End of training
Browse files- README.md +2 -2
- all_results.json +15 -0
- eval_results.json +9 -0
- predict_results.txt +0 -0
- train_results.json +9 -0
- trainer_state.json +976 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-large](https://huggingface.co/vinai/phobert-large) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- F1: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-large](https://huggingface.co/vinai/phobert-large) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2752
|
| 22 |
+
- F1: 0.9361
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9360692350642099,
|
| 4 |
+
"eval_loss": 0.2751626968383789,
|
| 5 |
+
"eval_runtime": 29.5871,
|
| 6 |
+
"eval_samples": 3582,
|
| 7 |
+
"eval_samples_per_second": 121.066,
|
| 8 |
+
"eval_steps_per_second": 5.07,
|
| 9 |
+
"total_flos": 2.3084949799922688e+17,
|
| 10 |
+
"train_loss": 0.688100525762774,
|
| 11 |
+
"train_runtime": 30496.5649,
|
| 12 |
+
"train_samples": 24771,
|
| 13 |
+
"train_samples_per_second": 32.49,
|
| 14 |
+
"train_steps_per_second": 1.355
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9360692350642099,
|
| 4 |
+
"eval_loss": 0.2751626968383789,
|
| 5 |
+
"eval_runtime": 29.5871,
|
| 6 |
+
"eval_samples": 3582,
|
| 7 |
+
"eval_samples_per_second": 121.066,
|
| 8 |
+
"eval_steps_per_second": 5.07
|
| 9 |
+
}
|
predict_results.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 2.3084949799922688e+17,
|
| 4 |
+
"train_loss": 0.688100525762774,
|
| 5 |
+
"train_runtime": 30496.5649,
|
| 6 |
+
"train_samples": 24771,
|
| 7 |
+
"train_samples_per_second": 32.49,
|
| 8 |
+
"train_steps_per_second": 1.355
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,976 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9360692350642099,
|
| 3 |
+
"best_model_checkpoint": "/tmp/classification_phobertlarge/checkpoint-6198",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 41320,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.484027105517909,
|
| 13 |
+
"grad_norm": 107.85002136230469,
|
| 14 |
+
"learning_rate": 2.9636979670861567e-05,
|
| 15 |
+
"loss": 0.4143,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.968054211035818,
|
| 20 |
+
"grad_norm": 15.951617240905762,
|
| 21 |
+
"learning_rate": 2.927395934172314e-05,
|
| 22 |
+
"loss": 0.3094,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_f1": 0.9327191513121161,
|
| 28 |
+
"eval_loss": 0.22949737310409546,
|
| 29 |
+
"eval_runtime": 30.2358,
|
| 30 |
+
"eval_samples_per_second": 118.469,
|
| 31 |
+
"eval_steps_per_second": 4.961,
|
| 32 |
+
"step": 1033
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.452081316553727,
|
| 36 |
+
"grad_norm": 0.18993481993675232,
|
| 37 |
+
"learning_rate": 2.8910939012584705e-05,
|
| 38 |
+
"loss": 0.2335,
|
| 39 |
+
"step": 1500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.936108422071636,
|
| 43 |
+
"grad_norm": 2.2120251655578613,
|
| 44 |
+
"learning_rate": 2.8547918683446275e-05,
|
| 45 |
+
"loss": 0.2346,
|
| 46 |
+
"step": 2000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_f1": 0.9279731993299832,
|
| 51 |
+
"eval_loss": 0.2650923430919647,
|
| 52 |
+
"eval_runtime": 30.2762,
|
| 53 |
+
"eval_samples_per_second": 118.311,
|
| 54 |
+
"eval_steps_per_second": 4.954,
|
| 55 |
+
"step": 2066
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.420135527589545,
|
| 59 |
+
"grad_norm": 1.64552640914917,
|
| 60 |
+
"learning_rate": 2.818489835430784e-05,
|
| 61 |
+
"loss": 0.1957,
|
| 62 |
+
"step": 2500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.904162633107454,
|
| 66 |
+
"grad_norm": 1.3047701120376587,
|
| 67 |
+
"learning_rate": 2.7821878025169413e-05,
|
| 68 |
+
"loss": 0.2359,
|
| 69 |
+
"step": 3000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_f1": 0.9338358458961474,
|
| 74 |
+
"eval_loss": 0.2357342690229416,
|
| 75 |
+
"eval_runtime": 30.2251,
|
| 76 |
+
"eval_samples_per_second": 118.511,
|
| 77 |
+
"eval_steps_per_second": 4.963,
|
| 78 |
+
"step": 3099
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.388189738625363,
|
| 82 |
+
"grad_norm": 21.052520751953125,
|
| 83 |
+
"learning_rate": 2.745885769603098e-05,
