End of training
Browse files- README.md +3 -2
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +2894 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: mit
|
|
| 4 |
base_model: klyang/MentaLLaMA-chat-7B-hf
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: MentaLLaMA-chat-7B-PsyCourse-fold6
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# MentaLLaMA-chat-7B-PsyCourse-fold6
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [klyang/MentaLLaMA-chat-7B-hf](https://huggingface.co/klyang/MentaLLaMA-chat-7B-hf) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 4 |
base_model: klyang/MentaLLaMA-chat-7B-hf
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: MentaLLaMA-chat-7B-PsyCourse-fold6
|
|
|
|
| 16 |
|
| 17 |
# MentaLLaMA-chat-7B-PsyCourse-fold6
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [klyang/MentaLLaMA-chat-7B-hf](https://huggingface.co/klyang/MentaLLaMA-chat-7B-hf) on the course-train-fold6 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0319
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.995305164319249,
|
| 3 |
+
"eval_loss": 0.03188331797719002,
|
| 4 |
+
"eval_runtime": 219.0038,
|
| 5 |
+
"eval_samples_per_second": 5.406,
|
| 6 |
+
"eval_steps_per_second": 5.406,
|
| 7 |
+
"total_flos": 8.987524405535048e+17,
|
| 8 |
+
"train_loss": 0.049058449685125426,
|
| 9 |
+
"train_runtime": 45792.1032,
|
| 10 |
+
"train_samples_per_second": 1.163,
|
| 11 |
+
"train_steps_per_second": 0.073
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.995305164319249,
|
| 3 |
+
"eval_loss": 0.03188331797719002,
|
| 4 |
+
"eval_runtime": 219.0038,
|
| 5 |
+
"eval_samples_per_second": 5.406,
|
| 6 |
+
"eval_steps_per_second": 5.406
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.995305164319249,
|
| 3 |
+
"total_flos": 8.987524405535048e+17,
|
| 4 |
+
"train_loss": 0.049058449685125426,
|
| 5 |
+
"train_runtime": 45792.1032,
|
| 6 |
+
"train_samples_per_second": 1.163,
|
| 7 |
+
"train_steps_per_second": 0.073
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2894 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03188331797719002,
|
| 3 |
+
"best_model_checkpoint": "saves/psy-course/MentaLLaMA-chat-7B/train/fold6/checkpoint-1900",
|
| 4 |
+
"epoch": 4.995305164319249,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 3325,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015023474178403756,
|
| 13 |
+
"grad_norm": 1.6643978357315063,
|
| 14 |
+
"learning_rate": 3.003003003003003e-06,
|
| 15 |
+
"loss": 1.6617,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03004694835680751,
|
| 20 |
+
"grad_norm": 1.970920443534851,
|
| 21 |
+
"learning_rate": 6.006006006006006e-06,
|
| 22 |
+
"loss": 1.7717,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04507042253521127,
|
| 27 |
+
"grad_norm": 2.2089600563049316,
|
| 28 |
+
"learning_rate": 9.00900900900901e-06,
|
| 29 |
+
"loss": 1.6755,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06009389671361502,
|
| 34 |
+
"grad_norm": 3.677905797958374,
|
| 35 |
+
"learning_rate": 1.2012012012012012e-05,
|
| 36 |
+
"loss": 1.3133,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07511737089201878,
|
| 41 |
+
"grad_norm": 1.0802251100540161,
|
| 42 |
+
"learning_rate": 1.5015015015015016e-05,
|
| 43 |
+
"loss": 0.8292,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07511737089201878,
|
| 48 |
+
"eval_loss": 0.6473308205604553,
|
| 49 |
+
"eval_runtime": 207.0843,
|
| 50 |
+
"eval_samples_per_second": 5.717,
|
| 51 |
+
"eval_steps_per_second": 5.717,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09014084507042254,
|
| 56 |
+
"grad_norm": 0.8090672492980957,
|
| 57 |
+
"learning_rate": 1.801801801801802e-05,
|
| 58 |
+
"loss": 0.6816,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.10516431924882629,
|
| 63 |
+
"grad_norm": 0.7934019565582275,
|
| 64 |
+
"learning_rate": 2.102102102102102e-05,
|
| 65 |
+
"loss": 0.4595,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.12018779342723004,
|
| 70 |
+
"grad_norm": 0.7469967007637024,
|
| 71 |
+
"learning_rate": 2.4024024024024024e-05,
|
| 72 |
+
"loss": 0.2868,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1352112676056338,
|
| 77 |
+
"grad_norm": 0.6989622712135315,
|
| 78 |
+
"learning_rate": 2.702702702702703e-05,
|
| 79 |
+
"loss": 0.1691,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.15023474178403756,
|
| 84 |
+
"grad_norm": 0.714993417263031,
|
| 85 |
+
"learning_rate": 3.0030030030030033e-05,
|
| 86 |
+
"loss": 0.1595,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.15023474178403756,
|
| 91 |
+
"eval_loss": 0.11690635979175568,
|
| 92 |
+
"eval_runtime": 208.0685,
|
| 93 |
+
"eval_samples_per_second": 5.69,
|
| 94 |
+
"eval_steps_per_second": 5.69,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.1652582159624413,
|
| 99 |
+
"grad_norm": 0.6915186643600464,
|
| 100 |
+
"learning_rate": 3.3033033033033035e-05,
|
| 101 |
+
"loss": 0.1358,
|
| 102 |
+
"step": 110
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.18028169014084508,
|
| 106 |
+
"grad_norm": 0.7759954333305359,
|
| 107 |
+
"learning_rate": 3.603603603603604e-05,
|
| 108 |
+
"loss": 0.1063,
|
| 109 |
+
"step": 120
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.19530516431924883,
|
| 113 |
+
"grad_norm": 0.8066917061805725,
|
| 114 |
+
"learning_rate": 3.903903903903904e-05,
|
| 115 |
+
"loss": 0.0996,
|
| 116 |
+
"step": 130
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.21032863849765257,
|
| 120 |
+
"grad_norm": 0.4547995924949646,
|
| 121 |
+
"learning_rate": 4.204204204204204e-05,
|
| 122 |
+
"loss": 0.1047,
|
| 123 |
+
"step": 140
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.22535211267605634,
|
| 127 |
+
"grad_norm": 0.7537536025047302,
|
| 128 |
+
"learning_rate": 4.5045045045045046e-05,
|
| 129 |
+
"loss": 0.0933,
|
| 130 |
+
"step": 150
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.22535211267605634,
|
| 134 |
+
"eval_loss": 0.07271095365285873,
|
| 135 |
+
"eval_runtime": 207.977,
|
| 136 |
+
"eval_samples_per_second": 5.693,
|
| 137 |
+
"eval_steps_per_second": 5.693,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.2403755868544601,
|
| 142 |
+
"grad_norm": 0.652043879032135,
|
| 143 |
+
"learning_rate": 4.804804804804805e-05,
|
| 144 |
+
"loss": 0.0726,
|
| 145 |
+
"step": 160
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.25539906103286386,
|
| 149 |
+
"grad_norm": 0.7913267016410828,
|
| 150 |
+
"learning_rate": 5.105105105105106e-05,
|
| 151 |
+
"loss": 0.0503,
|
| 152 |
+
"step": 170
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.2704225352112676,
|
| 156 |
+
"grad_norm": 0.6111043095588684,
|
| 157 |
+
"learning_rate": 5.405405405405406e-05,
|
| 158 |
+
"loss": 0.0682,
|
| 159 |
+
"step": 180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.28544600938967135,
|
| 163 |
+
"grad_norm": 0.6898893713951111,
|
| 164 |
+
"learning_rate": 5.705705705705706e-05,
|
| 165 |
+
"loss": 0.0651,
|
| 166 |
+
"step": 190
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.3004694835680751,
|
| 170 |
+
"grad_norm": 0.5223749876022339,
|
| 171 |
+
"learning_rate": 6.0060060060060066e-05,
|
| 172 |
+
"loss": 0.0512,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.3004694835680751,
|
| 177 |
+
"eval_loss": 0.05808287858963013,
|
| 178 |
+
"eval_runtime": 207.9535,
|
| 179 |
+
"eval_samples_per_second": 5.694,
|
| 180 |
+
"eval_steps_per_second": 5.694,
|
| 181 |
+
"step": 200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.3154929577464789,
|
| 185 |
+
"grad_norm": 0.7966965436935425,
|
| 186 |
+
"learning_rate": 6.306306306306306e-05,
|
| 187 |
+
"loss": 0.0562,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.3305164319248826,
|
| 192 |
+
"grad_norm": 0.4759034514427185,
|
| 193 |
+
"learning_rate": 6.606606606606607e-05,
|
| 194 |
+
"loss": 0.0591,
|
| 195 |
+
"step": 220
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.3455399061032864,
|
| 199 |
+
"grad_norm": 0.421832412481308,
|
| 200 |
+
"learning_rate": 6.906906906906907e-05,
|
| 201 |
+
"loss": 0.0504,
|
| 202 |
+
"step": 230
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.36056338028169016,
|
| 206 |
+
"grad_norm": 0.6044343113899231,
|
| 207 |
+
"learning_rate": 7.207207207207208e-05,
|
| 208 |
+
"loss": 0.0609,
|
| 209 |
+
"step": 240
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.3755868544600939,
|
| 213 |
+
"grad_norm": 0.702263355255127,
|
| 214 |
+
"learning_rate": 7.507507507507507e-05,
|
| 215 |
+
"loss": 0.0619,
|
| 216 |
+
"step": 250
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.3755868544600939,
|
| 220 |
+
"eval_loss": 0.04737601429224014,
|
| 221 |
+
"eval_runtime": 207.9514,
|
| 222 |
+
"eval_samples_per_second": 5.694,
|
| 223 |
+
"eval_steps_per_second": 5.694,
|
| 224 |
+
"step": 250
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.39061032863849765,
|
| 228 |
+
"grad_norm": 0.4086180627346039,
|
| 229 |
+
"learning_rate": 7.807807807807808e-05,
|
| 230 |
+
"loss": 0.0473,
|
| 231 |
+
"step": 260
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.4056338028169014,
|
| 235 |
+
"grad_norm": 0.3927551209926605,
|
| 236 |
+
"learning_rate": 8.108108108108109e-05,
|
| 237 |
+
"loss": 0.0552,
|
| 238 |
+
"step": 270
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.42065727699530514,
|
| 242 |
+
"grad_norm": 0.26502013206481934,
|
| 243 |
+
"learning_rate": 8.408408408408409e-05,
|
| 244 |
+
"loss": 0.0567,
|
| 245 |
+
"step": 280
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.4356807511737089,
|
| 249 |
+
"grad_norm": 0.8767386674880981,
|
| 250 |
+
"learning_rate": 8.70870870870871e-05,
|
| 251 |
+
"loss": 0.0572,
|
| 252 |
+
"step": 290
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.4507042253521127,
|
| 256 |
+
