End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +1013 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: other
|
|
| 4 |
base_model: mistralai/Ministral-8B-Instruct-2410
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: Ministral-8B-Instruct-2410-PsyCourse-doc-info-fold8
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# Ministral-8B-Instruct-2410-PsyCourse-doc-info-fold8
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.0502
|
| 21 |
|
|
|
|
| 4 |
base_model: mistralai/Ministral-8B-Instruct-2410
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: Ministral-8B-Instruct-2410-PsyCourse-doc-info-fold8
|
|
|
|
| 16 |
|
| 17 |
# Ministral-8B-Instruct-2410-PsyCourse-doc-info-fold8
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) on the course-doc-info-train-fold8 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.0502
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.938271604938271,
|
| 3 |
+
"eval_loss": 0.05018164962530136,
|
| 4 |
+
"eval_runtime": 28.7907,
|
| 5 |
+
"eval_samples_per_second": 1.563,
|
| 6 |
+
"eval_steps_per_second": 1.563,
|
| 7 |
+
"total_flos": 1.743556468483031e+17,
|
| 8 |
+
"train_loss": 0.07543200919032098,
|
| 9 |
+
"train_runtime": 5142.0517,
|
| 10 |
+
"train_samples_per_second": 0.394,
|
| 11 |
+
"train_steps_per_second": 0.024
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.938271604938271,
|
| 3 |
+
"eval_loss": 0.05018164962530136,
|
| 4 |
+
"eval_runtime": 28.7907,
|
| 5 |
+
"eval_samples_per_second": 1.563,
|
| 6 |
+
"eval_steps_per_second": 1.563
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.938271604938271,
|
| 3 |
+
"total_flos": 1.743556468483031e+17,
|
| 4 |
+
"train_loss": 0.07543200919032098,
|
| 5 |
+
"train_runtime": 5142.0517,
|
| 6 |
+
"train_samples_per_second": 0.394,
|
| 7 |
+
"train_steps_per_second": 0.024
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1013 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.05018164962530136,
|
| 3 |
+
"best_model_checkpoint": "saves/psy-course-doc-info/Ministral-8B-Instruct-2410/train/fold8/checkpoint-110",
|
| 4 |
+
"epoch": 4.938271604938271,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 125,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03950617283950617,
|
| 13 |
+
"grad_norm": 0.6986701488494873,
|
| 14 |
+
"learning_rate": 7.692307692307694e-06,
|
| 15 |
+
"loss": 0.3512,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07901234567901234,
|
| 20 |
+
"grad_norm": 0.7088757753372192,
|
| 21 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 22 |
+
"loss": 0.3588,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.11851851851851852,
|
| 27 |
+
"grad_norm": 0.8064380884170532,
|
| 28 |
+
"learning_rate": 2.307692307692308e-05,
|
| 29 |
+
"loss": 0.3718,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1580246913580247,
|
| 34 |
+
"grad_norm": 0.7222112417221069,
|
| 35 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 36 |
+
"loss": 0.3595,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.19753086419753085,
|
| 41 |
+
"grad_norm": 0.6226270794868469,
|
| 42 |
+
"learning_rate": 3.846153846153846e-05,
|
| 43 |
+
"loss": 0.3219,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.23703703703703705,
|
| 48 |
+
"grad_norm": 0.6451139450073242,
|
| 49 |
+
"learning_rate": 4.615384615384616e-05,
|
| 50 |
+
"loss": 0.3117,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2765432098765432,
|
| 55 |
+
"grad_norm": 0.4552927315235138,
|
| 56 |
+
"learning_rate": 5.384615384615385e-05,
|
| 57 |
+
"loss": 0.2792,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3160493827160494,
|
| 62 |
+
"grad_norm": 0.4108279347419739,
|
| 63 |
+
"learning_rate": 6.153846153846155e-05,
|
| 64 |
+
"loss": 0.2616,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.35555555555555557,
|
| 69 |
+
"grad_norm": 0.49120235443115234,
|
| 70 |
+
"learning_rate": 6.923076923076924e-05,
|
| 71 |
+
"loss": 0.2405,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3950617283950617,
|
| 76 |
+
"grad_norm": 0.634235680103302,
|
| 77 |
+
"learning_rate": 7.692307692307693e-05,
|
| 78 |
+
"loss": 0.2346,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3950617283950617,
|
| 83 |
+