|
| 84 |
+
"loss": 0.179,
|
| 85 |
+
"step": 3500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.872216844143272,
|
| 89 |
+
"grad_norm": 0.44238439202308655,
|
| 90 |
+
"learning_rate": 2.7095837366892545e-05,
|
| 91 |
+
"loss": 0.1734,
|
| 92 |
+
"step": 4000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.0,
|
| 96 |
+
"eval_f1": 0.9276940256839754,
|
| 97 |
+
"eval_loss": 0.2720405161380768,
|
| 98 |
+
"eval_runtime": 30.1857,
|
| 99 |
+
"eval_samples_per_second": 118.666,
|
| 100 |
+
"eval_steps_per_second": 4.969,
|
| 101 |
+
"step": 4132
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.356243949661181,
|
| 105 |
+
"grad_norm": 0.10865873098373413,
|
| 106 |
+
"learning_rate": 2.6732817037754114e-05,
|
| 107 |
+
"loss": 0.1593,
|
| 108 |
+
"step": 4500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.84027105517909,
|
| 112 |
+
"grad_norm": 35.33164978027344,
|
| 113 |
+
"learning_rate": 2.6369796708615684e-05,
|
| 114 |
+
"loss": 0.1551,
|
| 115 |
+
"step": 5000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 5.0,
|
| 119 |
+
"eval_f1": 0.9349525404801786,
|
| 120 |
+
"eval_loss": 0.29397937655448914,
|
| 121 |
+
"eval_runtime": 30.2085,
|
| 122 |
+
"eval_samples_per_second": 118.576,
|
| 123 |
+
"eval_steps_per_second": 4.965,
|
| 124 |
+
"step": 5165
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.3242981606969995,
|
| 128 |
+
"grad_norm": 0.6579780578613281,
|
| 129 |
+
"learning_rate": 2.6006776379477253e-05,
|
| 130 |
+
"loss": 0.1434,
|
| 131 |
+
"step": 5500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.808325266214908,
|
| 135 |
+
"grad_norm": 0.2094825804233551,
|
| 136 |
+
"learning_rate": 2.564375605033882e-05,
|
| 137 |
+
"loss": 0.1282,
|
| 138 |
+
"step": 6000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 6.0,
|
| 142 |
+
"eval_f1": 0.9360692350642099,
|
| 143 |
+
"eval_loss": 0.2751626968383789,
|
| 144 |
+
"eval_runtime": 30.1054,
|
| 145 |
+
"eval_samples_per_second": 118.982,
|
| 146 |
+
"eval_steps_per_second": 4.983,
|
| 147 |
+
"step": 6198
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 6.292352371732817,
|
| 151 |
+
"grad_norm": 0.12414942681789398,
|
| 152 |
+
"learning_rate": 2.5280735721200388e-05,
|
| 153 |
+
"loss": 0.1158,
|
| 154 |
+
"step": 6500
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.776379477250726,
|
| 158 |
+
"grad_norm": 4.030830383300781,
|
| 159 |
+
"learning_rate": 2.4917715392061957e-05,
|
| 160 |
+
"loss": 0.1208,
|
| 161 |
+
"step": 7000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 7.0,
|
| 165 |
+
"eval_f1": 0.9346733668341709,
|
| 166 |
+
"eval_loss": 0.3136898875236511,
|
| 167 |
+
"eval_runtime": 30.2127,
|
| 168 |
+
"eval_samples_per_second": 118.559,
|
| 169 |
+
"eval_steps_per_second": 4.965,
|
| 170 |
+
"step": 7231
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.260406582768635,
|
| 174 |
+
"grad_norm": 4.0914530754089355,
|
| 175 |
+
"learning_rate": 2.4554695062923523e-05,
|
| 176 |
+
"loss": 0.1127,
|
| 177 |
+
"step": 7500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.744433688286544,
|
| 181 |
+
"grad_norm": 0.10396777093410492,
|
| 182 |
+
"learning_rate": 2.4191674733785093e-05,
|
| 183 |
+
"loss": 0.1074,
|
| 184 |
+
"step": 8000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"eval_f1": 0.9355108877721943,
|
| 189 |
+
"eval_loss": 0.2966548800468445,
|
| 190 |
+
"eval_runtime": 30.2555,
|
| 191 |
+
"eval_samples_per_second": 118.392,
|
| 192 |
+
"eval_steps_per_second": 4.958,
|
| 193 |
+
"step": 8264
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.228460793804453,
|
| 197 |
+
"grad_norm": 0.22765442728996277,
|
| 198 |
+
"learning_rate": 2.382865440464666e-05,
|
| 199 |
+
"loss": 0.0993,
|
| 200 |
+
"step": 8500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.712487899322362,
|
| 204 |
+
"grad_norm": 0.5725598931312561,
|
| 205 |
+
"learning_rate": 2.346563407550823e-05,
|
| 206 |
+
"loss": 0.2719,
|
| 207 |
+
"step": 9000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 9.0,
|
| 211 |
+
"eval_f1": 0.6409826912339475,
|
| 212 |
+
"eval_loss": 0.8269146084785461,
|
| 213 |
+
"eval_runtime": 30.1812,