"grad_norm": 0.4874660074710846,
|
| 257 |
+
"learning_rate": 9.009009009009009e-05,
|
| 258 |
+
"loss": 0.0395,
|
| 259 |
+
"step": 300
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.4507042253521127,
|
| 263 |
+
"eval_loss": 0.04596845433115959,
|
| 264 |
+
"eval_runtime": 207.5439,
|
| 265 |
+
"eval_samples_per_second": 5.705,
|
| 266 |
+
"eval_steps_per_second": 5.705,
|
| 267 |
+
"step": 300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.46572769953051646,
|
| 271 |
+
"grad_norm": 0.43807005882263184,
|
| 272 |
+
"learning_rate": 9.30930930930931e-05,
|
| 273 |
+
"loss": 0.0444,
|
| 274 |
+
"step": 310
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.4807511737089202,
|
| 278 |
+
"grad_norm": 0.7260940074920654,
|
| 279 |
+
"learning_rate": 9.60960960960961e-05,
|
| 280 |
+
"loss": 0.063,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.49577464788732395,
|
| 285 |
+
"grad_norm": 0.40789633989334106,
|
| 286 |
+
"learning_rate": 9.90990990990991e-05,
|
| 287 |
+
"loss": 0.0497,
|
| 288 |
+
"step": 330
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.5107981220657277,
|
| 292 |
+
"grad_norm": 0.5078976154327393,
|
| 293 |
+
"learning_rate": 9.999864944989638e-05,
|
| 294 |
+
"loss": 0.058,
|
| 295 |
+
"step": 340
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.5258215962441315,
|
| 299 |
+
"grad_norm": 0.3737334907054901,
|
| 300 |
+
"learning_rate": 9.999203468625017e-05,
|
| 301 |
+
"loss": 0.0476,
|
| 302 |
+
"step": 350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.5258215962441315,
|
| 306 |
+
"eval_loss": 0.045401681214571,
|
| 307 |
+
"eval_runtime": 207.0191,
|
| 308 |
+
"eval_samples_per_second": 5.719,
|
| 309 |
+
"eval_steps_per_second": 5.719,
|
| 310 |
+
"step": 350
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5408450704225352,
|
| 314 |
+
"grad_norm": 0.5752517580986023,
|
| 315 |
+
"learning_rate": 9.997990837719421e-05,
|
| 316 |
+
"loss": 0.055,
|
| 317 |
+
"step": 360
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5558685446009389,
|
| 321 |
+
"grad_norm": 0.28732818365097046,
|
| 322 |
+
"learning_rate": 9.996227185963554e-05,
|
| 323 |
+
"loss": 0.0412,
|
| 324 |
+
"step": 370
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5708920187793427,
|
| 328 |
+
"grad_norm": 0.41930803656578064,
|
| 329 |
+
"learning_rate": 9.993912707797329e-05,
|
| 330 |
+
"loss": 0.0448,
|
| 331 |
+
"step": 380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5859154929577465,
|
| 335 |
+
"grad_norm": 0.35693687200546265,
|
| 336 |
+
"learning_rate": 9.99104765838842e-05,
|
| 337 |
+
"loss": 0.0423,
|
| 338 |
+
"step": 390
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6009389671361502,
|
| 342 |
+
"grad_norm": 0.11278364062309265,
|
| 343 |
+
"learning_rate": 9.987632353604151e-05,
|
| 344 |
+
"loss": 0.0444,
|
| 345 |
+
"step": 400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6009389671361502,
|
| 349 |
+
"eval_loss": 0.0407099612057209,
|
| 350 |
+
"eval_runtime": 207.012,
|
| 351 |
+
"eval_samples_per_second": 5.719,
|
| 352 |
+
"eval_steps_per_second": 5.719,
|
| 353 |
+
"step": 400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.615962441314554,
|
| 357 |
+
"grad_norm": 0.41261032223701477,
|
| 358 |
+
"learning_rate": 9.98366716997665e-05,
|
| 359 |
+
"loss": 0.0478,
|
| 360 |
+
"step": 410
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6309859154929578,
|
| 364 |
+
"grad_norm": 0.4349944293498993,
|
| 365 |
+
"learning_rate": 9.979152544661354e-05,
|
| 366 |
+
"loss": 0.0381,
|
| 367 |
+
"step": 420
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6460093896713615,
|
| 371 |
+
"grad_norm": 0.609887957572937,
|
| 372 |
+
"learning_rate": 9.974088975388802e-05,
|
| 373 |
+
"loss": 0.0435,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6610328638497652,
|
| 378 |
+
"grad_norm": 0.37665075063705444,
|
| 379 |
+
"learning_rate": 9.968477020409766e-05,
|
| 380 |
+
"loss": 0.0378,
|
| 381 |
+
"step": 440
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.676056338028169,
|
| 385 |
+
"grad_norm": 0.30656006932258606,
|
| 386 |
+
"learning_rate": 9.962317298433705e-05,
|
| 387 |
+
"loss": 0.0543,
|
| 388 |
+
"step": 450
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.676056338028169,
|
| 392 |
+
"eval_loss": 0.042538370937108994,
|
| 393 |
+
"eval_runtime": 207.26,
|
| 394 |
+
"eval_samples_per_second": 5.713,
|
| 395 |
+
"eval_steps_per_second": 5.713,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.6910798122065728,
|
| 400 |
+
"grad_norm": 0.2620968520641327,
|
| 401 |
+
"learning_rate": 9.955610488560551e-05,
|
| 402 |
+
"loss": 0.0529,
|
| 403 |
+
"step": 460
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.7061032863849765,
|
| 407 |
+
"grad_norm": 0.4962412118911743,
|
| 408 |
+
"learning_rate": 9.948357330205842e-05,
|
| 409 |
+
"loss": 0.0488,
|
| 410 |
+
"step": 470
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.7211267605633803,
|
| 414 |
+
"grad_norm": 0.11893105506896973,
|
| 415 |
+
"learning_rate": 9.940558623019201e-05,
|
| 416 |
+
"loss": 0.0282,
|
| 417 |
+
"step": 480
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.7361502347417841,
|
| 421 |
+
"grad_norm": 0.24164587259292603,
|
| 422 |
+
"learning_rate": 9.932215226796172e-05,
|
| 423 |
+
"loss": 0.0377,
|
| 424 |
+
"step": 490
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.7511737089201878,
|
| 428 |
+
"grad_norm": 0.23037177324295044,
|
| 429 |
+
"learning_rate": 9.923328061383435e-05,
|
| 430 |
+
"loss": 0.0454,
|
| 431 |
+
"step": 500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.7511737089201878,
|
| 435 |
+
"eval_loss": 0.0371524840593338,
|
| 436 |
+
"eval_runtime": 207.3218,
|
| 437 |
+
"eval_samples_per_second": 5.711,
|
| 438 |
+
"eval_steps_per_second": 5.711,
|
| 439 |
+
"step": 500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.7661971830985915,
|
| 443 |
+
"grad_norm": 0.1442870795726776,
|
| 444 |
+
"learning_rate": 9.913898106577393e-05,
|
| 445 |
+
"loss": 0.0382,
|
| 446 |
+
"step": 510
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.7812206572769953,
|
| 450 |
+
"grad_norm": 0.2532236576080322,
|
| 451 |
+
"learning_rate": 9.903926402016153e-05,
|
| 452 |
+
"loss": 0.0344,
|
| 453 |
+
"step": 520
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.7962441314553991,
|
| 457 |
+
"grad_norm": 0.14402006566524506,
|
| 458 |
+
"learning_rate": 9.893414047064897e-05,
|
| 459 |
+
"loss": 0.0398,
|
| 460 |
+
"step": 530
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.8112676056338028,
|
| 464 |
+
"grad_norm": 0.2404378205537796,
|
| 465 |
+
"learning_rate": 9.88236220069469e-05,
|
| 466 |
+
"loss": 0.0381,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.8262910798122066,
|
| 471 |
+
"grad_norm": 0.27102166414260864,
|
| 472 |
+
"learning_rate": 9.870772081354705e-05,
|
| 473 |
+
"loss": 0.0562,
|
| 474 |
+
"step": 550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.8262910798122066,
|
| 478 |
+
"eval_loss": 0.03766689449548721,
|
| 479 |
+
"eval_runtime": 207.5601,
|
| 480 |
+
"eval_samples_per_second": 5.704,
|
| 481 |
+
"eval_steps_per_second": 5.704,
|
| 482 |
+
"step": 550
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.8413145539906103,
|
| 486 |
+
"grad_norm": 0.2679292857646942,
|
| 487 |
+
"learning_rate": 9.858644966837878e-05,
|
| 488 |
+
"loss": 0.0406,
|
| 489 |
+
"step": 560
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.856338028169014,
|
| 493 |
+
"grad_norm": 0.0837322548031807,
|
| 494 |
+
"learning_rate": 9.845982194140051e-05,
|
| 495 |
+
"loss": 0.0381,
|
| 496 |
+
"step": 570
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.8713615023474178,
|
| 500 |
+
"grad_norm": 0.1658477932214737,
|
| 501 |
+
"learning_rate": 9.832785159312559e-05,
|
| 502 |
+
"loss": 0.0404,
|
| 503 |
+
"step": 580
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.8863849765258216,
|
| 507 |
+
"grad_norm": 0.2504975199699402,
|
| 508 |
+
"learning_rate": 9.819055317308317e-05,
|
| 509 |
+
"loss": 0.0365,
|
| 510 |
+
"step": 590
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.9014084507042254,
|
| 514 |
+
"grad_norm": 0.24881432950496674,
|
| 515 |
+
"learning_rate": 9.804794181821422e-05,
|
| 516 |
+
"loss": 0.0336,
|
| 517 |
+
"step": 600
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 0.9014084507042254,
|
| 521 |
+
"eval_loss": 0.036089226603507996,
|
| 522 |
+
"eval_runtime": 208.6413,
|
| 523 |
+
"eval_samples_per_second": 5.675,
|
| 524 |
+
"eval_steps_per_second": 5.675,
|
| 525 |
+
"step": 600
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.9164319248826291,
|
| 529 |
+
"grad_norm": 0.20400799810886383,
|
| 530 |
+
"learning_rate": 9.790003325120261e-05,
|
| 531 |
+
"loss": 0.0375,
|
| 532 |
+
"step": 610
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.9314553990610329,
|
| 536 |
+
"grad_norm": 0.4139688313007355,
|
| 537 |
+
"learning_rate": 9.774684377874178e-05,
|
| 538 |
+
"loss": 0.0466,
|
| 539 |
+
"step": 620
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.9464788732394366,
|
| 543 |
+
"grad_norm": 0.302133172750473,
|
| 544 |
+
"learning_rate": 9.758839028973692e-05,
|
| 545 |
+
"loss": 0.0308,
|
| 546 |
+
"step": 630
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.9615023474178404,
|
| 550 |
+
"grad_norm": 0.26886606216430664,
|
| 551 |
+
"learning_rate": 9.742469025344298e-05,
|
| 552 |
+
"loss": 0.0416,
|
| 553 |
+
"step": 640
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.9765258215962441,
|
| 557 |
+
"grad_norm": 0.22003549337387085,
|
| 558 |
+
"learning_rate": 9.725576171753874e-05,
|
| 559 |
+
"loss": 0.0494,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.9765258215962441,
|
| 564 |
+
"eval_loss": 0.03678149729967117,
|
| 565 |
+
"eval_runtime": 209.7916,