"eval_loss": 0.22451305389404297,
|
| 84 |
+
"eval_runtime": 29.4003,
|
| 85 |
+
"eval_samples_per_second": 1.531,
|
| 86 |
+
"eval_steps_per_second": 1.531,
|
| 87 |
+
"step": 10
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.4345679012345679,
|
| 91 |
+
"grad_norm": 0.6467899680137634,
|
| 92 |
+
"learning_rate": 8.461538461538461e-05,
|
| 93 |
+
"loss": 0.2449,
|
| 94 |
+
"step": 11
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.4740740740740741,
|
| 98 |
+
"grad_norm": 0.5100688338279724,
|
| 99 |
+
"learning_rate": 9.230769230769232e-05,
|
| 100 |
+
"loss": 0.213,
|
| 101 |
+
"step": 12
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.5135802469135803,
|
| 105 |
+
"grad_norm": 0.48423299193382263,
|
| 106 |
+
"learning_rate": 0.0001,
|
| 107 |
+
"loss": 0.1846,
|
| 108 |
+
"step": 13
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.5530864197530864,
|
| 112 |
+
"grad_norm": 0.4744335114955902,
|
| 113 |
+
"learning_rate": 9.998033131915266e-05,
|
| 114 |
+
"loss": 0.1631,
|
| 115 |
+
"step": 14
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.5925925925925926,
|
| 119 |
+
"grad_norm": 0.4236028492450714,
|
| 120 |
+
"learning_rate": 9.992134075089084e-05,
|
| 121 |
+
"loss": 0.1391,
|
| 122 |
+
"step": 15
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.6320987654320988,
|
| 126 |
+
"grad_norm": 0.46361225843429565,
|
| 127 |
+
"learning_rate": 9.982307470588098e-05,
|
| 128 |
+
"loss": 0.137,
|
| 129 |
+
"step": 16
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.671604938271605,
|
| 133 |
+
"grad_norm": 0.6843778491020203,
|
| 134 |
+
"learning_rate": 9.968561049466214e-05,
|
| 135 |
+
"loss": 0.1323,
|
| 136 |
+
"step": 17
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.7111111111111111,
|
| 140 |
+
"grad_norm": 0.3910984396934509,
|
| 141 |
+
"learning_rate": 9.950905626682228e-05,
|
| 142 |
+
"loss": 0.1141,
|
| 143 |
+
"step": 18
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.7506172839506173,
|
| 147 |
+
"grad_norm": 0.464354008436203,
|
| 148 |
+
"learning_rate": 9.92935509259118e-05,
|
| 149 |
+
"loss": 0.1178,
|
| 150 |
+
"step": 19
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.7901234567901234,
|
| 154 |
+
"grad_norm": 0.45064210891723633,
|
| 155 |
+
"learning_rate": 9.903926402016153e-05,
|
| 156 |
+
"loss": 0.1064,
|
| 157 |
+
"step": 20
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.7901234567901234,
|
| 161 |
+
"eval_loss": 0.10692152380943298,
|
| 162 |
+
"eval_runtime": 29.2333,
|
| 163 |
+
"eval_samples_per_second": 1.539,
|
| 164 |
+
"eval_steps_per_second": 1.539,
|
| 165 |
+
"step": 20
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.8296296296296296,
|
| 169 |
+
"grad_norm": 0.5124718546867371,
|
| 170 |
+
"learning_rate": 9.874639560909117e-05,
|
| 171 |
+
"loss": 0.1217,
|
| 172 |
+
"step": 21
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.8691358024691358,
|
| 176 |
+
"grad_norm": 0.4468311667442322,
|
| 177 |
+
"learning_rate": 9.841517610611309e-05,
|
| 178 |
+
"loss": 0.097,
|
| 179 |
+
"step": 22
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.908641975308642,
|
| 183 |
+
"grad_norm": 0.3930855095386505,
|
| 184 |
+
"learning_rate": 9.804586609725499e-05,
|
| 185 |
+
"loss": 0.0871,
|
| 186 |
+
"step": 23
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.9481481481481482,
|
| 190 |
+
"grad_norm": 0.44672492146492004,
|
| 191 |
+
"learning_rate": 9.763875613614482e-05,
|
| 192 |
+
"loss": 0.1053,
|
| 193 |
+
"step": 24
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.9876543209876543,
|
| 197 |
+
"grad_norm": 0.3032624125480652,
|
| 198 |
+
"learning_rate": 9.719416651541839e-05,
|
| 199 |
+
"loss": 0.0961,
|
| 200 |
+
"step": 25
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.0271604938271606,
|
| 204 |
+
"grad_norm": 0.3262697756290436,
|
| 205 |
+
"learning_rate": 9.671244701472999e-05,
|
| 206 |
+
"loss": 0.0775,
|
| 207 |
+
"step": 26
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.0666666666666667,
|
| 211 |
+
"grad_norm": 0.30042222142219543,
|
| 212 |
+
"learning_rate": 9.619397662556435e-05,
|
| 213 |
+
"loss": 0.0706,
|
| 214 |
+
"step": 27
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.106172839506173,
|