|
| 214 |
+
"eval_samples_per_second": 118.683,
|
| 215 |
+
"eval_steps_per_second": 4.97,
|
| 216 |
+
"step": 9297
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 9.196515004840272,
|
| 220 |
+
"grad_norm": 1.2485491037368774,
|
| 221 |
+
"learning_rate": 2.3102613746369797e-05,
|
| 222 |
+
"loss": 0.8488,
|
| 223 |
+
"step": 9500
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 9.68054211035818,
|
| 227 |
+
"grad_norm": 0.6287474036216736,
|
| 228 |
+
"learning_rate": 2.2739593417231366e-05,
|
| 229 |
+
"loss": 0.8456,
|
| 230 |
+
"step": 10000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 10.0,
|
| 234 |
+
"eval_f1": 0.6409826912339475,
|
| 235 |
+
"eval_loss": 0.8219632506370544,
|
| 236 |
+
"eval_runtime": 30.1694,
|
| 237 |
+
"eval_samples_per_second": 118.729,
|
| 238 |
+
"eval_steps_per_second": 4.972,
|
| 239 |
+
"step": 10330
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 10.164569215876089,
|
| 243 |
+
"grad_norm": 1.4201635122299194,
|
| 244 |
+
"learning_rate": 2.2376573088092936e-05,
|
| 245 |
+
"loss": 0.8446,
|
| 246 |
+
"step": 10500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 10.648596321393999,
|
| 250 |
+
"grad_norm": 1.115445613861084,
|
| 251 |
+
"learning_rate": 2.20135527589545e-05,
|
| 252 |
+
"loss": 0.84,
|
| 253 |
+
"step": 11000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 11.0,
|
| 257 |
+
"eval_f1": 0.6409826912339475,
|
| 258 |
+
"eval_loss": 0.8211880922317505,
|
| 259 |
+
"eval_runtime": 30.1617,
|
| 260 |
+
"eval_samples_per_second": 118.76,
|
| 261 |
+
"eval_steps_per_second": 4.973,
|
| 262 |
+
"step": 11363
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 11.132623426911907,
|
| 266 |
+
"grad_norm": 1.784400463104248,
|
| 267 |
+
"learning_rate": 2.165053242981607e-05,
|
| 268 |
+
"loss": 0.8435,
|
| 269 |
+
"step": 11500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 11.616650532429816,
|
| 273 |
+
"grad_norm": 1.7544752359390259,
|
| 274 |
+
"learning_rate": 2.1287512100677637e-05,
|
| 275 |
+
"loss": 0.841,
|
| 276 |
+
"step": 12000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 12.0,
|
| 280 |
+
"eval_f1": 0.6409826912339475,
|
| 281 |
+
"eval_loss": 0.8207370042800903,
|
| 282 |
+
"eval_runtime": 30.1494,
|
| 283 |
+
"eval_samples_per_second": 118.808,
|
| 284 |
+
"eval_steps_per_second": 4.975,
|
| 285 |
+
"step": 12396
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 12.100677637947726,
|
| 289 |
+
"grad_norm": 0.660882830619812,
|
| 290 |
+
"learning_rate": 2.092449177153921e-05,
|
| 291 |
+
"loss": 0.8421,
|
| 292 |
+
"step": 12500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 12.584704743465634,
|
| 296 |
+
"grad_norm": 0.5564029812812805,
|
| 297 |
+
"learning_rate": 2.0561471442400775e-05,
|
| 298 |
+
"loss": 0.8383,
|
| 299 |
+
"step": 13000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 13.0,
|
| 303 |
+
"eval_f1": 0.6409826912339475,
|
| 304 |
+
"eval_loss": 0.8198010325431824,
|
| 305 |
+
"eval_runtime": 30.2024,
|
| 306 |
+
"eval_samples_per_second": 118.6,
|
| 307 |
+
"eval_steps_per_second": 4.966,
|
| 308 |
+
"step": 13429
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 13.068731848983543,
|
| 312 |
+
"grad_norm": 0.45510706305503845,
|
| 313 |
+
"learning_rate": 2.019845111326234e-05,
|
| 314 |
+
"loss": 0.8494,
|
| 315 |
+
"step": 13500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 13.552758954501453,
|
| 319 |
+
"grad_norm": 0.9993030428886414,
|
| 320 |
+
"learning_rate": 1.983543078412391e-05,
|
| 321 |
+
"loss": 0.8371,
|
| 322 |
+
"step": 14000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 14.0,
|
| 326 |
+
"eval_f1": 0.6409826912339475,
|
| 327 |
+
"eval_loss": 0.8276057839393616,
|
| 328 |
+
"eval_runtime": 30.1498,
|
| 329 |
+
"eval_samples_per_second": 118.807,
|
| 330 |
+
"eval_steps_per_second": 4.975,
|
| 331 |
+
"step": 14462
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.036786060019361,
|
| 335 |
+
"grad_norm": 0.8151612281799316,
|
| 336 |
+
"learning_rate": 1.947241045498548e-05,
|
| 337 |
+
"loss": 0.8465,
|
| 338 |
+
"step": 14500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 14.52081316553727,