|
| 566 |
+
"eval_samples_per_second": 5.644,
|
| 567 |
+
"eval_steps_per_second": 5.644,
|
| 568 |
+
"step": 650
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.9915492957746479,
|
| 572 |
+
"grad_norm": 0.686739981174469,
|
| 573 |
+
"learning_rate": 9.708162330613708e-05,
|
| 574 |
+
"loss": 0.0367,
|
| 575 |
+
"step": 660
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.0065727699530516,
|
| 579 |
+
"grad_norm": 0.1825464963912964,
|
| 580 |
+
"learning_rate": 9.690229421773167e-05,
|
| 581 |
+
"loss": 0.0369,
|
| 582 |
+
"step": 670
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.0215962441314554,
|
| 586 |
+
"grad_norm": 0.442874014377594,
|
| 587 |
+
"learning_rate": 9.67177942230804e-05,
|
| 588 |
+
"loss": 0.0312,
|
| 589 |
+
"step": 680
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.036619718309859,
|
| 593 |
+
"grad_norm": 0.45542624592781067,
|
| 594 |
+
"learning_rate": 9.652814366302568e-05,
|
| 595 |
+
"loss": 0.0392,
|
| 596 |
+
"step": 690
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.051643192488263,
|
| 600 |
+
"grad_norm": 0.16704720258712769,
|
| 601 |
+
"learning_rate": 9.633336344625185e-05,
|
| 602 |
+
"loss": 0.0354,
|
| 603 |
+
"step": 700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.051643192488263,
|
| 607 |
+
"eval_loss": 0.0386400930583477,
|
| 608 |
+
"eval_runtime": 210.8244,
|
| 609 |
+
"eval_samples_per_second": 5.616,
|
| 610 |
+
"eval_steps_per_second": 5.616,
|
| 611 |
+
"step": 700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.0666666666666667,
|
| 615 |
+
"grad_norm": 0.2750037610530853,
|
| 616 |
+
"learning_rate": 9.61334750469801e-05,
|
| 617 |
+
"loss": 0.0282,
|
| 618 |
+
"step": 710
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.0816901408450703,
|
| 622 |
+
"grad_norm": 0.5251227021217346,
|
| 623 |
+
"learning_rate": 9.592850050260089e-05,
|
| 624 |
+
"loss": 0.0349,
|
| 625 |
+
"step": 720
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.0967136150234742,
|
| 629 |
+
"grad_norm": 0.13910022377967834,
|
| 630 |
+
"learning_rate": 9.571846241124446e-05,
|
| 631 |
+
"loss": 0.0291,
|
| 632 |
+
"step": 730
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1117370892018779,
|
| 636 |
+
"grad_norm": 0.17034609615802765,
|
| 637 |
+
"learning_rate": 9.55033839292893e-05,
|
| 638 |
+
"loss": 0.0257,
|
| 639 |
+
"step": 740
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1267605633802817,
|
| 643 |
+
"grad_norm": 0.4655674397945404,
|
| 644 |
+
"learning_rate": 9.52832887688093e-05,
|
| 645 |
+
"loss": 0.029,
|
| 646 |
+
"step": 750
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1267605633802817,
|
| 650 |
+
"eval_loss": 0.037642158567905426,
|
| 651 |
+
"eval_runtime": 212.6123,
|
| 652 |
+
"eval_samples_per_second": 5.569,
|
| 653 |
+
"eval_steps_per_second": 5.569,
|
| 654 |
+
"step": 750
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.1417840375586854,
|
| 658 |
+
"grad_norm": 0.33910053968429565,
|
| 659 |
+
"learning_rate": 9.50582011949595e-05,
|
| 660 |
+
"loss": 0.0329,
|
| 661 |
+
"step": 760
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.1568075117370893,
|
| 665 |
+
"grad_norm": 0.1620519608259201,
|
| 666 |
+
"learning_rate": 9.482814602330084e-05,
|
| 667 |
+
"loss": 0.0328,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.171830985915493,
|
| 672 |
+
"grad_norm": 0.3181590735912323,
|
| 673 |
+
"learning_rate": 9.459314861706435e-05,
|
| 674 |
+
"loss": 0.0363,
|
| 675 |
+
"step": 780
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.1868544600938966,
|
| 679 |
+
"grad_norm": 0.3142762780189514,
|
| 680 |
+
"learning_rate": 9.435323488435488e-05,
|
| 681 |
+
"loss": 0.0279,
|
| 682 |
+
"step": 790
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2018779342723005,
|
| 686 |
+
"grad_norm": 0.34480199217796326,
|
| 687 |
+
"learning_rate": 9.410843127529473e-05,
|
| 688 |
+
"loss": 0.0301,
|
| 689 |
+
"step": 800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2018779342723005,
|
| 693 |
+
"eval_loss": 0.03519855812191963,
|
| 694 |
+
"eval_runtime": 213.9318,
|
| 695 |
+
"eval_samples_per_second": 5.534,
|
| 696 |
+
"eval_steps_per_second": 5.534,
|
| 697 |
+
"step": 800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.2169014084507042,
|
| 701 |
+
"grad_norm": 0.3214387893676758,
|
| 702 |
+
"learning_rate": 9.385876477910765e-05,
|
| 703 |
+
"loss": 0.0317,
|
| 704 |
+
"step": 810
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.231924882629108,
|
| 708 |
+
"grad_norm": 0.20507577061653137,
|
| 709 |
+
"learning_rate": 9.360426292114314e-05,
|
| 710 |
+
"loss": 0.0291,
|
| 711 |
+
"step": 820
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.2469483568075117,
|
| 715 |
+
"grad_norm": 0.2585867941379547,
|
| 716 |
+
"learning_rate": 9.334495375984212e-05,
|
| 717 |
+
"loss": 0.0279,
|
| 718 |
+
"step": 830
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.2619718309859156,
|
| 722 |
+
"grad_norm": 0.21939821541309357,
|
| 723 |
+
"learning_rate": 9.30808658836432e-05,
|
| 724 |
+
"loss": 0.0293,
|
| 725 |
+
"step": 840
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.2769953051643192,
|
| 729 |
+
"grad_norm": 0.42434459924697876,
|
| 730 |
+
"learning_rate": 9.281202840783108e-05,
|
| 731 |
+
"loss": 0.0321,
|
| 732 |
+
"step": 850
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.2769953051643192,
|
| 736 |
+
"eval_loss": 0.034056928008794785,
|
| 737 |
+
"eval_runtime": 215.0317,
|
| 738 |
+
"eval_samples_per_second": 5.506,
|
| 739 |
+
"eval_steps_per_second": 5.506,
|
| 740 |
+
"step": 850
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.292018779342723,
|
| 744 |
+
"grad_norm": 0.22999003529548645,
|
| 745 |
+
"learning_rate": 9.253847097132655e-05,
|
| 746 |
+
"loss": 0.0243,
|
| 747 |
+
"step": 860
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.3070422535211268,
|
| 751 |
+
"grad_norm": 0.1908729076385498,
|
| 752 |
+
"learning_rate": 9.226022373341882e-05,
|
| 753 |
+
"loss": 0.0247,
|
| 754 |
+
"step": 870
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.3220657276995305,
|
| 758 |
+
"grad_norm": 0.16463080048561096,
|
| 759 |
+
"learning_rate": 9.19773173704406e-05,
|
| 760 |
+
"loss": 0.0245,
|
| 761 |
+
"step": 880
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.3370892018779343,
|
| 765 |
+
"grad_norm": 0.5442132949829102,
|
| 766 |
+
"learning_rate": 9.168978307238594e-05,
|
| 767 |
+
"loss": 0.032,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.352112676056338,
|
| 772 |
+
"grad_norm": 0.11218319833278656,
|
| 773 |
+
"learning_rate": 9.13976525394717e-05,
|
| 774 |
+
"loss": 0.0271,
|
| 775 |
+
"step": 900
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.352112676056338,
|
| 779 |
+
"eval_loss": 0.03430046886205673,
|
| 780 |
+
"eval_runtime": 215.5861,
|
| 781 |
+
"eval_samples_per_second": 5.492,
|
| 782 |
+
"eval_steps_per_second": 5.492,
|
| 783 |
+
"step": 900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.3671361502347419,
|
| 787 |
+
"grad_norm": 0.13387706875801086,
|
| 788 |
+
"learning_rate": 9.110095797864263e-05,
|
| 789 |
+
"loss": 0.0224,
|
| 790 |
+
"step": 910
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.3821596244131455,
|
| 794 |
+
"grad_norm": 0.20326955616474152,
|
| 795 |
+
"learning_rate": 9.079973210002051e-05,
|
| 796 |
+
"loss": 0.0318,
|
| 797 |
+
"step": 920
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.3971830985915492,
|
| 801 |
+
"grad_norm": 0.2666754424571991,
|
| 802 |
+
"learning_rate": 9.049400811329807e-05,
|
| 803 |
+
"loss": 0.0322,
|
| 804 |
+
"step": 930
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.412206572769953,
|
| 808 |
+
"grad_norm": 0.29467517137527466,
|
| 809 |
+
"learning_rate": 9.01838197240775e-05,
|
| 810 |
+
"loss": 0.0305,
|
| 811 |
+
"step": 940
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.4272300469483568,
|
| 815 |
+
"grad_norm": 0.27433130145072937,
|
| 816 |
+
"learning_rate": 8.986920113015461e-05,
|
| 817 |
+
"loss": 0.0351,
|
| 818 |
+
"step": 950
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.4272300469483568,
|
| 822 |
+
"eval_loss": 0.03298981115221977,
|
| 823 |
+
"eval_runtime": 216.0148,
|
| 824 |
+
"eval_samples_per_second": 5.481,
|
| 825 |
+
"eval_steps_per_second": 5.481,
|
| 826 |
+
"step": 950
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.4422535211267606,
|
| 830 |
+
"grad_norm": 0.23651893436908722,
|
| 831 |
+
"learning_rate": 8.955018701774846e-05,
|
| 832 |
+
"loss": 0.0262,
|
| 833 |
+
"step": 960
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.4572769953051643,
|
| 837 |
+
"grad_norm": 0.2778637111186981,
|
| 838 |
+
"learning_rate": 8.922681255767731e-05,
|
| 839 |
+
"loss": 0.0347,
|
| 840 |
+
"step": 970
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.4723004694835682,
|
| 844 |
+
"grad_norm": 0.18652495741844177,
|
| 845 |
+
"learning_rate": 8.889911340148112e-05,
|
| 846 |
+
"loss": 0.0308,
|
| 847 |
+
"step": 980
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 1.4873239436619718,
|
| 851 |
+
"grad_norm": 0.17595553398132324,
|
| 852 |
+
"learning_rate": 8.856712567749095e-05,
|
| 853 |
+
"loss": 0.0278,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.5023474178403755,
|
| 858 |
+
"grad_norm": 0.4016423523426056,
|
| 859 |
+
"learning_rate": 8.82308859868459e-05,
|
| 860 |
+
"loss": 0.0244,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 1.5023474178403755,
|
| 865 |
+
"eval_loss": 0.033029552549123764,
|
| 866 |
+
"eval_runtime": 217.6657,
|
| 867 |
+
"eval_samples_per_second": 5.44,
|
| 868 |
+
"eval_steps_per_second": 5.44,
|
| 869 |
+
"step": 1000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.5173708920187794,
|
| 873 |
+
"grad_norm": 0.2475142925977707,
|
| 874 |