| 218 |
+
"grad_norm": 0.26015129685401917,
|
| 219 |
+
"learning_rate": 9.563916325306594e-05,
|
| 220 |
+
"loss": 0.0815,
|
| 221 |
+
"step": 28
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.145679012345679,
|
| 225 |
+
"grad_norm": 0.38614243268966675,
|
| 226 |
+
"learning_rate": 9.504844339512095e-05,
|
| 227 |
+
"loss": 0.0786,
|
| 228 |
+
"step": 29
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.1851851851851851,
|
| 232 |
+
"grad_norm": 0.2820267081260681,
|
| 233 |
+
"learning_rate": 9.442228179894362e-05,
|
| 234 |
+
"loss": 0.0661,
|
| 235 |
+
"step": 30
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.1851851851851851,
|
| 239 |
+
"eval_loss": 0.0779971331357956,
|
| 240 |
+
"eval_runtime": 29.2289,
|
| 241 |
+
"eval_samples_per_second": 1.54,
|
| 242 |
+
"eval_steps_per_second": 1.54,
|
| 243 |
+
"step": 30
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.2246913580246914,
|
| 247 |
+
"grad_norm": 0.31274303793907166,
|
| 248 |
+
"learning_rate": 9.376117109543769e-05,
|
| 249 |
+
"loss": 0.0675,
|
| 250 |
+
"step": 31
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.2641975308641975,
|
| 254 |
+
"grad_norm": 0.31122153997421265,
|
| 255 |
+
"learning_rate": 9.306563141162046e-05,
|
| 256 |
+
"loss": 0.0717,
|
| 257 |
+
"step": 32
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.3037037037037038,
|
| 261 |
+
"grad_norm": 0.4171571433544159,
|
| 262 |
+
"learning_rate": 9.233620996141421e-05,
|
| 263 |
+
"loss": 0.0705,
|
| 264 |
+
"step": 33
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.34320987654321,
|
| 268 |
+
"grad_norm": 0.39130330085754395,
|
| 269 |
+
"learning_rate": 9.157348061512727e-05,
|
| 270 |
+
"loss": 0.0643,
|
| 271 |
+
"step": 34
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.382716049382716,
|
| 275 |
+
"grad_norm": 0.26237934827804565,
|
| 276 |
+
"learning_rate": 9.077804344796302e-05,
|
| 277 |
+
"loss": 0.0685,
|
| 278 |
+
"step": 35
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.4222222222222223,
|
| 282 |
+
"grad_norm": 0.3337700366973877,
|
| 283 |
+
"learning_rate": 8.995052426791247e-05,
|
| 284 |
+
"loss": 0.0599,
|
| 285 |
+
"step": 36
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.4617283950617284,
|
| 289 |
+
"grad_norm": 0.34181517362594604,
|
| 290 |
+
"learning_rate": 8.90915741234015e-05,
|
| 291 |
+
"loss": 0.0605,
|
| 292 |
+
"step": 37
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.5012345679012347,
|
| 296 |
+
"grad_norm": 0.4122283160686493,
|
| 297 |
+
"learning_rate": 8.820186879108038e-05,
|
| 298 |
+
"loss": 0.0659,
|
| 299 |
+
"step": 38
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.5407407407407407,
|
| 303 |
+
"grad_norm": 0.392880380153656,
|
| 304 |
+
"learning_rate": 8.728210824415827e-05,
|
| 305 |
+
"loss": 0.0681,
|
| 306 |
+
"step": 39
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.5802469135802468,
|
| 310 |
+
"grad_norm": 0.3869952857494354,
|
| 311 |
+
"learning_rate": 8.633301610170135e-05,
|
| 312 |
+
"loss": 0.0574,
|
| 313 |
+
"step": 40
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.5802469135802468,
|
| 317 |
+
"eval_loss": 0.06701873987913132,
|
| 318 |
+
"eval_runtime": 29.226,
|
| 319 |
+
"eval_samples_per_second": 1.54,
|
| 320 |
+
"eval_steps_per_second": 1.54,
|
| 321 |
+
"step": 40
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.6197530864197531,
|
| 325 |
+
"grad_norm": 0.3228525221347809,
|
| 326 |
+
"learning_rate": 8.535533905932738e-05,
|
| 327 |
+
"loss": 0.0736,
|
| 328 |
+
"step": 41
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.6592592592592592,
|
| 332 |
+
"grad_norm": 0.380230188369751,
|
| 333 |
+
"learning_rate": 8.434984630174509e-05,
|
| 334 |
+
"loss": 0.0485,
|
| 335 |
+
"step": 42
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.6987654320987655,
|
| 339 |
+
"grad_norm": 0.23963190615177155,
|
| 340 |
+
"learning_rate": 8.33173288976002e-05,
|
| 341 |
+
"loss": 0.0603,
|
| 342 |
+
"step": 43
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.7382716049382716,
|
| 346 |
+
"grad_norm": 0.2060232162475586,
|
| 347 |
+
"learning_rate": 8.225859917710439e-05,
|
| 348 |
+
"loss": 0.0528,
|
| 349 |
+
"step": 44