|
| 342 |
+
"grad_norm": 1.312504768371582,
|
| 343 |
+
"learning_rate": 1.910939012584705e-05,
|
| 344 |
+
"loss": 0.8486,
|
| 345 |
+
"step": 15000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 15.0,
|
| 349 |
+
"eval_f1": 0.6409826912339475,
|
| 350 |
+
"eval_loss": 0.8243775367736816,
|
| 351 |
+
"eval_runtime": 30.177,
|
| 352 |
+
"eval_samples_per_second": 118.7,
|
| 353 |
+
"eval_steps_per_second": 4.971,
|
| 354 |
+
"step": 15495
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 15.00484027105518,
|
| 358 |
+
"grad_norm": 1.09600031375885,
|
| 359 |
+
"learning_rate": 1.8746369796708615e-05,
|
| 360 |
+
"loss": 0.8345,
|
| 361 |
+
"step": 15500
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 15.488867376573088,
|
| 365 |
+
"grad_norm": 1.3698476552963257,
|
| 366 |
+
"learning_rate": 1.8383349467570184e-05,
|
| 367 |
+
"loss": 0.8365,
|
| 368 |
+
"step": 16000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 15.972894482090997,
|
| 372 |
+
"grad_norm": 0.5735417008399963,
|
| 373 |
+
"learning_rate": 1.8020329138431754e-05,
|
| 374 |
+
"loss": 0.844,
|
| 375 |
+
"step": 16500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 16.0,
|
| 379 |
+
"eval_f1": 0.6409826912339475,
|
| 380 |
+
"eval_loss": 0.8390949964523315,
|
| 381 |
+
"eval_runtime": 30.2033,
|
| 382 |
+
"eval_samples_per_second": 118.596,
|
| 383 |
+
"eval_steps_per_second": 4.966,
|
| 384 |
+
"step": 16528
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 16.456921587608907,
|
| 388 |
+
"grad_norm": 1.3362003564834595,
|
| 389 |
+
"learning_rate": 1.765730880929332e-05,
|
| 390 |
+
"loss": 0.8349,
|
| 391 |
+
"step": 17000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 16.940948693126813,
|
| 395 |
+
"grad_norm": 0.4842800796031952,
|
| 396 |
+
"learning_rate": 1.729428848015489e-05,
|
| 397 |
+
"loss": 0.837,
|
| 398 |
+
"step": 17500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 17.0,
|
| 402 |
+
"eval_f1": 0.6409826912339475,
|
| 403 |
+
"eval_loss": 0.8234853744506836,
|
| 404 |
+
"eval_runtime": 30.2033,
|
| 405 |
+
"eval_samples_per_second": 118.596,
|
| 406 |
+
"eval_steps_per_second": 4.966,
|
| 407 |
+
"step": 17561
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 17.424975798644724,
|
| 411 |
+
"grad_norm": 0.564506471157074,
|
| 412 |
+
"learning_rate": 1.6931268151016455e-05,
|
| 413 |
+
"loss": 0.8368,
|
| 414 |
+
"step": 18000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 17.909002904162634,
|
| 418 |
+
"grad_norm": 1.0698164701461792,
|
| 419 |
+
"learning_rate": 1.6568247821878027e-05,
|
| 420 |
+
"loss": 0.8438,
|
| 421 |
+
"step": 18500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 18.0,
|
| 425 |
+
"eval_f1": 0.6409826912339475,
|
| 426 |
+
"eval_loss": 0.8246738314628601,
|
| 427 |
+
"eval_runtime": 30.1891,
|
| 428 |
+
"eval_samples_per_second": 118.652,
|
| 429 |
+
"eval_steps_per_second": 4.969,
|
| 430 |
+
"step": 18594
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 18.393030009680544,
|
| 434 |
+
"grad_norm": 0.8105427026748657,
|
| 435 |
+
"learning_rate": 1.6205227492739593e-05,
|
| 436 |
+
"loss": 0.8367,
|
| 437 |
+
"step": 19000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 18.87705711519845,
|
| 441 |
+
"grad_norm": 0.630368173122406,
|
| 442 |
+
"learning_rate": 1.5842207163601163e-05,
|
| 443 |
+
"loss": 0.8418,
|
| 444 |
+
"step": 19500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 19.0,
|
| 448 |
+
"eval_f1": 0.6409826912339475,
|
| 449 |
+
"eval_loss": 0.8236768841743469,
|
| 450 |
+
"eval_runtime": 30.1711,
|
| 451 |
+
"eval_samples_per_second": 118.723,
|
| 452 |
+
"eval_steps_per_second": 4.972,
|
| 453 |
+
"step": 19627
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 19.36108422071636,
|
| 457 |
+
"grad_norm": 0.884172797203064,
|
| 458 |
+
"learning_rate": 1.5479186834462732e-05,
|
| 459 |
+
"loss": 0.8503,
|
| 460 |
+
"step": 20000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 19.845111326234267,
|
| 464 |
+
"grad_norm": 0.585003674030304,
|
| 465 |
+
"learning_rate": 1.51161665053243e-05,
|
| 466 |
+
"loss": 0.8384,
|
| 467 |
+