+
"learning_rate": 8.789043139945795e-05,
|
| 875 |
+
"loss": 0.0298,
|
| 876 |
+
"step": 1010
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.532394366197183,
|
| 880 |
+
"grad_norm": 0.2059515416622162,
|
| 881 |
+
"learning_rate": 8.754579944992491e-05,
|
| 882 |
+
"loss": 0.036,
|
| 883 |
+
"step": 1020
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.5474178403755867,
|
| 887 |
+
"grad_norm": 0.16493824124336243,
|
| 888 |
+
"learning_rate": 8.719702813339248e-05,
|
| 889 |
+
"loss": 0.0288,
|
| 890 |
+
"step": 1030
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.5624413145539906,
|
| 894 |
+
"grad_norm": 0.20951515436172485,
|
| 895 |
+
"learning_rate": 8.684415590136518e-05,
|
| 896 |
+
"loss": 0.0277,
|
| 897 |
+
"step": 1040
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.5774647887323945,
|
| 901 |
+
"grad_norm": 0.0918218195438385,
|
| 902 |
+
"learning_rate": 8.648722165746722e-05,
|
| 903 |
+
"loss": 0.0277,
|
| 904 |
+
"step": 1050
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.5774647887323945,
|
| 908 |
+
"eval_loss": 0.034053802490234375,
|
| 909 |
+
"eval_runtime": 219.1262,
|
| 910 |
+
"eval_samples_per_second": 5.403,
|
| 911 |
+
"eval_steps_per_second": 5.403,
|
| 912 |
+
"step": 1050
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.5924882629107981,
|
| 916 |
+
"grad_norm": 0.09690333157777786,
|
| 917 |
+
"learning_rate": 8.61262647531534e-05,
|
| 918 |
+
"loss": 0.0251,
|
| 919 |
+
"step": 1060
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.6075117370892018,
|
| 923 |
+
"grad_norm": 0.19759806990623474,
|
| 924 |
+
"learning_rate": 8.576132498337068e-05,
|
| 925 |
+
"loss": 0.0362,
|
| 926 |
+
"step": 1070
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6225352112676057,
|
| 930 |
+
"grad_norm": 0.2693615257740021,
|
| 931 |
+
"learning_rate": 8.539244258217088e-05,
|
| 932 |
+
"loss": 0.0319,
|
| 933 |
+
"step": 1080
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.6375586854460094,
|
| 937 |
+
"grad_norm": 0.20964917540550232,
|
| 938 |
+
"learning_rate": 8.501965821827485e-05,
|
| 939 |
+
"loss": 0.032,
|
| 940 |
+
"step": 1090
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.652582159624413,
|
| 944 |
+
"grad_norm": 0.2406897395849228,
|
| 945 |
+
"learning_rate": 8.464301299058892e-05,
|
| 946 |
+
"loss": 0.0231,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.652582159624413,
|
| 951 |
+
"eval_loss": 0.034014929085969925,
|
| 952 |
+
"eval_runtime": 219.9896,
|
| 953 |
+
"eval_samples_per_second": 5.382,
|
| 954 |
+
"eval_steps_per_second": 5.382,
|
| 955 |
+
"step": 1100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.667605633802817,
|
| 959 |
+
"grad_norm": 0.25754356384277344,
|
| 960 |
+
"learning_rate": 8.426254842367374e-05,
|
| 961 |
+
"loss": 0.0258,
|
| 962 |
+
"step": 1110
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.6826291079812208,
|
| 966 |
+
"grad_norm": 0.2006315290927887,
|
| 967 |
+
"learning_rate": 8.387830646316623e-05,
|
| 968 |
+
"loss": 0.0334,
|
| 969 |
+
"step": 1120
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.6976525821596244,
|
| 973 |
+
"grad_norm": 0.24149750173091888,
|
| 974 |
+
"learning_rate": 8.349032947115525e-05,
|
| 975 |
+
"loss": 0.0249,
|
| 976 |
+
"step": 1130
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.712676056338028,
|
| 980 |
+
"grad_norm": 0.30418938398361206,
|
| 981 |
+
"learning_rate": 8.309866022151107e-05,
|
| 982 |
+
"loss": 0.0382,
|
| 983 |
+
"step": 1140
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.727699530516432,
|
| 987 |
+
"grad_norm": 0.18970516324043274,
|
| 988 |
+
"learning_rate": 8.270334189516983e-05,
|
| 989 |
+
"loss": 0.0261,
|
| 990 |
+
"step": 1150
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.727699530516432,
|
| 994 |
+
"eval_loss": 0.03269151970744133,
|
| 995 |
+
"eval_runtime": 220.3747,
|
| 996 |
+
"eval_samples_per_second": 5.373,
|
| 997 |
+
"eval_steps_per_second": 5.373,
|
| 998 |
+
"step": 1150
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.7427230046948357,
|
| 1002 |
+
"grad_norm": 0.11434226483106613,
|
| 1003 |
+
"learning_rate": 8.230441807537277e-05,
|
| 1004 |
+
"loss": 0.0393,
|
| 1005 |
+
"step": 1160
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.7577464788732393,
|
| 1009 |
+
"grad_norm": 0.18536998331546783,
|
| 1010 |
+
"learning_rate": 8.190193274286122e-05,
|
| 1011 |
+
"loss": 0.026,
|
| 1012 |
+
"step": 1170
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.7727699530516432,
|
| 1016 |
+
"grad_norm": 0.1809949427843094,
|
| 1017 |
+
"learning_rate": 8.149593027102789e-05,
|
| 1018 |
+
"loss": 0.0295,
|
| 1019 |
+
"step": 1180
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.787793427230047,
|
| 1023 |
+
"grad_norm": 0.2668783664703369,
|
| 1024 |
+
"learning_rate": 8.108645542102469e-05,
|
| 1025 |
+
"loss": 0.0255,
|
| 1026 |
+
"step": 1190
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.8028169014084507,
|
| 1030 |
+
"grad_norm": 0.19622312486171722,
|
| 1031 |
+
"learning_rate": 8.067355333682798e-05,
|
| 1032 |
+
"loss": 0.0297,
|
| 1033 |
+
"step": 1200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.8028169014084507,
|
| 1037 |
+
"eval_loss": 0.03482316806912422,
|
| 1038 |
+
"eval_runtime": 220.3204,
|
| 1039 |
+
"eval_samples_per_second": 5.374,
|
| 1040 |
+
"eval_steps_per_second": 5.374,
|
| 1041 |
+
"step": 1200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.8178403755868544,
|
| 1045 |
+
"grad_norm": 0.28984394669532776,
|
| 1046 |
+
"learning_rate": 8.025726954026138e-05,
|
| 1047 |
+
"loss": 0.0228,
|
| 1048 |
+
"step": 1210
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.8328638497652583,
|
| 1052 |
+
"grad_norm": 0.2442556470632553,
|
| 1053 |
+
"learning_rate": 7.983764992597716e-05,
|
| 1054 |
+
"loss": 0.0272,
|
| 1055 |
+
"step": 1220
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.847887323943662,
|
| 1059 |
+
"grad_norm": 0.33816081285476685,
|
| 1060 |
+
"learning_rate": 7.94147407563964e-05,
|
| 1061 |
+
"loss": 0.0296,
|
| 1062 |
+
"step": 1230
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.8629107981220656,
|
| 1066 |
+
"grad_norm": 0.3621574938297272,
|
| 1067 |
+
"learning_rate": 7.89885886566086e-05,
|
| 1068 |
+
"loss": 0.0333,
|
| 1069 |
+
"step": 1240
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 1.8779342723004695,
|
| 1073 |
+
"grad_norm": 0.27948886156082153,
|
| 1074 |
+
"learning_rate": 7.855924060923141e-05,
|
| 1075 |
+
"loss": 0.027,
|
| 1076 |
+
"step": 1250
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.8779342723004695,
|
| 1080 |
+
"eval_loss": 0.03339933604001999,
|
| 1081 |
+
"eval_runtime": 220.8261,
|
| 1082 |
+
"eval_samples_per_second": 5.362,
|
| 1083 |
+
"eval_steps_per_second": 5.362,
|
| 1084 |
+
"step": 1250
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.8929577464788734,
|
| 1088 |
+
"grad_norm": 0.19844475388526917,
|
| 1089 |
+
"learning_rate": 7.812674394923077e-05,
|
| 1090 |
+
"loss": 0.03,
|
| 1091 |
+
"step": 1260
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.907981220657277,
|
| 1095 |
+
"grad_norm": 0.17690996825695038,
|
| 1096 |
+
"learning_rate": 7.769114635870231e-05,
|
| 1097 |
+
"loss": 0.0275,
|
| 1098 |
+
"step": 1270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.9230046948356807,
|
| 1102 |
+
"grad_norm": 0.26047441363334656,
|
| 1103 |
+
"learning_rate": 7.725249586161463e-05,
|
| 1104 |
+
"loss": 0.0294,
|
| 1105 |
+
"step": 1280
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.9380281690140846,
|
| 1109 |
+
"grad_norm": 0.16973556578159332,
|
| 1110 |
+
"learning_rate": 7.68108408185145e-05,
|
| 1111 |
+
"loss": 0.031,
|
| 1112 |
+
"step": 1290
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.9530516431924883,
|
| 1116 |
+
"grad_norm": 0.18663664162158966,
|
| 1117 |
+
"learning_rate": 7.636622992119536e-05,
|
| 1118 |
+
"loss": 0.0417,
|
| 1119 |
+
"step": 1300
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.9530516431924883,
|
| 1123 |
+
"eval_loss": 0.03484388440847397,
|
| 1124 |
+
"eval_runtime": 221.5077,
|
| 1125 |
+
"eval_samples_per_second": 5.345,
|
| 1126 |
+
"eval_steps_per_second": 5.345,
|
| 1127 |
+
"step": 1300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 1.968075117370892,
|
| 1131 |
+
"grad_norm": 0.2905583083629608,
|
| 1132 |
+
"learning_rate": 7.591871218732902e-05,
|
| 1133 |
+
"loss": 0.0329,
|
| 1134 |
+
"step": 1310
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.9830985915492958,
|
| 1138 |
+
"grad_norm": 0.18613983690738678,
|
| 1139 |
+
"learning_rate": 7.54683369550616e-05,
|
| 1140 |
+
"loss": 0.0364,
|
| 1141 |
+
"step": 1320
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 1.9981220657276997,
|
| 1145 |
+
"grad_norm": 0.1276523619890213,
|
| 1146 |
+
"learning_rate": 7.501515387757404e-05,
|
| 1147 |
+
"loss": 0.0269,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.013145539906103,
|
| 1152 |
+
"grad_norm": 0.22376187145709991,
|
| 1153 |
+
"learning_rate": 7.455921291760796e-05,
|
| 1154 |
+
"loss": 0.0313,
|
| 1155 |
+
"step": 1340
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.028169014084507,
|
| 1159 |
+
"grad_norm": 0.09134189784526825,
|
| 1160 |
+
"learning_rate": 7.410056434195725e-05,
|
| 1161 |
+
"loss": 0.0173,
|
| 1162 |
+
"step": 1350
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 2.028169014084507,
|
| 1166 |
+
"eval_loss": 0.03278001397848129,
|
| 1167 |
+
"eval_runtime": 221.4183,
|
| 1168 |
+
"eval_samples_per_second": 5.347,
|
| 1169 |
+
"eval_steps_per_second": 5.347,
|
| 1170 |
+
"step": 1350
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 2.043192488262911,
|
| 1174 |
+
"grad_norm": 0.13673311471939087,
|
| 1175 |
+
"learning_rate": 7.363925871592629e-05,
|