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.7777777777777777,
|
| 353 |
+
"grad_norm": 0.32117360830307007,
|
| 354 |
+
"learning_rate": 8.117449009293668e-05,
|
| 355 |
+
"loss": 0.0547,
|
| 356 |
+
"step": 45
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.817283950617284,
|
| 360 |
+
"grad_norm": 0.2599658668041229,
|
| 361 |
+
"learning_rate": 8.006585456492029e-05,
|
| 362 |
+
"loss": 0.0527,
|
| 363 |
+
"step": 46
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.8567901234567903,
|
| 367 |
+
"grad_norm": 0.2622724771499634,
|
| 368 |
+
"learning_rate": 7.89335648089903e-05,
|
| 369 |
+
"loss": 0.0671,
|
| 370 |
+
"step": 47
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.8962962962962964,
|
| 374 |
+
"grad_norm": 0.23480120301246643,
|
| 375 |
+
"learning_rate": 7.777851165098012e-05,
|
| 376 |
+
"loss": 0.0539,
|
| 377 |
+
"step": 48
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.9358024691358025,
|
| 381 |
+
"grad_norm": 0.2417857050895691,
|
| 382 |
+
"learning_rate": 7.660160382576683e-05,
|
| 383 |
+
"loss": 0.0603,
|
| 384 |
+
"step": 49
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.9753086419753085,
|
| 388 |
+
"grad_norm": 0.2173357754945755,
|
| 389 |
+
"learning_rate": 7.540376726232648e-05,
|
| 390 |
+
"loss": 0.0454,
|
| 391 |
+
"step": 50
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.9753086419753085,
|
| 395 |
+
"eval_loss": 0.05982673540711403,
|
| 396 |
+
"eval_runtime": 29.235,
|
| 397 |
+
"eval_samples_per_second": 1.539,
|
| 398 |
+
"eval_steps_per_second": 1.539,
|
| 399 |
+
"step": 50
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.0148148148148146,
|
| 403 |
+
"grad_norm": 0.22790434956550598,
|
| 404 |
+
"learning_rate": 7.4185944355262e-05,
|
| 405 |
+
"loss": 0.0492,
|
| 406 |
+
"step": 51
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.054320987654321,
|
| 410 |
+
"grad_norm": 0.24523094296455383,
|
| 411 |
+
"learning_rate": 7.294909322337689e-05,
|
| 412 |
+
"loss": 0.0436,
|
| 413 |
+
"step": 52
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.093827160493827,
|
| 417 |
+
"grad_norm": 0.172611802816391,
|
| 418 |
+
"learning_rate": 7.169418695587791e-05,
|
| 419 |
+
"loss": 0.0382,
|
| 420 |
+
"step": 53
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.1333333333333333,
|
| 424 |
+
"grad_norm": 0.19519709050655365,
|
| 425 |
+
"learning_rate": 7.042221284679982e-05,
|
| 426 |
+
"loss": 0.0482,
|
| 427 |
+
"step": 54
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.1728395061728394,
|
| 431 |
+
"grad_norm": 0.29516759514808655,
|
| 432 |
+
"learning_rate": 6.91341716182545e-05,
|
| 433 |
+
"loss": 0.0428,
|
| 434 |
+
"step": 55
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.212345679012346,
|
| 438 |
+
"grad_norm": 0.3404478132724762,
|
| 439 |
+
"learning_rate": 6.783107663311565e-05,
|
| 440 |
+
"loss": 0.0378,
|
| 441 |
+
"step": 56
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 2.251851851851852,
|
| 445 |
+
"grad_norm": 0.20548518002033234,
|
| 446 |
+
"learning_rate": 6.651395309775837e-05,
|
| 447 |
+
"loss": 0.0421,
|
| 448 |
+
"step": 57
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 2.291358024691358,
|
| 452 |
+
"grad_norm": 0.27528247237205505,
|
| 453 |
+
"learning_rate": 6.518383725548074e-05,
|
| 454 |
+
"loss": 0.0414,
|
| 455 |
+
"step": 58
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 2.330864197530864,
|
| 459 |
+
"grad_norm": 0.23328714072704315,
|
| 460 |
+
"learning_rate": 6.384177557124247e-05,
|
| 461 |
+
"loss": 0.0482,
|
| 462 |
+
"step": 59
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.3703703703703702,
|
| 466 |
+
"grad_norm": 0.20340496301651,
|
| 467 |
+
"learning_rate": 6.248882390836135e-05,
|
| 468 |
+
"loss": 0.0416,
|
| 469 |
+
"step": 60
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 2.3703703703703702,
|
| 473 |
+
"eval_loss": 0.05630980432033539,
|
| 474 |
+
"eval_runtime": 29.2424,
|
| 475 |
+
"eval_samples_per_second": 1.539,
|
| 476 |
+
"eval_steps_per_second": 1.539,
|
| 477 |
+
"step": 60
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.4098765432098768,
|
| 481 |
+
"grad_norm": 0.23710526525974274,
|
| 482 |
+
"learning_rate": 6.112604669781572e-05,
|
| 483 |
+
"loss": 0.0396,