"step": 20500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 20.0,
|
| 471 |
+
"eval_f1": 0.6409826912339475,
|
| 472 |
+
"eval_loss": 0.8199198246002197,
|
| 473 |
+
"eval_runtime": 30.1335,
|
| 474 |
+
"eval_samples_per_second": 118.871,
|
| 475 |
+
"eval_steps_per_second": 4.978,
|
| 476 |
+
"step": 20660
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 20.329138431752177,
|
| 480 |
+
"grad_norm": 1.194004774093628,
|
| 481 |
+
"learning_rate": 1.4753146176185867e-05,
|
| 482 |
+
"loss": 0.845,
|
| 483 |
+
"step": 21000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 20.813165537270088,
|
| 487 |
+
"grad_norm": 0.378979355096817,
|
| 488 |
+
"learning_rate": 1.4390125847047436e-05,
|
| 489 |
+
"loss": 0.8387,
|
| 490 |
+
"step": 21500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 21.0,
|
| 494 |
+
"eval_f1": 0.6409826912339475,
|
| 495 |
+
"eval_loss": 0.8225679993629456,
|
| 496 |
+
"eval_runtime": 29.9919,
|
| 497 |
+
"eval_samples_per_second": 119.432,
|
| 498 |
+
"eval_steps_per_second": 5.001,
|
| 499 |
+
"step": 21693
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 21.297192642787998,
|
| 503 |
+
"grad_norm": 0.6615312099456787,
|
| 504 |
+
"learning_rate": 1.4027105517909002e-05,
|
| 505 |
+
"loss": 0.833,
|
| 506 |
+
"step": 22000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 21.781219748305904,
|
| 510 |
+
"grad_norm": 0.4334174394607544,
|
| 511 |
+
"learning_rate": 1.3664085188770572e-05,
|
| 512 |
+
"loss": 0.8478,
|
| 513 |
+
"step": 22500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 22.0,
|
| 517 |
+
"eval_f1": 0.6409826912339475,
|
| 518 |
+
"eval_loss": 0.8205086588859558,
|
| 519 |
+
"eval_runtime": 29.9624,
|
| 520 |
+
"eval_samples_per_second": 119.55,
|
| 521 |
+
"eval_steps_per_second": 5.006,
|
| 522 |
+
"step": 22726
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 22.265246853823815,
|
| 526 |
+
"grad_norm": 0.8907693028450012,
|
| 527 |
+
"learning_rate": 1.3301064859632139e-05,
|
| 528 |
+
"loss": 0.8442,
|
| 529 |
+
"step": 23000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 22.749273959341725,
|
| 533 |
+
"grad_norm": 0.5629915595054626,
|
| 534 |
+
"learning_rate": 1.2938044530493708e-05,
|
| 535 |
+
"loss": 0.8364,
|
| 536 |
+
"step": 23500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 23.0,
|
| 540 |
+
"eval_f1": 0.6409826912339475,
|
| 541 |
+
"eval_loss": 0.8259473443031311,
|
| 542 |
+
"eval_runtime": 29.9524,
|
| 543 |
+
"eval_samples_per_second": 119.59,
|
| 544 |
+
"eval_steps_per_second": 5.008,
|
| 545 |
+
"step": 23759
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 23.23330106485963,
|
| 549 |
+
"grad_norm": 0.47178810834884644,
|
| 550 |
+
"learning_rate": 1.2575024201355276e-05,
|
| 551 |
+
"loss": 0.83,
|
| 552 |
+
"step": 24000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 23.71732817037754,
|
| 556 |
+
"grad_norm": 0.5889131426811218,
|
| 557 |
+
"learning_rate": 1.2212003872216845e-05,
|
| 558 |
+
"loss": 0.8325,
|
| 559 |
+
"step": 24500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 24.0,
|
| 563 |
+
"eval_f1": 0.6409826912339475,
|
| 564 |
+
"eval_loss": 0.8245280385017395,
|
| 565 |
+
"eval_runtime": 29.9556,
|
| 566 |
+
"eval_samples_per_second": 119.577,
|
| 567 |
+
"eval_steps_per_second": 5.007,
|
| 568 |
+
"step": 24792
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 24.20135527589545,
|
| 572 |
+
"grad_norm": 0.8042486310005188,
|
| 573 |
+
"learning_rate": 1.1848983543078413e-05,
|
| 574 |
+
"loss": 0.8357,
|
| 575 |
+
"step": 25000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 24.68538238141336,
|
| 579 |
+
"grad_norm": 0.7427254319190979,
|
| 580 |
+
"learning_rate": 1.148596321393998e-05,
|
| 581 |
+
"loss": 0.8289,
|
| 582 |
+
"step": 25500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 25.0,
|
| 586 |
+
"eval_f1": 0.6409826912339475,
|
| 587 |
+
"eval_loss": 0.8248125314712524,
|
| 588 |
+
"eval_runtime": 30.0337,
|
| 589 |
+
"eval_samples_per_second": 119.266,
|
| 590 |
+
"eval_steps_per_second": 4.994,
|
| 591 |
+
"step": 25825
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 25.16940948693127,
|
| 595 |