| 1176 |
+
"loss": 0.0205,
|
| 1177 |
+
"step": 1360
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 2.0582159624413148,
|
| 1181 |
+
"grad_norm": 0.161691814661026,
|
| 1182 |
+
"learning_rate": 7.317534689775528e-05,
|
| 1183 |
+
"loss": 0.0201,
|
| 1184 |
+
"step": 1370
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.073239436619718,
|
| 1188 |
+
"grad_norm": 0.1258237212896347,
|
| 1189 |
+
"learning_rate": 7.270888003301304e-05,
|
| 1190 |
+
"loss": 0.0159,
|
| 1191 |
+
"step": 1380
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 2.088262910798122,
|
| 1195 |
+
"grad_norm": 0.29005277156829834,
|
| 1196 |
+
"learning_rate": 7.22399095489584e-05,
|
| 1197 |
+
"loss": 0.0222,
|
| 1198 |
+
"step": 1390
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 2.103286384976526,
|
| 1202 |
+
"grad_norm": 0.11337249726057053,
|
| 1203 |
+
"learning_rate": 7.176848714887042e-05,
|
| 1204 |
+
"loss": 0.0207,
|
| 1205 |
+
"step": 1400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 2.103286384976526,
|
| 1209 |
+
"eval_loss": 0.03226166218519211,
|
| 1210 |
+
"eval_runtime": 221.8527,
|
| 1211 |
+
"eval_samples_per_second": 5.337,
|
| 1212 |
+
"eval_steps_per_second": 5.337,
|
| 1213 |
+
"step": 1400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.1183098591549294,
|
| 1217 |
+
"grad_norm": 0.16560769081115723,
|
| 1218 |
+
"learning_rate": 7.129466480634806e-05,
|
| 1219 |
+
"loss": 0.0225,
|
| 1220 |
+
"step": 1410
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.1333333333333333,
|
| 1224 |
+
"grad_norm": 0.3505042493343353,
|
| 1225 |
+
"learning_rate": 7.081849475958042e-05,
|
| 1226 |
+
"loss": 0.0187,
|
| 1227 |
+
"step": 1420
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.148356807511737,
|
| 1231 |
+
"grad_norm": 0.16660155355930328,
|
| 1232 |
+
"learning_rate": 7.034002950558723e-05,
|
| 1233 |
+
"loss": 0.0185,
|
| 1234 |
+
"step": 1430
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.1633802816901406,
|
| 1238 |
+
"grad_norm": 0.13852401077747345,
|
| 1239 |
+
"learning_rate": 6.985932179443144e-05,
|
| 1240 |
+
"loss": 0.0192,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.1784037558685445,
|
| 1245 |
+
"grad_norm": 0.21416670083999634,
|
| 1246 |
+
"learning_rate": 6.937642462340342e-05,
|
| 1247 |
+
"loss": 0.0223,
|
| 1248 |
+
"step": 1450
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.1784037558685445,
|
| 1252 |
+
"eval_loss": 0.03254189342260361,
|
| 1253 |
+
"eval_runtime": 222.3283,
|
| 1254 |
+
"eval_samples_per_second": 5.325,
|
| 1255 |
+
"eval_steps_per_second": 5.325,
|
| 1256 |
+
"step": 1450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.1934272300469484,
|
| 1260 |
+
"grad_norm": 0.1465296745300293,
|
| 1261 |
+
"learning_rate": 6.889139123117817e-05,
|
| 1262 |
+
"loss": 0.0191,
|
| 1263 |
+
"step": 1460
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.2084507042253523,
|
| 1267 |
+
"grad_norm": 0.19770440459251404,
|
| 1268 |
+
"learning_rate": 6.840427509194575e-05,
|
| 1269 |
+
"loss": 0.0209,
|
| 1270 |
+
"step": 1470
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.2234741784037557,
|
| 1274 |
+
"grad_norm": 0.22087684273719788,
|
| 1275 |
+
"learning_rate": 6.791512990951597e-05,
|
| 1276 |
+
"loss": 0.0149,
|
| 1277 |
+
"step": 1480
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.2384976525821596,
|
| 1281 |
+
"grad_norm": 0.34373584389686584,
|
| 1282 |
+
"learning_rate": 6.74240096113975e-05,
|
| 1283 |
+
"loss": 0.0211,
|
| 1284 |
+
"step": 1490
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.2535211267605635,
|
| 1288 |
+
"grad_norm": 0.12940266728401184,
|
| 1289 |
+
"learning_rate": 6.693096834285256e-05,
|
| 1290 |
+
"loss": 0.0107,
|
| 1291 |
+
"step": 1500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.2535211267605635,
|
| 1295 |
+
"eval_loss": 0.035872478038072586,
|
| 1296 |
+
"eval_runtime": 223.0953,
|
| 1297 |
+
"eval_samples_per_second": 5.307,
|
| 1298 |
+
"eval_steps_per_second": 5.307,
|
| 1299 |
+
"step": 1500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.2685446009389674,
|
| 1303 |
+
"grad_norm": 0.39470982551574707,
|
| 1304 |
+
"learning_rate": 6.643606046092732e-05,
|
| 1305 |
+
"loss": 0.0226,
|
| 1306 |
+
"step": 1510
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.283568075117371,
|
| 1310 |
+
"grad_norm": 0.22834685444831848,
|
| 1311 |
+
"learning_rate": 6.593934052845929e-05,
|
| 1312 |
+
"loss": 0.0253,
|
| 1313 |
+
"step": 1520
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.2985915492957747,
|
| 1317 |
+
"grad_norm": 0.41707947850227356,
|
| 1318 |
+
"learning_rate": 6.544086330806181e-05,
|
| 1319 |
+
"loss": 0.0205,
|
| 1320 |
+
"step": 1530
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.3136150234741786,
|
| 1324 |
+
"grad_norm": 0.1707378327846527,
|
| 1325 |
+
"learning_rate": 6.494068375608646e-05,
|
| 1326 |
+
"loss": 0.0197,
|
| 1327 |
+
"step": 1540
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.328638497652582,
|
| 1331 |
+
"grad_norm": 0.17165277898311615,
|
| 1332 |
+
"learning_rate": 6.443885701656432e-05,
|
| 1333 |
+
"loss": 0.0182,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.328638497652582,
|
| 1338 |
+
"eval_loss": 0.03318268060684204,
|
| 1339 |
+
"eval_runtime": 223.282,
|
| 1340 |
+
"eval_samples_per_second": 5.303,
|
| 1341 |
+
"eval_steps_per_second": 5.303,
|
| 1342 |
+
"step": 1550
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.343661971830986,
|
| 1346 |
+
"grad_norm": 0.14184924960136414,
|
| 1347 |
+
"learning_rate": 6.393543841512632e-05,
|
| 1348 |
+
"loss": 0.02,
|
| 1349 |
+
"step": 1560
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.35868544600939,
|
| 1353 |
+
"grad_norm": 0.31733861565589905,
|
| 1354 |
+
"learning_rate": 6.343048345290386e-05,
|
| 1355 |
+
"loss": 0.0144,
|
| 1356 |
+
"step": 1570
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.3737089201877932,
|
| 1360 |
+
"grad_norm": 0.16232426464557648,
|
| 1361 |
+
"learning_rate": 6.292404780040961e-05,
|
| 1362 |
+
"loss": 0.023,
|
| 1363 |
+
"step": 1580
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.388732394366197,
|
| 1367 |
+
"grad_norm": 0.18042533099651337,
|
| 1368 |
+
"learning_rate": 6.241618729140018e-05,
|
| 1369 |
+
"loss": 0.0157,
|
| 1370 |
+
"step": 1590
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.403755868544601,
|
| 1374 |
+
"grad_norm": 0.4602644741535187,
|
| 1375 |
+
"learning_rate": 6.190695791672042e-05,
|
| 1376 |
+
"loss": 0.0187,
|
| 1377 |
+
"step": 1600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.403755868544601,
|
| 1381 |
+
"eval_loss": 0.032255951315164566,
|
| 1382 |
+
"eval_runtime": 222.5999,
|
| 1383 |
+
"eval_samples_per_second": 5.319,
|
| 1384 |
+
"eval_steps_per_second": 5.319,
|
| 1385 |
+
"step": 1600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.418779342723005,
|
| 1389 |
+
"grad_norm": 0.0981522649526596,
|
| 1390 |
+
"learning_rate": 6.139641581813052e-05,
|
| 1391 |
+
"loss": 0.0183,
|
| 1392 |
+
"step": 1610
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 2.4338028169014083,
|
| 1396 |
+
"grad_norm": 0.1687208116054535,
|
| 1397 |
+
"learning_rate": 6.088461728211642e-05,
|
| 1398 |
+
"loss": 0.0177,
|
| 1399 |
+
"step": 1620
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 2.448826291079812,
|
| 1403 |
+
"grad_norm": 0.19458502531051636,
|
| 1404 |
+
"learning_rate": 6.0371618733684474e-05,
|
| 1405 |
+
"loss": 0.0117,
|
| 1406 |
+
"step": 1630
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.463849765258216,
|
| 1410 |
+
"grad_norm": 0.12449292838573456,
|
| 1411 |
+
"learning_rate": 5.9857476730140485e-05,
|
| 1412 |
+
"loss": 0.0175,
|
| 1413 |
+
"step": 1640
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 2.4788732394366195,
|
| 1417 |
+
"grad_norm": 0.20214103162288666,
|
| 1418 |
+
"learning_rate": 5.9342247954854466e-05,
|
| 1419 |
+
"loss": 0.018,
|
| 1420 |
+
"step": 1650
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 2.4788732394366195,
|
| 1424 |
+
"eval_loss": 0.03270378336310387,
|
| 1425 |
+
"eval_runtime": 221.3125,
|
| 1426 |
+
"eval_samples_per_second": 5.35,
|
| 1427 |
+
"eval_steps_per_second": 5.35,
|
| 1428 |
+
"step": 1650
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.4938967136150234,
|
| 1432 |
+
"grad_norm": 0.5592941641807556,
|
| 1433 |
+
"learning_rate": 5.8825989211011335e-05,
|
| 1434 |
+
"loss": 0.022,
|
| 1435 |
+
"step": 1660
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 2.5089201877934273,
|
| 1439 |
+
"grad_norm": 0.20930063724517822,
|
| 1440 |
+
"learning_rate": 5.830875741534852e-05,
|
| 1441 |
+
"loss": 0.0159,
|
| 1442 |
+
"step": 1670
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.523943661971831,
|
| 1446 |
+
"grad_norm": 0.24804644286632538,
|
| 1447 |
+
"learning_rate": 5.7790609591880826e-05,
|
| 1448 |
+
"loss": 0.0239,
|
| 1449 |
+
"step": 1680
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.5389671361502346,
|
| 1453 |
+
"grad_norm": 0.17634353041648865,
|
| 1454 |
+
"learning_rate": 5.727160286561386e-05,
|
| 1455 |
+
"loss": 0.0224,
|
| 1456 |
+
"step": 1690
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.5539906103286385,
|
| 1460 |
+
"grad_norm": 0.23160183429718018,
|
| 1461 |
+
"learning_rate": 5.675179445624581e-05,
|
| 1462 |
+
"loss": 0.0205,
|
| 1463 |
+
"step": 1700
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.5539906103286385,
|
| 1467 |
+
"eval_loss": 0.03496889770030975,
|
| 1468 |
+
"eval_runtime": 221.4535,
|
| 1469 |
+
"eval_samples_per_second": 5.346,
|
| 1470 |
+
"eval_steps_per_second": 5.346,
|
| 1471 |
+
"step": 1700
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 2.5690140845070424,
|
| 1475 |
+
"grad_norm": 0.09332422912120819,
|
| 1476 |