|
| 484 |
+
"step": 61
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 2.449382716049383,
|
| 488 |
+
"grad_norm": 0.21828940510749817,
|
| 489 |
+
"learning_rate": 5.9754516100806423e-05,
|
| 490 |
+
"loss": 0.04,
|
| 491 |
+
"step": 62
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 2.488888888888889,
|
| 495 |
+
"grad_norm": 0.24057340621948242,
|
| 496 |
+
"learning_rate": 5.837531116523682e-05,
|
| 497 |
+
"loss": 0.0486,
|
| 498 |
+
"step": 63
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.528395061728395,
|
| 502 |
+
"grad_norm": 0.24865029752254486,
|
| 503 |
+
"learning_rate": 5.698951697677498e-05,
|
| 504 |
+
"loss": 0.0442,
|
| 505 |
+
"step": 64
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 2.567901234567901,
|
| 509 |
+
"grad_norm": 0.2211083471775055,
|
| 510 |
+
"learning_rate": 5.559822380516539e-05,
|
| 511 |
+
"loss": 0.0483,
|
| 512 |
+
"step": 65
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.6074074074074076,
|
| 516 |
+
"grad_norm": 0.21438068151474,
|
| 517 |
+
"learning_rate": 5.420252624646238e-05,
|
| 518 |
+
"loss": 0.0517,
|
| 519 |
+
"step": 66
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.6469135802469137,
|
| 523 |
+
"grad_norm": 0.2499130219221115,
|
| 524 |
+
"learning_rate": 5.2803522361859594e-05,
|
| 525 |
+
"loss": 0.0482,
|
| 526 |
+
"step": 67
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.68641975308642,
|
| 530 |
+
"grad_norm": 0.1926133632659912,
|
| 531 |
+
"learning_rate": 5.140231281379345e-05,
|
| 532 |
+
"loss": 0.0348,
|
| 533 |
+
"step": 68
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.725925925925926,
|
| 537 |
+
"grad_norm": 0.18554463982582092,
|
| 538 |
+
"learning_rate": 5e-05,
|
| 539 |
+
"loss": 0.0384,
|
| 540 |
+
"step": 69
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.765432098765432,
|
| 544 |
+
"grad_norm": 0.22229990363121033,
|
| 545 |
+
"learning_rate": 4.859768718620656e-05,
|
| 546 |
+
"loss": 0.0462,
|
| 547 |
+
"step": 70
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 2.765432098765432,
|
| 551 |
+
"eval_loss": 0.05303001031279564,
|
| 552 |
+
"eval_runtime": 29.236,
|
| 553 |
+
"eval_samples_per_second": 1.539,
|
| 554 |
+
"eval_steps_per_second": 1.539,
|
| 555 |
+
"step": 70
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.8049382716049385,
|
| 559 |
+
"grad_norm": 0.302468866109848,
|
| 560 |
+
"learning_rate": 4.7196477638140404e-05,
|
| 561 |
+
"loss": 0.0434,
|
| 562 |
+
"step": 71
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.8444444444444446,
|
| 566 |
+
"grad_norm": 0.2213534712791443,
|
| 567 |
+
"learning_rate": 4.579747375353763e-05,
|
| 568 |
+
"loss": 0.0394,
|
| 569 |
+
"step": 72
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.8839506172839506,
|
| 573 |
+
"grad_norm": 0.21534393727779388,
|
| 574 |
+
"learning_rate": 4.4401776194834613e-05,
|
| 575 |
+
"loss": 0.0414,
|
| 576 |
+
"step": 73
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.9234567901234567,
|
| 580 |
+
"grad_norm": 0.26027560234069824,
|
| 581 |
+
"learning_rate": 4.3010483023225045e-05,
|
| 582 |
+
"loss": 0.0428,
|
| 583 |
+
"step": 74
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.962962962962963,
|
| 587 |
+
"grad_norm": 0.2458033412694931,
|
| 588 |
+
"learning_rate": 4.162468883476319e-05,
|
| 589 |
+
"loss": 0.0383,
|
| 590 |
+
"step": 75
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 3.0024691358024693,
|
| 594 |
+
"grad_norm": 0.17401082813739777,
|
| 595 |
+
"learning_rate": 4.0245483899193595e-05,
|
| 596 |
+
"loss": 0.0394,
|
| 597 |
+
"step": 76
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.0419753086419754,
|
| 601 |
+
"grad_norm": 0.15630896389484406,
|
| 602 |
+
"learning_rate": 3.887395330218429e-05,
|
| 603 |
+
"loss": 0.0317,
|
| 604 |
+
"step": 77
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 3.0814814814814815,
|
| 608 |
+
"grad_norm": 0.16098880767822266,
|
| 609 |
+
"learning_rate": 3.7511176091638653e-05,
|
| 610 |
+
"loss": 0.0267,
|
| 611 |
+
"step": 78
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 3.1209876543209876,
|
| 615 |
+
"grad_norm": 0.21407581865787506,
|
| 616 |
+
"learning_rate": 3.6158224428757535e-05,
|
| 617 |
+