+
"grad_norm": 1.4444160461425781,
|
| 596 |
+
"learning_rate": 1.1122942884801548e-05,
|
| 597 |
+
"loss": 0.8344,
|
| 598 |
+
"step": 26000
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 25.65343659244918,
|
| 602 |
+
"grad_norm": 1.6647661924362183,
|
| 603 |
+
"learning_rate": 1.0759922555663117e-05,
|
| 604 |
+
"loss": 0.8251,
|
| 605 |
+
"step": 26500
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 26.0,
|
| 609 |
+
"eval_f1": 0.6409826912339475,
|
| 610 |
+
"eval_loss": 0.8246968984603882,
|
| 611 |
+
"eval_runtime": 30.0343,
|
| 612 |
+
"eval_samples_per_second": 119.263,
|
| 613 |
+
"eval_steps_per_second": 4.994,
|
| 614 |
+
"step": 26858
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 26.137463697967085,
|
| 618 |
+
"grad_norm": 0.560504138469696,
|
| 619 |
+
"learning_rate": 1.0396902226524685e-05,
|
| 620 |
+
"loss": 0.8265,
|
| 621 |
+
"step": 27000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 26.621490803484996,
|
| 625 |
+
"grad_norm": 2.6784770488739014,
|
| 626 |
+
"learning_rate": 1.0033881897386254e-05,
|
| 627 |
+
"loss": 0.824,
|
| 628 |
+
"step": 27500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 27.0,
|
| 632 |
+
"eval_f1": 0.6409826912339475,
|
| 633 |
+
"eval_loss": 0.821439802646637,
|
| 634 |
+
"eval_runtime": 29.9327,
|
| 635 |
+
"eval_samples_per_second": 119.668,
|
| 636 |
+
"eval_steps_per_second": 5.011,
|
| 637 |
+
"step": 27891
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 27.105517909002906,
|
| 641 |
+
"grad_norm": 0.9694743156433105,
|
| 642 |
+
"learning_rate": 9.670861568247822e-06,
|
| 643 |
+
"loss": 0.8304,
|
| 644 |
+
"step": 28000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 27.589545014520812,
|
| 648 |
+
"grad_norm": 0.8298953771591187,
|
| 649 |
+
"learning_rate": 9.307841239109391e-06,
|
| 650 |
+
"loss": 0.8197,
|
| 651 |
+
"step": 28500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 28.0,
|
| 655 |
+
"eval_f1": 0.6409826912339475,
|
| 656 |
+
"eval_loss": 0.8281795382499695,
|
| 657 |
+
"eval_runtime": 29.9052,
|
| 658 |
+
"eval_samples_per_second": 119.778,
|
| 659 |
+
"eval_steps_per_second": 5.016,
|
| 660 |
+
"step": 28924
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 28.073572120038722,
|
| 664 |
+
"grad_norm": 1.530045986175537,
|
| 665 |
+
"learning_rate": 8.944820909970959e-06,
|
| 666 |
+
"loss": 0.8214,
|
| 667 |
+
"step": 29000
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 28.557599225556633,
|
| 671 |
+
"grad_norm": 1.6190696954727173,
|
| 672 |
+
"learning_rate": 8.581800580832526e-06,
|
| 673 |
+
"loss": 0.8241,
|
| 674 |
+
"step": 29500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 29.0,
|
| 678 |
+
"eval_f1": 0.6409826912339475,
|
| 679 |
+
"eval_loss": 0.8340330123901367,
|
| 680 |
+
"eval_runtime": 29.9993,
|
| 681 |
+
"eval_samples_per_second": 119.403,
|
| 682 |
+
"eval_steps_per_second": 5.0,
|
| 683 |
+
"step": 29957
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 29.04162633107454,
|
| 687 |
+
"grad_norm": 0.6487633585929871,
|
| 688 |
+
"learning_rate": 8.218780251694096e-06,
|
| 689 |
+
"loss": 0.8209,
|
| 690 |
+
"step": 30000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 29.52565343659245,
|
| 694 |
+
"grad_norm": 1.1463664770126343,
|
| 695 |
+
"learning_rate": 7.855759922555663e-06,
|
| 696 |
+
"loss": 0.8285,
|
| 697 |
+
"step": 30500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 30.0,
|
| 701 |
+
"eval_f1": 0.6409826912339475,
|
| 702 |
+
"eval_loss": 0.8359894156455994,
|
| 703 |
+
"eval_runtime": 29.9709,
|
| 704 |
+
"eval_samples_per_second": 119.516,
|
| 705 |
+
"eval_steps_per_second": 5.005,
|
| 706 |
+
"step": 30990
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 30.00968054211036,
|
| 710 |
+
"grad_norm": 1.2427114248275757,
|
| 711 |
+
"learning_rate": 7.492739593417232e-06,
|
| 712 |
+
"loss": 0.8103,
|
| 713 |
+
"step": 31000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 30.493707647628266,
|
| 717 |
+
"grad_norm": 1.2282146215438843,
|
| 718 |
+
"learning_rate": 7.129719264278799e-06,
|
| 719 |
+
"loss": 0.814,
|
| 720 |
+