+
"learning_rate": 5.62312416718593e-05,
|
| 1477 |
+
"loss": 0.0164,
|
| 1478 |
+
"step": 1710
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 2.584037558685446,
|
| 1482 |
+
"grad_norm": 0.09970489144325256,
|
| 1483 |
+
"learning_rate": 5.5710001902603116e-05,
|
| 1484 |
+
"loss": 0.0257,
|
| 1485 |
+
"step": 1720
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 2.5990610328638497,
|
| 1489 |
+
"grad_norm": 0.17599700391292572,
|
| 1490 |
+
"learning_rate": 5.5188132614365094e-05,
|
| 1491 |
+
"loss": 0.0198,
|
| 1492 |
+
"step": 1730
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 2.6140845070422536,
|
| 1496 |
+
"grad_norm": 0.11823566257953644,
|
| 1497 |
+
"learning_rate": 5.4665691342436565e-05,
|
| 1498 |
+
"loss": 0.0152,
|
| 1499 |
+
"step": 1740
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 2.629107981220657,
|
| 1503 |
+
"grad_norm": 0.31226423382759094,
|
| 1504 |
+
"learning_rate": 5.414273568516919e-05,
|
| 1505 |
+
"loss": 0.0182,
|
| 1506 |
+
"step": 1750
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 2.629107981220657,
|
| 1510 |
+
"eval_loss": 0.032272376120090485,
|
| 1511 |
+
"eval_runtime": 221.7599,
|
| 1512 |
+
"eval_samples_per_second": 5.339,
|
| 1513 |
+
"eval_steps_per_second": 5.339,
|
| 1514 |
+
"step": 1750
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.644131455399061,
|
| 1518 |
+
"grad_norm": 0.12536101043224335,
|
| 1519 |
+
"learning_rate": 5.361932329762481e-05,
|
| 1520 |
+
"loss": 0.0197,
|
| 1521 |
+
"step": 1760
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.659154929577465,
|
| 1525 |
+
"grad_norm": 0.11700065433979034,
|
| 1526 |
+
"learning_rate": 5.309551188521914e-05,
|
| 1527 |
+
"loss": 0.0197,
|
| 1528 |
+
"step": 1770
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.6741784037558687,
|
| 1532 |
+
"grad_norm": 0.30704525113105774,
|
| 1533 |
+
"learning_rate": 5.2571359197359704e-05,
|
| 1534 |
+
"loss": 0.0161,
|
| 1535 |
+
"step": 1780
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.6892018779342726,
|
| 1539 |
+
"grad_norm": 0.06998223811388016,
|
| 1540 |
+
"learning_rate": 5.2046923021079175e-05,
|
| 1541 |
+
"loss": 0.0151,
|
| 1542 |
+
"step": 1790
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.704225352112676,
|
| 1546 |
+
"grad_norm": 0.09223470091819763,
|
| 1547 |
+
"learning_rate": 5.1522261174664346e-05,
|
| 1548 |
+
"loss": 0.0202,
|
| 1549 |
+
"step": 1800
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.704225352112676,
|
| 1553 |
+
"eval_loss": 0.032467469573020935,
|
| 1554 |
+
"eval_runtime": 221.8494,
|
| 1555 |
+
"eval_samples_per_second": 5.337,
|
| 1556 |
+
"eval_steps_per_second": 5.337,
|
| 1557 |
+
"step": 1800
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.71924882629108,
|
| 1561 |
+
"grad_norm": 0.3536123037338257,
|
| 1562 |
+
"learning_rate": 5.0997431501281835e-05,
|
| 1563 |
+
"loss": 0.0262,
|
| 1564 |
+
"step": 1810
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.7342723004694838,
|
| 1568 |
+
"grad_norm": 0.2920973300933838,
|
| 1569 |
+
"learning_rate": 5.0472491862600915e-05,
|
| 1570 |
+
"loss": 0.0202,
|
| 1571 |
+
"step": 1820
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.749295774647887,
|
| 1575 |
+
"grad_norm": 0.23328785598278046,
|
| 1576 |
+
"learning_rate": 4.994750013241435e-05,
|
| 1577 |
+
"loss": 0.022,
|
| 1578 |
+
"step": 1830
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.764319248826291,
|
| 1582 |
+
"grad_norm": 0.13965994119644165,
|
| 1583 |
+
"learning_rate": 4.9422514190257974e-05,
|
| 1584 |
+
"loss": 0.0201,
|
| 1585 |
+
"step": 1840
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.779342723004695,
|
| 1589 |
+
"grad_norm": 0.12157478928565979,
|
| 1590 |
+
"learning_rate": 4.88975919150294e-05,
|
| 1591 |
+
"loss": 0.0218,
|
| 1592 |
+
"step": 1850
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.779342723004695,
|
| 1596 |
+
"eval_loss": 0.03232384845614433,
|
| 1597 |
+
"eval_runtime": 221.306,
|
| 1598 |
+
"eval_samples_per_second": 5.35,
|
| 1599 |
+
"eval_steps_per_second": 5.35,
|
| 1600 |
+
"step": 1850
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 2.7943661971830984,
|
| 1604 |
+
"grad_norm": 0.17367467284202576,
|
| 1605 |
+
"learning_rate": 4.83727911786071e-05,
|
| 1606 |
+
"loss": 0.022,
|
| 1607 |
+
"step": 1860
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.8093896713615023,
|
| 1611 |
+
"grad_norm": 0.30343097448349,
|
| 1612 |
+
"learning_rate": 4.7848169839470145e-05,
|
| 1613 |
+
"loss": 0.0224,
|
| 1614 |
+
"step": 1870
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.824413145539906,
|
| 1618 |
+
"grad_norm": 0.12888950109481812,
|
| 1619 |
+
"learning_rate": 4.7323785736319244e-05,
|
| 1620 |
+
"loss": 0.021,
|
| 1621 |
+
"step": 1880
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.8394366197183096,
|
| 1625 |
+
"grad_norm": 0.15008972585201263,
|
| 1626 |
+
"learning_rate": 4.679969668170024e-05,
|
| 1627 |
+
"loss": 0.0232,
|
| 1628 |
+
"step": 1890
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.8544600938967135,
|
| 1632 |
+
"grad_norm": 0.10453104227781296,
|
| 1633 |
+
"learning_rate": 4.627596045563031e-05,
|
| 1634 |
+
"loss": 0.0179,
|
| 1635 |
+
"step": 1900
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 2.8544600938967135,
|
| 1639 |
+
"eval_loss": 0.03188331797719002,
|
| 1640 |
+
"eval_runtime": 221.3325,
|
| 1641 |
+
"eval_samples_per_second": 5.349,
|
| 1642 |
+
"eval_steps_per_second": 5.349,
|
| 1643 |
+
"step": 1900
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.8694835680751174,
|
| 1647 |
+
"grad_norm": 0.20146887004375458,
|
| 1648 |
+
"learning_rate": 4.575263479922783e-05,
|
| 1649 |
+
"loss": 0.0147,
|
| 1650 |
+
"step": 1910
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.8845070422535213,
|
| 1654 |
+
"grad_norm": 0.10789839178323746,
|
| 1655 |
+
"learning_rate": 4.522977740834651e-05,
|
| 1656 |
+
"loss": 0.0118,
|
| 1657 |
+
"step": 1920
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 2.8995305164319247,
|
| 1661 |
+
"grad_norm": 0.1628926396369934,
|
| 1662 |
+
"learning_rate": 4.4707445927214456e-05,
|
| 1663 |
+
"loss": 0.0126,
|
| 1664 |
+
"step": 1930
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.9145539906103286,
|
| 1668 |
+
"grad_norm": 0.14242199063301086,
|
| 1669 |
+
"learning_rate": 4.4185697942079115e-05,
|
| 1670 |
+
"loss": 0.0199,
|
| 1671 |
+
"step": 1940
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 2.9295774647887325,
|
| 1675 |
+
"grad_norm": 0.15721061825752258,
|
| 1676 |
+
"learning_rate": 4.366459097485832e-05,
|
| 1677 |
+
"loss": 0.0213,
|
| 1678 |
+
"step": 1950
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.9295774647887325,
|
| 1682 |
+
"eval_loss": 0.03299366310238838,
|
| 1683 |
+
"eval_runtime": 221.4895,
|
| 1684 |
+
"eval_samples_per_second": 5.346,
|
| 1685 |
+
"eval_steps_per_second": 5.346,
|
| 1686 |
+
"step": 1950
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 2.9446009389671364,
|
| 1690 |
+
"grad_norm": 0.2670339047908783,
|
| 1691 |
+
"learning_rate": 4.314418247679866e-05,
|
| 1692 |
+
"loss": 0.0231,
|
| 1693 |
+
"step": 1960
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 2.95962441314554,
|
| 1697 |
+
"grad_norm": 0.18777424097061157,
|
| 1698 |
+
"learning_rate": 4.26245298221416e-05,
|
| 1699 |
+
"loss": 0.0141,
|
| 1700 |
+
"step": 1970
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.9746478873239437,
|
| 1704 |
+
"grad_norm": 0.08959437906742096,
|
| 1705 |
+
"learning_rate": 4.2105690301798014e-05,
|
| 1706 |
+
"loss": 0.0143,
|
| 1707 |
+
"step": 1980
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 2.9896713615023476,
|
| 1711 |
+
"grad_norm": 0.23774972558021545,
|
| 1712 |
+
"learning_rate": 4.158772111703194e-05,
|
| 1713 |
+
"loss": 0.0167,
|
| 1714 |
+
"step": 1990
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 3.004694835680751,
|
| 1718 |
+
"grad_norm": 0.025070570409297943,
|
| 1719 |
+
"learning_rate": 4.107067937315429e-05,
|
| 1720 |
+
"loss": 0.0104,
|
| 1721 |
+
"step": 2000
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 3.004694835680751,
|
| 1725 |
+
"eval_loss": 0.03279737010598183,
|
| 1726 |
+
"eval_runtime": 222.0291,
|
| 1727 |
+
"eval_samples_per_second": 5.333,
|
| 1728 |
+
"eval_steps_per_second": 5.333,
|
| 1729 |
+
"step": 2000
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 3.019718309859155,
|
| 1733 |
+
"grad_norm": 0.12388351559638977,
|
| 1734 |
+
"learning_rate": 4.055462207322698e-05,
|
| 1735 |
+
"loss": 0.01,
|
| 1736 |
+
"step": 2010
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 3.034741784037559,
|
| 1740 |
+
"grad_norm": 0.024688435718417168,
|
| 1741 |
+
"learning_rate": 4.003960611177855e-05,
|
| 1742 |
+
"loss": 0.0101,
|
| 1743 |
+
"step": 2020
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 3.0497652582159622,
|
| 1747 |
+
"grad_norm": 0.24434754252433777,
|
| 1748 |
+
"learning_rate": 3.952568826853152e-05,
|
| 1749 |
+
"loss": 0.0094,
|
| 1750 |
+
"step": 2030
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 3.064788732394366,
|
| 1754 |
+
"grad_norm": 0.05700072646141052,
|
| 1755 |
+
"learning_rate": 3.901292520214256e-05,
|
| 1756 |
+
"loss": 0.0068,
|
| 1757 |
+
"step": 2040
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 3.07981220657277,
|
| 1761 |
+
"grad_norm": 0.6166403889656067,
|
| 1762 |
+
"learning_rate": 3.850137344395598e-05,
|
| 1763 |
+
"loss": 0.0097,
|
| 1764 |
+
"step": 2050
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 3.07981220657277,
|
| 1768 |
+
"eval_loss": 0.035948336124420166,
|
| 1769 |
+
"eval_runtime": 221.527,
|
| 1770 |
+
"eval_samples_per_second": 5.345,
|
| 1771 |
+
"eval_steps_per_second": 5.345,
|
| 1772 |
+
"step": 2050