"loss": 0.0345,
|
| 618 |
+
"step": 79
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 3.1604938271604937,
|
| 622 |
+
"grad_norm": 0.18872970342636108,
|
| 623 |
+
"learning_rate": 3.4816162744519263e-05,
|
| 624 |
+
"loss": 0.0333,
|
| 625 |
+
"step": 80
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 3.1604938271604937,
|
| 629 |
+
"eval_loss": 0.05208759754896164,
|
| 630 |
+
"eval_runtime": 29.23,
|
| 631 |
+
"eval_samples_per_second": 1.54,
|
| 632 |
+
"eval_steps_per_second": 1.54,
|
| 633 |
+
"step": 80
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.2,
|
| 637 |
+
"grad_norm": 0.18860113620758057,
|
| 638 |
+
"learning_rate": 3.3486046902241664e-05,
|
| 639 |
+
"loss": 0.0355,
|
| 640 |
+
"step": 81
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 3.2395061728395063,
|
| 644 |
+
"grad_norm": 0.19854336977005005,
|
| 645 |
+
"learning_rate": 3.216892336688435e-05,
|
| 646 |
+
"loss": 0.0355,
|
| 647 |
+
"step": 82
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.2790123456790123,
|
| 651 |
+
"grad_norm": 0.19552765786647797,
|
| 652 |
+
"learning_rate": 3.086582838174551e-05,
|
| 653 |
+
"loss": 0.0319,
|
| 654 |
+
"step": 83
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.3185185185185184,
|
| 658 |
+
"grad_norm": 0.17172157764434814,
|
| 659 |
+
"learning_rate": 2.9577787153200197e-05,
|
| 660 |
+
"loss": 0.0312,
|
| 661 |
+
"step": 84
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.3580246913580245,
|
| 665 |
+
"grad_norm": 0.16564936935901642,
|
| 666 |
+
"learning_rate": 2.8305813044122097e-05,
|
| 667 |
+
"loss": 0.032,
|
| 668 |
+
"step": 85
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.397530864197531,
|
| 672 |
+
"grad_norm": 0.19835302233695984,
|
| 673 |
+
"learning_rate": 2.705090677662311e-05,
|
| 674 |
+
"loss": 0.0392,
|
| 675 |
+
"step": 86
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 3.437037037037037,
|
| 679 |
+
"grad_norm": 0.24695785343647003,
|
| 680 |
+
"learning_rate": 2.581405564473801e-05,
|
| 681 |
+
"loss": 0.0372,
|
| 682 |
+
"step": 87
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 3.476543209876543,
|
| 686 |
+
"grad_norm": 0.1902298927307129,
|
| 687 |
+
"learning_rate": 2.459623273767354e-05,
|
| 688 |
+
"loss": 0.0335,
|
| 689 |
+
"step": 88
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 3.5160493827160493,
|
| 693 |
+
"grad_norm": 0.23312592506408691,
|
| 694 |
+
"learning_rate": 2.3398396174233178e-05,
|
| 695 |
+
"loss": 0.0359,
|
| 696 |
+
"step": 89
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 3.5555555555555554,
|
| 700 |
+
"grad_norm": 0.2210913747549057,
|
| 701 |
+
"learning_rate": 2.2221488349019903e-05,
|
| 702 |
+
"loss": 0.0389,
|
| 703 |
+
"step": 90
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 3.5555555555555554,
|
| 707 |
+
"eval_loss": 0.05143319070339203,
|
| 708 |
+
"eval_runtime": 29.2368,
|
| 709 |
+
"eval_samples_per_second": 1.539,
|
| 710 |
+
"eval_steps_per_second": 1.539,
|
| 711 |
+
"step": 90
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.595061728395062,
|
| 715 |
+
"grad_norm": 0.21577894687652588,
|
| 716 |
+
"learning_rate": 2.1066435191009715e-05,
|
| 717 |
+
"loss": 0.0362,
|
| 718 |
+
"step": 91
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 3.634567901234568,
|
| 722 |
+
"grad_norm": 0.19230662286281586,
|
| 723 |
+
"learning_rate": 1.9934145435079702e-05,
|
| 724 |
+
"loss": 0.0318,
|
| 725 |
+
"step": 92
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 3.674074074074074,
|
| 729 |
+
"grad_norm": 0.20312044024467468,
|
| 730 |
+
"learning_rate": 1.8825509907063327e-05,
|
| 731 |
+
"loss": 0.0302,
|
| 732 |
+
"step": 93
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 3.71358024691358,
|
| 736 |
+
"grad_norm": 0.2292894870042801,
|
| 737 |
+
"learning_rate": 1.774140082289563e-05,
|
| 738 |
+
"loss": 0.0397,
|
| 739 |
+
"step": 94
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 3.753086419753086,
|
| 743 |
+
"grad_norm": 0.3409978151321411,
|
| 744 |
+
"learning_rate": 1.6682671102399805e-05,
|
| 745 |
+
"loss": 0.0318,
|
| 746 |
+
"step": 95
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 3.7925925925925927,
|
| 750 |
+
"grad_norm": 0.1991758942604065,
