |
+
"grad_norm": 0.8911744356155396,
|
| 761 |
+
"learning_rate": 0.0007064220183486239,
|
| 762 |
+
"loss": 0.6808,
|
| 763 |
+
"step": 940
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 8.68,
|
| 767 |
+
"grad_norm": 0.8479019999504089,
|
| 768 |
+
"learning_rate": 0.0006605504587155964,
|
| 769 |
+
"loss": 0.6848,
|
| 770 |
+
"step": 950
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 8.77,
|
| 774 |
+
"grad_norm": 1.6710033416748047,
|
| 775 |
+
"learning_rate": 0.0006146788990825688,
|
| 776 |
+
"loss": 0.6663,
|
| 777 |
+
"step": 960
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 8.86,
|
| 781 |
+
"grad_norm": 0.7447875142097473,
|
| 782 |
+
"learning_rate": 0.0005688073394495413,
|
| 783 |
+
"loss": 0.6389,
|
| 784 |
+
"step": 970
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 8.95,
|
| 788 |
+
"grad_norm": 1.6779547929763794,
|
| 789 |
+
"learning_rate": 0.0005229357798165138,
|
| 790 |
+
"loss": 0.6985,
|
| 791 |
+
"step": 980
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 9.0,
|
| 795 |
+
"eval_accuracy": 0.7647058823529411,
|
| 796 |
+
"eval_f1": 0.49146885132144036,
|
| 797 |
+
"eval_loss": 0.6264088749885559,
|
| 798 |
+
"eval_precision": 0.6450867892155376,
|
| 799 |
+
"eval_recall": 0.4691799634422585,
|
| 800 |
+
"eval_runtime": 5.2889,
|
| 801 |
+
"eval_samples_per_second": 189.641,
|
| 802 |
+
"eval_steps_per_second": 11.912,
|
| 803 |
+
"step": 985
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 9.04,
|
| 807 |
+
"grad_norm": 0.9829887747764587,
|
| 808 |
+
"learning_rate": 0.00047706422018348627,
|
| 809 |
+
"loss": 0.608,
|
| 810 |
+
"step": 990
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 9.13,
|
| 814 |
+
"grad_norm": 0.8272444009780884,
|
| 815 |
+
"learning_rate": 0.00043119266055045873,
|
| 816 |
+
"loss": 0.6661,
|
| 817 |
+
"step": 1000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 9.22,
|
| 821 |
+
"grad_norm": 1.056658387184143,
|
| 822 |
+
"learning_rate": 0.0003853211009174312,
|
| 823 |
+
"loss": 0.6982,
|
| 824 |
+
"step": 1010
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 9.32,
|
| 828 |
+
"grad_norm": 0.8954740166664124,
|
| 829 |
+
"learning_rate": 0.0003394495412844037,
|
| 830 |
+
"loss": 0.5824,
|
| 831 |
+
"step": 1020
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 9.41,
|
| 835 |
+
"grad_norm": 1.664057731628418,
|
| 836 |
+
"learning_rate": 0.0002935779816513762,
|
| 837 |
+
"loss": 0.657,
|
| 838 |
+
"step": 1030
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 9.5,
|
| 842 |
+
"grad_norm": 0.9140777587890625,
|
| 843 |
+
"learning_rate": 0.00024770642201834864,
|
| 844 |
+
"loss": 0.5975,
|
| 845 |
+
"step": 1040
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 9.59,
|
| 849 |
+
"grad_norm": 1.4085102081298828,
|
| 850 |
+
"learning_rate": 0.0002018348623853211,
|
| 851 |
+
"loss": 0.6377,
|
| 852 |
+
"step": 1050
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 9.68,
|
| 856 |
+
"grad_norm": 0.9538210034370422,
|
| 857 |
+
"learning_rate": 0.0001559633027522936,
|
| 858 |
+
"loss": 0.6508,
|
| 859 |
+
"step": 1060
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 9.77,
|
| 863 |
+
"grad_norm": 1.5182217359542847,
|
| 864 |
+
"learning_rate": 0.00011009174311926606,
|
| 865 |
+
"loss": 0.6776,
|
| 866 |
+
"step": 1070
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 9.86,
|
| 870 |
+
"grad_norm": 2.307227849960327,
|
| 871 |
+
"learning_rate": 6.422018348623854e-05,
|
| 872 |
+
"loss": 0.6758,
|
| 873 |
+
"step": 1080
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 9.95,
|
| 877 |
+
"grad_norm": 0.9896465539932251,
|
| 878 |
+
"learning_rate": 1.834862385321101e-05,
|
| 879 |
+
"loss": 0.6283,
|
| 880 |
+
"step": 1090
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 9.95,
|
| 884 |
+
"eval_accuracy": 0.7676969092721835,
|
| 885 |
+
"eval_f1": 0.50879281522548,
|
| 886 |
+
"eval_loss": 0.6179494857788086,
|
| 887 |
+
"eval_precision": 0.5889156507510112,
|
| 888 |
+
"eval_recall": 0.4796382542284182,
|
| 889 |
+
"eval_runtime": 5.4809,
|
| 890 |
+
"eval_samples_per_second": 182.998,
|
| 891 |
+
"eval_steps_per_second": 11.494,
|
| 892 |
+
"step": 1090
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 9.95,
|
| 896 |
+
"step": 1090,
|
| 897 |
+
"total_flos": 5.442882169274339e+18,
|
| 898 |
+
"train_loss": 0.7558124568484245,
|
| 899 |
+
"train_runtime": 825.7866,
|
| 900 |
+
"train_samples_per_second": 84.852,
|
| 901 |
+
"train_steps_per_second": 1.32
|
| 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": 5.442882169274339e+18,
|
| 910 |
+
"train_batch_size": 16,
|
| 911 |
+
"trial_name": null,
|
| 912 |
+
"trial_params": null
|
| 913 |
+
}
|