"step": 31500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 30.977734753146176,
|
| 724 |
+
"grad_norm": 1.6168450117111206,
|
| 725 |
+
"learning_rate": 6.766698935140368e-06,
|
| 726 |
+
"loss": 0.8169,
|
| 727 |
+
"step": 32000
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 31.0,
|
| 731 |
+
"eval_f1": 0.6409826912339475,
|
| 732 |
+
"eval_loss": 0.8400696516036987,
|
| 733 |
+
"eval_runtime": 29.9824,
|
| 734 |
+
"eval_samples_per_second": 119.47,
|
| 735 |
+
"eval_steps_per_second": 5.003,
|
| 736 |
+
"step": 32023
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 31.461761858664087,
|
| 740 |
+
"grad_norm": 2.838101863861084,
|
| 741 |
+
"learning_rate": 6.403678606001936e-06,
|
| 742 |
+
"loss": 0.8113,
|
| 743 |
+
"step": 32500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 31.945788964181993,
|
| 747 |
+
"grad_norm": 1.7709537744522095,
|
| 748 |
+
"learning_rate": 6.040658276863505e-06,
|
| 749 |
+
"loss": 0.811,
|
| 750 |
+
"step": 33000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 32.0,
|
| 754 |
+
"eval_f1": 0.6409826912339475,
|
| 755 |
+
"eval_loss": 0.853431224822998,
|
| 756 |
+
"eval_runtime": 30.0224,
|
| 757 |
+
"eval_samples_per_second": 119.311,
|
| 758 |
+
"eval_steps_per_second": 4.996,
|
| 759 |
+
"step": 33056
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 32.4298160696999,
|
| 763 |
+
"grad_norm": 1.901810646057129,
|
| 764 |
+
"learning_rate": 5.677637947725073e-06,
|
| 765 |
+
"loss": 0.8153,
|
| 766 |
+
"step": 33500
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 32.91384317521781,
|
| 770 |
+
"grad_norm": 3.1245410442352295,
|
| 771 |
+
"learning_rate": 5.3146176185866415e-06,
|
| 772 |
+
"loss": 0.8056,
|
| 773 |
+
"step": 34000
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 33.0,
|
| 777 |
+
"eval_f1": 0.6409826912339475,
|
| 778 |
+
"eval_loss": 0.869020938873291,
|
| 779 |
+
"eval_runtime": 29.9807,
|
| 780 |
+
"eval_samples_per_second": 119.477,
|
| 781 |
+
"eval_steps_per_second": 5.003,
|
| 782 |
+
"step": 34089
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 33.397870280735724,
|
| 786 |
+
"grad_norm": 2.18399977684021,
|
| 787 |
+
"learning_rate": 4.951597289448209e-06,
|
| 788 |
+
"loss": 0.8193,
|
| 789 |
+
"step": 34500
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 33.88189738625363,
|
| 793 |
+
"grad_norm": 0.905960202217102,
|
| 794 |
+
"learning_rate": 4.588576960309778e-06,
|
| 795 |
+
"loss": 0.8023,
|
| 796 |
+
"step": 35000
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 34.0,
|
| 800 |
+
"eval_f1": 0.6409826912339475,
|
| 801 |
+
"eval_loss": 0.863983154296875,
|
| 802 |
+
"eval_runtime": 29.8457,
|
| 803 |
+
"eval_samples_per_second": 120.017,
|
| 804 |
+
"eval_steps_per_second": 5.026,
|
| 805 |
+
"step": 35122
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 34.36592449177154,
|
| 809 |
+
"grad_norm": 2.841273069381714,
|
| 810 |
+
"learning_rate": 4.225556631171346e-06,
|
| 811 |
+
"loss": 0.8067,
|
| 812 |
+
"step": 35500
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 34.84995159728945,
|
| 816 |
+
"grad_norm": 0.9258007407188416,
|
| 817 |
+
"learning_rate": 3.8625363020329145e-06,
|
| 818 |
+
"loss": 0.8146,
|
| 819 |
+
"step": 36000
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 35.0,
|
| 823 |
+
"eval_f1": 0.6409826912339475,
|
| 824 |
+
"eval_loss": 0.870430052280426,
|
| 825 |
+
"eval_runtime": 29.8816,
|
| 826 |
+
"eval_samples_per_second": 119.873,
|
| 827 |
+
"eval_steps_per_second": 5.02,
|
| 828 |
+
"step": 36155
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 35.33397870280736,
|
| 832 |
+
"grad_norm": 0.954750120639801,
|
| 833 |
+
"learning_rate": 3.499515972894482e-06,
|
| 834 |
+
"loss": 0.7988,
|
| 835 |
+
"step": 36500
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 35.81800580832527,
|
| 839 |
+
"grad_norm": 1.9225430488586426,
|
| 840 |
+
"learning_rate": 3.1364956437560505e-06,
|
| 841 |
+
"loss": 0.8079,
|
| 842 |
+
"step": 37000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 36.0,
|
| 846 |
+
"eval_f1": 0.6409826912339475,
|
| 847 |
+
"eval_loss": 0.8958960175514221,
|
| 848 |
+
"eval_runtime": 29.9806,
|
| 849 |