|
| 751 |
+
"learning_rate": 1.5650153698254916e-05,
|
| 752 |
+
"loss": 0.0255,
|
| 753 |
+
"step": 96
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.832098765432099,
|
| 757 |
+
"grad_norm": 0.1859804391860962,
|
| 758 |
+
"learning_rate": 1.4644660940672627e-05,
|
| 759 |
+
"loss": 0.0338,
|
| 760 |
+
"step": 97
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 3.871604938271605,
|
| 764 |
+
"grad_norm": 0.3077707290649414,
|
| 765 |
+
"learning_rate": 1.3666983898298657e-05,
|
| 766 |
+
"loss": 0.0348,
|
| 767 |
+
"step": 98
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 3.911111111111111,
|
| 771 |
+
"grad_norm": 0.174801304936409,
|
| 772 |
+
"learning_rate": 1.2717891755841722e-05,
|
| 773 |
+
"loss": 0.028,
|
| 774 |
+
"step": 99
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.950617283950617,
|
| 778 |
+
"grad_norm": 0.196518212556839,
|
| 779 |
+
"learning_rate": 1.1798131208919627e-05,
|
| 780 |
+
"loss": 0.0338,
|
| 781 |
+
"step": 100
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.950617283950617,
|
| 785 |
+
"eval_loss": 0.0505523718893528,
|
| 786 |
+
"eval_runtime": 29.2378,
|
| 787 |
+
"eval_samples_per_second": 1.539,
|
| 788 |
+
"eval_steps_per_second": 1.539,
|
| 789 |
+
"step": 100
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.9901234567901236,
|
| 793 |
+
"grad_norm": 0.20451030135154724,
|
| 794 |
+
"learning_rate": 1.090842587659851e-05,
|
| 795 |
+
"loss": 0.0377,
|
| 796 |
+
"step": 101
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 4.029629629629629,
|
| 800 |
+
"grad_norm": 0.17839042842388153,
|
| 801 |
+
"learning_rate": 1.004947573208756e-05,
|
| 802 |
+
"loss": 0.0314,
|
| 803 |
+
"step": 102
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 4.069135802469136,
|
| 807 |
+
"grad_norm": 0.21361440420150757,
|
| 808 |
+
"learning_rate": 9.221956552036992e-06,
|
| 809 |
+
"loss": 0.0339,
|
| 810 |
+
"step": 103
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 4.108641975308642,
|
| 814 |
+
"grad_norm": 0.19031283259391785,
|
| 815 |
+
"learning_rate": 8.426519384872733e-06,
|
| 816 |
+
"loss": 0.0288,
|
| 817 |
+
"step": 104
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 4.148148148148148,
|
| 821 |
+
"grad_norm": 0.1789381355047226,
|
| 822 |
+
"learning_rate": 7.663790038585793e-06,
|
| 823 |
+
"loss": 0.0323,
|
| 824 |
+
"step": 105
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 4.187654320987654,
|
| 828 |
+
"grad_norm": 0.18017913401126862,
|
| 829 |
+
"learning_rate": 6.934368588379553e-06,
|
| 830 |
+
"loss": 0.035,
|
| 831 |
+
"step": 106
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 4.22716049382716,
|
| 835 |
+
"grad_norm": 0.18314358592033386,
|
| 836 |
+
"learning_rate": 6.238828904562316e-06,
|
| 837 |
+
"loss": 0.0263,
|
| 838 |
+
"step": 107
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 4.266666666666667,
|
| 842 |
+
"grad_norm": 0.20920193195343018,
|
| 843 |
+
"learning_rate": 5.577718201056392e-06,
|
| 844 |
+
"loss": 0.0337,
|
| 845 |
+
"step": 108
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 4.306172839506173,
|
| 849 |
+
"grad_norm": 0.17229609191417694,
|
| 850 |
+
"learning_rate": 4.951556604879048e-06,
|
| 851 |
+
"loss": 0.0303,
|
| 852 |
+
"step": 109
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 4.345679012345679,
|
| 856 |
+
"grad_norm": 0.1544303596019745,
|
| 857 |
+
"learning_rate": 4.360836746934055e-06,
|
| 858 |
+
"loss": 0.0269,
|
| 859 |
+
"step": 110
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 4.345679012345679,
|
| 863 |
+
"eval_loss": 0.05018164962530136,
|
| 864 |
+
"eval_runtime": 29.2504,
|
| 865 |
+
"eval_samples_per_second": 1.538,
|
| 866 |
+
"eval_steps_per_second": 1.538,
|
| 867 |
+
"step": 110
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 4.385185185185185,
|
| 871 |
+
"grad_norm": 0.18308553099632263,
|
| 872 |
+
"learning_rate": 3.8060233744356633e-06,
|
| 873 |
+
"loss": 0.03,
|
| 874 |
+
"step": 111
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.424691358024692,
|
| 878 |
+
"grad_norm": 0.19447027146816254,
|
| 879 |
+
"learning_rate": 3.2875529852700147e-06,
|
| 880 |
+
"loss": 0.0294,
|
| 881 |
+
"step": 112