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 3.094835680751174,
|
| 1776 |
+
"grad_norm": 0.2868582308292389,
|
| 1777 |
+
"learning_rate": 3.799108939177118e-05,
|
| 1778 |
+
"loss": 0.0076,
|
| 1779 |
+
"step": 2060
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.1098591549295773,
|
| 1783 |
+
"grad_norm": 0.201212540268898,
|
| 1784 |
+
"learning_rate": 3.7482129303624934e-05,
|
| 1785 |
+
"loss": 0.0148,
|
| 1786 |
+
"step": 2070
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.124882629107981,
|
| 1790 |
+
"grad_norm": 0.512485921382904,
|
| 1791 |
+
"learning_rate": 3.697454929158901e-05,
|
| 1792 |
+
"loss": 0.0115,
|
| 1793 |
+
"step": 2080
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 3.139906103286385,
|
| 1797 |
+
"grad_norm": 0.23441621661186218,
|
| 1798 |
+
"learning_rate": 3.6468405315583854e-05,
|
| 1799 |
+
"loss": 0.0147,
|
| 1800 |
+
"step": 2090
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 3.1549295774647885,
|
| 1804 |
+
"grad_norm": 0.20293329656124115,
|
| 1805 |
+
"learning_rate": 3.59637531772092e-05,
|
| 1806 |
+
"loss": 0.0103,
|
| 1807 |
+
"step": 2100
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 3.1549295774647885,
|
| 1811 |
+
"eval_loss": 0.0363353006541729,
|
| 1812 |
+
"eval_runtime": 221.0841,
|
| 1813 |
+
"eval_samples_per_second": 5.355,
|
| 1814 |
+
"eval_steps_per_second": 5.355,
|
| 1815 |
+
"step": 2100
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 3.1699530516431924,
|
| 1819 |
+
"grad_norm": 0.2901355028152466,
|
| 1820 |
+
"learning_rate": 3.546064851359192e-05,
|
| 1821 |
+
"loss": 0.0082,
|
| 1822 |
+
"step": 2110
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 3.1849765258215963,
|
| 1826 |
+
"grad_norm": 0.11046724766492844,
|
| 1827 |
+
"learning_rate": 3.495914679125212e-05,
|
| 1828 |
+
"loss": 0.0097,
|
| 1829 |
+
"step": 2120
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 3.2,
|
| 1833 |
+
"grad_norm": 0.07721813023090363,
|
| 1834 |
+
"learning_rate": 3.445930329998819e-05,
|
| 1835 |
+
"loss": 0.0103,
|
| 1836 |
+
"step": 2130
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 3.2150234741784036,
|
| 1840 |
+
"grad_norm": 0.26007962226867676,
|
| 1841 |
+
"learning_rate": 3.396117314678097e-05,
|
| 1842 |
+
"loss": 0.0142,
|
| 1843 |
+
"step": 2140
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 3.2300469483568075,
|
| 1847 |
+
"grad_norm": 0.4068683087825775,
|
| 1848 |
+
"learning_rate": 3.3464811249718474e-05,
|
| 1849 |
+
"loss": 0.0131,
|
| 1850 |
+
"step": 2150
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 3.2300469483568075,
|
| 1854 |
+
"eval_loss": 0.03587425872683525,
|
| 1855 |
+
"eval_runtime": 221.6748,
|
| 1856 |
+
"eval_samples_per_second": 5.341,
|
| 1857 |
+
"eval_steps_per_second": 5.341,
|
| 1858 |
+
"step": 2150
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 3.2450704225352114,
|
| 1862 |
+
"grad_norm": 0.18533091247081757,
|
| 1863 |
+
"learning_rate": 3.297027233194114e-05,
|
| 1864 |
+
"loss": 0.0126,
|
| 1865 |
+
"step": 2160
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 3.260093896713615,
|
| 1869 |
+
"grad_norm": 0.10798130184412003,
|
| 1870 |
+
"learning_rate": 3.2477610915608704e-05,
|
| 1871 |
+
"loss": 0.0081,
|
| 1872 |
+
"step": 2170
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 3.2751173708920187,
|
| 1876 |
+
"grad_norm": 0.19456113874912262,
|
| 1877 |
+
"learning_rate": 3.1986881315889315e-05,
|
| 1878 |
+
"loss": 0.0091,
|
| 1879 |
+
"step": 2180
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 3.2901408450704226,
|
| 1883 |
+
"grad_norm": 0.26297393441200256,
|
| 1884 |
+
"learning_rate": 3.149813763497124e-05,
|
| 1885 |
+
"loss": 0.0155,
|
| 1886 |
+
"step": 2190
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 3.3051643192488265,
|
| 1890 |
+
"grad_norm": 0.26686403155326843,
|
| 1891 |
+
"learning_rate": 3.101143375609818e-05,
|
| 1892 |
+
"loss": 0.0149,
|
| 1893 |
+
"step": 2200
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 3.3051643192488265,
|
| 1897 |
+
"eval_loss": 0.036220405250787735,
|
| 1898 |
+
"eval_runtime": 221.9816,
|
| 1899 |
+
"eval_samples_per_second": 5.334,
|
| 1900 |
+
"eval_steps_per_second": 5.334,
|
| 1901 |
+
"step": 2200
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 3.32018779342723,
|
| 1905 |
+
"grad_norm": 0.3246915638446808,
|
| 1906 |
+
"learning_rate": 3.0526823337628915e-05,
|
| 1907 |
+
"loss": 0.0129,
|
| 1908 |
+
"step": 2210
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 3.335211267605634,
|
| 1912 |
+
"grad_norm": 0.4109603464603424,
|
| 1913 |
+
"learning_rate": 3.004435980712129e-05,
|
| 1914 |
+
"loss": 0.0098,
|
| 1915 |
+
"step": 2220
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 3.3502347417840377,
|
| 1919 |
+
"grad_norm": 0.4035777747631073,
|
| 1920 |
+
"learning_rate": 2.9564096355442116e-05,
|
| 1921 |
+
"loss": 0.0128,
|
| 1922 |
+
"step": 2230
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 3.365258215962441,
|
| 1926 |
+
"grad_norm": 0.1403612494468689,
|
| 1927 |
+
"learning_rate": 2.9086085930902824e-05,
|
| 1928 |
+
"loss": 0.0137,
|
| 1929 |
+
"step": 2240
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 3.380281690140845,
|
| 1933 |
+
"grad_norm": 0.21675720810890198,
|
| 1934 |
+
"learning_rate": 2.8610381233422058e-05,
|
| 1935 |
+
"loss": 0.0083,
|
| 1936 |
+
"step": 2250
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 3.380281690140845,
|
| 1940 |
+
"eval_loss": 0.03651123493909836,
|
| 1941 |
+
"eval_runtime": 222.0068,
|
| 1942 |
+
"eval_samples_per_second": 5.333,
|
| 1943 |
+
"eval_steps_per_second": 5.333,
|
| 1944 |
+
"step": 2250
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 3.395305164319249,
|
| 1948 |
+
"grad_norm": 0.30155134201049805,
|
| 1949 |
+
"learning_rate": 2.8137034708715592e-05,
|
| 1950 |
+
"loss": 0.0098,
|
| 1951 |
+
"step": 2260
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 3.4103286384976528,
|
| 1955 |
+
"grad_norm": 0.05509026348590851,
|
| 1956 |
+
"learning_rate": 2.7666098542514273e-05,
|
| 1957 |
+
"loss": 0.0096,
|
| 1958 |
+
"step": 2270
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 3.425352112676056,
|
| 1962 |
+
"grad_norm": 0.25339803099632263,
|
| 1963 |
+
"learning_rate": 2.719762465481055e-05,
|
| 1964 |
+
"loss": 0.0123,
|
| 1965 |
+
"step": 2280
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 3.44037558685446,
|
| 1969 |
+
"grad_norm": 0.052862461656332016,
|
| 1970 |
+
"learning_rate": 2.6731664694134473e-05,
|
| 1971 |
+
"loss": 0.0118,
|
| 1972 |
+
"step": 2290
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 3.455399061032864,
|
| 1976 |
+
"grad_norm": 0.16749408841133118,
|
| 1977 |
+
"learning_rate": 2.6268270031859476e-05,
|
| 1978 |
+
"loss": 0.0115,
|
| 1979 |
+
"step": 2300
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 3.455399061032864,
|
| 1983 |
+
"eval_loss": 0.035897351801395416,
|
| 1984 |
+
"eval_runtime": 221.6066,
|
| 1985 |
+
"eval_samples_per_second": 5.343,
|
| 1986 |
+
"eval_steps_per_second": 5.343,
|
| 1987 |
+
"step": 2300
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 3.4704225352112674,
|
| 1991 |
+
"grad_norm": 0.28338125348091125,
|
| 1992 |
+
"learning_rate": 2.580749175653877e-05,
|
| 1993 |
+
"loss": 0.0097,
|
| 1994 |
+
"step": 2310
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 3.4854460093896713,
|
| 1998 |
+
"grad_norm": 0.36096012592315674,
|
| 1999 |
+
"learning_rate": 2.5349380668272905e-05,
|
| 2000 |
+
"loss": 0.0122,
|
| 2001 |
+
"step": 2320
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 3.500469483568075,
|
| 2005 |
+
"grad_norm": 0.20076704025268555,
|
| 2006 |
+
"learning_rate": 2.489398727310908e-05,
|
| 2007 |
+
"loss": 0.0096,
|
| 2008 |
+
"step": 2330
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 3.5154929577464786,
|
| 2012 |
+
"grad_norm": 0.28074732422828674,
|
| 2013 |
+
"learning_rate": 2.4441361777473066e-05,
|
| 2014 |
+
"loss": 0.0088,
|
| 2015 |
+
"step": 2340
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 3.5305164319248825,
|
| 2019 |
+
"grad_norm": 0.5027658343315125,
|
| 2020 |
+
"learning_rate": 2.3991554082633912e-05,
|
| 2021 |
+
"loss": 0.0111,
|
| 2022 |
+
"step": 2350
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 3.5305164319248825,
|
| 2026 |
+
"eval_loss": 0.038713712245225906,
|
| 2027 |
+
"eval_runtime": 221.2043,
|
| 2028 |
+
"eval_samples_per_second": 5.353,
|
| 2029 |
+
"eval_steps_per_second": 5.353,
|
| 2030 |
+
"step": 2350
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 3.5455399061032864,
|
| 2034 |
+
"grad_norm": 0.11489449441432953,
|
| 2035 |
+
"learning_rate": 2.354461377920239e-05,
|
| 2036 |
+
"loss": 0.009,
|
| 2037 |
+
"step": 2360
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 3.5605633802816903,
|
| 2041 |
+
"grad_norm": 0.42795881628990173,
|
| 2042 |
+
"learning_rate": 2.3100590141663807e-05,
|
| 2043 |
+
"loss": 0.0123,
|
| 2044 |
+
"step": 2370
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 3.575586854460094,
|
| 2048 |
+
"grad_norm": 0.34739652276039124,
|
| 2049 |
+
"learning_rate": 2.265953212294551e-05,
|
| 2050 |
+
"loss": 0.0143,
|
| 2051 |
+
"step": 2380
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 3.5906103286384976,
|
| 2055 |
+
"grad_norm": 0.5267793536186218,
|
| 2056 |
+
"learning_rate": 2.2221488349019903e-05,
|
| 2057 |
+
"loss": 0.015,
|
| 2058 |
+
"step": 2390
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 3.6056338028169015,
|
| 2062 |
+
"grad_norm": 0.2632468342781067,
|
| 2063 |
+
"learning_rate": 2.1786507113543457e-05,
|
| 2064 |
+
"loss": 0.0094,
|
| 2065 |
+
"step": 2400
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 3.6056338028169015,
|
| 2069 |
+
"eval_loss": 0.03762707859277725,