+
"eval_samples_per_second": 119.477,
|
| 850 |
+
"eval_steps_per_second": 5.003,
|
| 851 |
+
"step": 37188
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 36.30203291384318,
|
| 855 |
+
"grad_norm": 1.3240045309066772,
|
| 856 |
+
"learning_rate": 2.7734753146176185e-06,
|
| 857 |
+
"loss": 0.8054,
|
| 858 |
+
"step": 37500
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 36.78606001936109,
|
| 862 |
+
"grad_norm": 5.058154106140137,
|
| 863 |
+
"learning_rate": 2.410454985479187e-06,
|
| 864 |
+
"loss": 0.8081,
|
| 865 |
+
"step": 38000
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 37.0,
|
| 869 |
+
"eval_f1": 0.6409826912339475,
|
| 870 |
+
"eval_loss": 0.8801546692848206,
|
| 871 |
+
"eval_runtime": 29.8881,
|
| 872 |
+
"eval_samples_per_second": 119.847,
|
| 873 |
+
"eval_steps_per_second": 5.019,
|
| 874 |
+
"step": 38221
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 37.27008712487899,
|
| 878 |
+
"grad_norm": 2.5231587886810303,
|
| 879 |
+
"learning_rate": 2.047434656340755e-06,
|
| 880 |
+
"loss": 0.8066,
|
| 881 |
+
"step": 38500
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 37.7541142303969,
|
| 885 |
+
"grad_norm": 1.007765769958496,
|
| 886 |
+
"learning_rate": 1.6844143272023235e-06,
|
| 887 |
+
"loss": 0.8059,
|
| 888 |
+
"step": 39000
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 38.0,
|
| 892 |
+
"eval_f1": 0.6409826912339475,
|
| 893 |
+
"eval_loss": 0.890051543712616,
|
| 894 |
+
"eval_runtime": 29.9175,
|
| 895 |
+
"eval_samples_per_second": 119.729,
|
| 896 |
+
"eval_steps_per_second": 5.014,
|
| 897 |
+
"step": 39254
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 38.23814133591481,
|
| 901 |
+
"grad_norm": 1.9777470827102661,
|
| 902 |
+
"learning_rate": 1.3213939980638917e-06,
|
| 903 |
+
"loss": 0.8077,
|
| 904 |
+
"step": 39500
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 38.72216844143272,
|
| 908 |
+
"grad_norm": 2.1567304134368896,
|
| 909 |
+
"learning_rate": 9.5837366892546e-07,
|
| 910 |
+
"loss": 0.8045,
|
| 911 |
+
"step": 40000
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 39.0,
|
| 915 |
+
"eval_f1": 0.6409826912339475,
|
| 916 |
+
"eval_loss": 0.888158917427063,
|
| 917 |
+
"eval_runtime": 29.9571,
|
| 918 |
+
"eval_samples_per_second": 119.571,
|
| 919 |
+
"eval_steps_per_second": 5.007,
|
| 920 |
+
"step": 40287
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 39.20619554695063,
|
| 924 |
+
"grad_norm": 1.4379490613937378,
|
| 925 |
+
"learning_rate": 5.953533397870282e-07,
|
| 926 |
+
"loss": 0.7972,
|
| 927 |
+
"step": 40500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 39.690222652468535,
|
| 931 |
+
"grad_norm": 1.5997909307479858,
|
| 932 |
+
"learning_rate": 2.3233301064859634e-07,
|
| 933 |
+
"loss": 0.8024,
|
| 934 |
+
"step": 41000
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 40.0,
|
| 938 |
+
"eval_f1": 0.6409826912339475,
|
| 939 |
+
"eval_loss": 0.8918312788009644,
|
| 940 |
+
"eval_runtime": 29.6532,
|
| 941 |
+
"eval_samples_per_second": 120.796,
|
| 942 |
+
"eval_steps_per_second": 5.058,
|
| 943 |
+
"step": 41320
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 40.0,
|
| 947 |
+
"step": 41320,
|
| 948 |
+
"total_flos": 2.3084949799922688e+17,
|
| 949 |
+
"train_loss": 0.688100525762774,
|
| 950 |
+
"train_runtime": 30496.5649,
|
| 951 |
+
"train_samples_per_second": 32.49,
|
| 952 |
+
"train_steps_per_second": 1.355
|
| 953 |
+
}
|
| 954 |
+
],
|
| 955 |
+
"logging_steps": 500,
|
| 956 |
+
"max_steps": 41320,
|
| 957 |
+
"num_input_tokens_seen": 0,
|
| 958 |
+
"num_train_epochs": 40,
|
| 959 |
+
"save_steps": 500,
|
| 960 |
+
"stateful_callbacks": {
|
| 961 |
+
"TrainerControl": {
|
| 962 |
+
"args": {
|
| 963 |
+
"should_epoch_stop": false,
|
| 964 |
+
"should_evaluate": false,
|
| 965 |
+
"should_log": false,
|
| 966 |
+
"should_save": true,
|
| 967 |
+
"should_training_stop": true
|
| 968 |
+
},
|
| 969 |
+
"attributes": {}
|
| 970 |
+
}
|
| 971 |
+
},
|
| 972 |
+
"total_flos": 2.3084949799922688e+17,
|
| 973 |
+
"train_batch_size": 24,
|
| 974 |
+
"trial_name": null,
|
| 975 |
+
"trial_params": null
|
| 976 |
+
}
|