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.4641975308641975,
|
| 885 |
+
"grad_norm": 0.15953049063682556,
|
| 886 |
+
"learning_rate": 2.8058334845816213e-06,
|
| 887 |
+
"loss": 0.0249,
|
| 888 |
+
"step": 113
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 4.503703703703704,
|
| 892 |
+
"grad_norm": 0.18681061267852783,
|
| 893 |
+
"learning_rate": 2.361243863855184e-06,
|
| 894 |
+
"loss": 0.0278,
|
| 895 |
+
"step": 114
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 4.54320987654321,
|
| 899 |
+
"grad_norm": 0.16752764582633972,
|
| 900 |
+
"learning_rate": 1.9541339027450256e-06,
|
| 901 |
+
"loss": 0.0262,
|
| 902 |
+
"step": 115
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 4.582716049382716,
|
| 906 |
+
"grad_norm": 0.18282541632652283,
|
| 907 |
+
"learning_rate": 1.584823893886933e-06,
|
| 908 |
+
"loss": 0.0349,
|
| 909 |
+
"step": 116
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 4.622222222222222,
|
| 913 |
+
"grad_norm": 0.1863105297088623,
|
| 914 |
+
"learning_rate": 1.2536043909088191e-06,
|
| 915 |
+
"loss": 0.0295,
|
| 916 |
+
"step": 117
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 4.661728395061728,
|
| 920 |
+
"grad_norm": 0.17035965621471405,
|
| 921 |
+
"learning_rate": 9.607359798384785e-07,
|
| 922 |
+
"loss": 0.0305,
|
| 923 |
+
"step": 118
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 4.701234567901235,
|
| 927 |
+
"grad_norm": 0.17452426254749298,
|
| 928 |
+
"learning_rate": 7.064490740882057e-07,
|
| 929 |
+
"loss": 0.0319,
|
| 930 |
+
"step": 119
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 4.7407407407407405,
|
| 934 |
+
"grad_norm": 0.17947931587696075,
|
| 935 |
+
"learning_rate": 4.909437331777179e-07,
|
| 936 |
+
"loss": 0.0289,
|
| 937 |
+
"step": 120
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 4.7407407407407405,
|
| 941 |
+
"eval_loss": 0.050184786319732666,
|
| 942 |
+
"eval_runtime": 29.264,
|
| 943 |
+
"eval_samples_per_second": 1.538,
|
| 944 |
+
"eval_steps_per_second": 1.538,
|
| 945 |
+
"step": 120
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 4.780246913580247,
|
| 949 |
+
"grad_norm": 0.19741973280906677,
|
| 950 |
+
"learning_rate": 3.143895053378698e-07,
|
| 951 |
+
"loss": 0.0266,
|
| 952 |
+
"step": 121
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 4.8197530864197535,
|
| 956 |
+
"grad_norm": 0.17615406215190887,
|
| 957 |
+
"learning_rate": 1.7692529411904578e-07,
|
| 958 |
+
"loss": 0.0299,
|
| 959 |
+
"step": 122
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 4.859259259259259,
|
| 963 |
+
"grad_norm": 0.15755115449428558,
|
| 964 |
+
"learning_rate": 7.865924910916977e-08,
|
| 965 |
+
"loss": 0.0284,
|
| 966 |
+
"step": 123
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 4.898765432098766,
|
| 970 |
+
"grad_norm": 0.1805194765329361,
|
| 971 |
+
"learning_rate": 1.9668680847356735e-08,
|
| 972 |
+
"loss": 0.0279,
|
| 973 |
+
"step": 124
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 4.938271604938271,
|
| 977 |
+
"grad_norm": 0.1919160634279251,
|
| 978 |
+
"learning_rate": 0.0,
|
| 979 |
+
"loss": 0.0373,
|
| 980 |
+
"step": 125
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 4.938271604938271,
|
| 984 |
+
"step": 125,
|
| 985 |
+
"total_flos": 1.743556468483031e+17,
|
| 986 |
+
"train_loss": 0.07543200919032098,
|
| 987 |
+
"train_runtime": 5142.0517,
|
| 988 |
+
"train_samples_per_second": 0.394,
|
| 989 |
+
"train_steps_per_second": 0.024
|
| 990 |
+
}
|
| 991 |
+
],
|
| 992 |
+
"logging_steps": 1,
|
| 993 |
+
"max_steps": 125,
|
| 994 |
+
"num_input_tokens_seen": 0,
|
| 995 |
+
"num_train_epochs": 5,
|
| 996 |
+
"save_steps": 10,
|
| 997 |
+
"stateful_callbacks": {
|
| 998 |
+
"TrainerControl": {
|
| 999 |
+
"args": {
|
| 1000 |
+
"should_epoch_stop": false,
|
| 1001 |
+
"should_evaluate": false,
|
| 1002 |
+
"should_log": false,
|
| 1003 |
+
"should_save": true,
|
| 1004 |
+
"should_training_stop": true
|
| 1005 |
+
},
|
| 1006 |
+
"attributes": {}
|
| 1007 |
+
}
|
| 1008 |
+
},
|
| 1009 |
+
"total_flos": 1.743556468483031e+17,
|
| 1010 |
+
"train_batch_size": 1,
|
| 1011 |
+
"trial_name": null,
|
| 1012 |
+
"trial_params": null
|
| 1013 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|