|
| 2070 |
+
"eval_runtime": 222.4043,
|
| 2071 |
+
"eval_samples_per_second": 5.324,
|
| 2072 |
+
"eval_steps_per_second": 5.324,
|
| 2073 |
+
"step": 2400
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 3.6206572769953054,
|
| 2077 |
+
"grad_norm": 0.15667402744293213,
|
| 2078 |
+
"learning_rate": 2.1354636372532523e-05,
|
| 2079 |
+
"loss": 0.0149,
|
| 2080 |
+
"step": 2410
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 3.635680751173709,
|
| 2084 |
+
"grad_norm": 0.22469666600227356,
|
| 2085 |
+
"learning_rate": 2.092592373907617e-05,
|
| 2086 |
+
"loss": 0.0106,
|
| 2087 |
+
"step": 2420
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 3.6507042253521127,
|
| 2091 |
+
"grad_norm": 0.2074466347694397,
|
| 2092 |
+
"learning_rate": 2.0500416478086932e-05,
|
| 2093 |
+
"loss": 0.0079,
|
| 2094 |
+
"step": 2430
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 3.6657276995305166,
|
| 2098 |
+
"grad_norm": 0.21211783587932587,
|
| 2099 |
+
"learning_rate": 2.0078161501089954e-05,
|
| 2100 |
+
"loss": 0.0106,
|
| 2101 |
+
"step": 2440
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 3.68075117370892,
|
| 2105 |
+
"grad_norm": 0.13224627077579498,
|
| 2106 |
+
"learning_rate": 1.9659205361050982e-05,
|
| 2107 |
+
"loss": 0.0051,
|
| 2108 |
+
"step": 2450
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 3.68075117370892,
|
| 2112 |
+
"eval_loss": 0.03760899230837822,
|
| 2113 |
+
"eval_runtime": 222.6155,
|
| 2114 |
+
"eval_samples_per_second": 5.319,
|
| 2115 |
+
"eval_steps_per_second": 5.319,
|
| 2116 |
+
"step": 2450
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.695774647887324,
|
| 2120 |
+
"grad_norm": 0.07832111418247223,
|
| 2121 |
+
"learning_rate": 1.924359424724408e-05,
|
| 2122 |
+
"loss": 0.0143,
|
| 2123 |
+
"step": 2460
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 3.710798122065728,
|
| 2127 |
+
"grad_norm": 0.34279316663742065,
|
| 2128 |
+
"learning_rate": 1.8831373980159296e-05,
|
| 2129 |
+
"loss": 0.0129,
|
| 2130 |
+
"step": 2470
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 3.7258215962441312,
|
| 2134 |
+
"grad_norm": 0.13393940031528473,
|
| 2135 |
+
"learning_rate": 1.8422590006450947e-05,
|
| 2136 |
+
"loss": 0.0062,
|
| 2137 |
+
"step": 2480
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 3.740845070422535,
|
| 2141 |
+
"grad_norm": 0.05350172892212868,
|
| 2142 |
+
"learning_rate": 1.801728739392731e-05,
|
| 2143 |
+
"loss": 0.0087,
|
| 2144 |
+
"step": 2490
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.755868544600939,
|
| 2148 |
+
"grad_norm": 0.0817003846168518,
|
| 2149 |
+
"learning_rate": 1.7615510826581904e-05,
|
| 2150 |
+
"loss": 0.0053,
|
| 2151 |
+
"step": 2500
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 3.755868544600939,
|
| 2155 |
+
"eval_loss": 0.03752273693680763,
|
| 2156 |
+
"eval_runtime": 222.372,
|
| 2157 |
+
"eval_samples_per_second": 5.324,
|
| 2158 |
+
"eval_steps_per_second": 5.324,
|
| 2159 |
+
"step": 2500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 3.770892018779343,
|
| 2163 |
+
"grad_norm": 0.21435320377349854,
|
| 2164 |
+
"learning_rate": 1.7217304599667146e-05,
|
| 2165 |
+
"loss": 0.0111,
|
| 2166 |
+
"step": 2510
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 3.7859154929577463,
|
| 2170 |
+
"grad_norm": 0.15997354686260223,
|
| 2171 |
+
"learning_rate": 1.6822712614810893e-05,
|
| 2172 |
+
"loss": 0.0101,
|
| 2173 |
+
"step": 2520
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 3.80093896713615,
|
| 2177 |
+
"grad_norm": 0.182434543967247,
|
| 2178 |
+
"learning_rate": 1.643177837517631e-05,
|
| 2179 |
+
"loss": 0.0093,
|
| 2180 |
+
"step": 2530
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 3.815962441314554,
|
| 2184 |
+
"grad_norm": 0.22743846476078033,
|
| 2185 |
+
"learning_rate": 1.6044544980665767e-05,
|
| 2186 |
+
"loss": 0.0125,
|
| 2187 |
+
"step": 2540
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 3.830985915492958,
|
| 2191 |
+
"grad_norm": 0.33443447947502136,
|
| 2192 |
+
"learning_rate": 1.5661055123169126e-05,
|
| 2193 |
+
"loss": 0.0078,
|
| 2194 |
+
"step": 2550
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 3.830985915492958,
|
| 2198 |
+
"eval_loss": 0.037725865840911865,
|
| 2199 |
+
"eval_runtime": 221.8726,
|
| 2200 |
+
"eval_samples_per_second": 5.336,
|
| 2201 |
+
"eval_steps_per_second": 5.336,
|
| 2202 |
+
"step": 2550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 3.8460093896713614,
|
| 2206 |
+
"grad_norm": 0.5252001285552979,
|
| 2207 |
+
"learning_rate": 1.5281351081856974e-05,
|
| 2208 |
+
"loss": 0.0118,
|
| 2209 |
+
"step": 2560
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 3.8610328638497653,
|
| 2213 |
+
"grad_norm": 0.17551793158054352,
|
| 2214 |
+
"learning_rate": 1.4905474718519491e-05,
|
| 2215 |
+
"loss": 0.007,
|
| 2216 |
+
"step": 2570
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 3.876056338028169,
|
| 2220 |
+
"grad_norm": 0.1480369120836258,
|
| 2221 |
+
"learning_rate": 1.453346747295119e-05,
|
| 2222 |
+
"loss": 0.0141,
|
| 2223 |
+
"step": 2580
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 3.8910798122065726,
|
| 2227 |
+
"grad_norm": 0.19130997359752655,
|
| 2228 |
+
"learning_rate": 1.4165370358382274e-05,
|
| 2229 |
+
"loss": 0.0093,
|
| 2230 |
+
"step": 2590
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 3.9061032863849765,
|
| 2234 |
+
"grad_norm": 0.1769220381975174,
|
| 2235 |
+
"learning_rate": 1.3801223956956994e-05,
|
| 2236 |
+
"loss": 0.0105,
|
| 2237 |
+
"step": 2600
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 3.9061032863849765,
|
| 2241 |
+
"eval_loss": 0.0372486487030983,
|
| 2242 |
+
"eval_runtime": 221.9339,
|
| 2243 |
+
"eval_samples_per_second": 5.335,
|
| 2244 |
+
"eval_steps_per_second": 5.335,
|
| 2245 |
+
"step": 2600
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 3.9211267605633804,
|
| 2249 |
+
"grad_norm": 0.11678332090377808,
|
| 2250 |
+
"learning_rate": 1.344106841525946e-05,
|
| 2251 |
+
"loss": 0.0141,
|
| 2252 |
+
"step": 2610
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 3.936150234741784,
|
| 2256 |
+
"grad_norm": 0.24193385243415833,
|
| 2257 |
+
"learning_rate": 1.3084943439887659e-05,
|
| 2258 |
+
"loss": 0.0085,
|
| 2259 |
+
"step": 2620
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 3.9511737089201877,
|
| 2263 |
+
"grad_norm": 0.10747163742780685,
|
| 2264 |
+
"learning_rate": 1.273288829307579e-05,
|
| 2265 |
+
"loss": 0.0054,
|
| 2266 |
+
"step": 2630
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 3.9661971830985916,
|
| 2270 |
+
"grad_norm": 0.28762826323509216,
|
| 2271 |
+
"learning_rate": 1.2384941788365622e-05,
|
| 2272 |
+
"loss": 0.0078,
|
| 2273 |
+
"step": 2640
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 3.981220657276995,
|
| 2277 |
+
"grad_norm": 0.11588622629642487,
|
| 2278 |
+
"learning_rate": 1.2041142286327477e-05,
|
| 2279 |
+
"loss": 0.0105,
|
| 2280 |
+
"step": 2650
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 3.981220657276995,
|
| 2284 |
+
"eval_loss": 0.03705061599612236,
|
| 2285 |
+
"eval_runtime": 222.758,
|
| 2286 |
+
"eval_samples_per_second": 5.315,
|
| 2287 |
+
"eval_steps_per_second": 5.315,
|
| 2288 |
+
"step": 2650
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 3.996244131455399,
|
| 2292 |
+
"grad_norm": 0.2062222808599472,
|
| 2293 |
+
"learning_rate": 1.170152769033095e-05,
|
| 2294 |
+
"loss": 0.0085,
|
| 2295 |
+
"step": 2660
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"epoch": 4.011267605633803,
|
| 2299 |
+
"grad_norm": 0.15399198234081268,
|
| 2300 |
+
"learning_rate": 1.1366135442366127e-05,
|
| 2301 |
+
"loss": 0.0057,
|
| 2302 |
+
"step": 2670
|
| 2303 |
+
},
|
| 2304 |
+
{
|
| 2305 |
+
"epoch": 4.026291079812206,
|
| 2306 |
+
"grad_norm": 0.12908077239990234,
|
| 2307 |
+
"learning_rate": 1.103500251891571e-05,
|
| 2308 |
+
"loss": 0.0074,
|
| 2309 |
+
"step": 2680
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 4.041314553990611,
|
| 2313 |
+
"grad_norm": 0.06553179025650024,
|
| 2314 |
+
"learning_rate": 1.0708165426878325e-05,
|
| 2315 |
+
"loss": 0.0049,
|
| 2316 |
+
"step": 2690
|
| 2317 |
+
},
|
| 2318 |
+
{
|
| 2319 |
+
"epoch": 4.056338028169014,
|
| 2320 |
+
"grad_norm": 0.12032724171876907,
|
| 2321 |
+
"learning_rate": 1.0385660199543812e-05,
|
| 2322 |
+
"loss": 0.0064,
|
| 2323 |
+
"step": 2700
|
| 2324 |
+
},
|
| 2325 |
+
{
|
| 2326 |
+
"epoch": 4.056338028169014,
|
| 2327 |
+
"eval_loss": 0.038154859095811844,
|
| 2328 |
+
"eval_runtime": 220.5195,
|
| 2329 |
+
"eval_samples_per_second": 5.369,
|
| 2330 |
+
"eval_steps_per_second": 5.369,
|
| 2331 |
+
"step": 2700
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 4.0713615023474174,
|
| 2335 |
+
"grad_norm": 0.06930918991565704,
|
| 2336 |
+
"learning_rate": 1.0067522392620537e-05,
|
| 2337 |
+
"loss": 0.0052,
|
| 2338 |
+
"step": 2710
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 4.086384976525822,
|
| 2342 |
+
"grad_norm": 0.10133420675992966,
|
| 2343 |
+
"learning_rate": 9.753787080315385e-06,
|
| 2344 |
+
"loss": 0.0075,
|
| 2345 |
+
"step": 2720
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 4.101408450704225,
|
| 2349 |
+
"grad_norm": 0.01755693554878235,
|
| 2350 |
+
"learning_rate": 9.444488851467042e-06,
|
| 2351 |
+
"loss": 0.0059,
|
| 2352 |
+
"step": 2730
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 4.1164319248826295,
|
| 2356 |
+
"grad_norm": 0.1588931679725647,
|
| 2357 |
+
"learning_rate": 9.139661805732435e-06,
|
| 2358 |
+
"loss": 0.0042,
|
| 2359 |
+
"step": 2740
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 4.131455399061033,
|
| 2363 |
+
"grad_norm": 0.39881432056427,
|
| 2364 |
+
"learning_rate": 8.839339549827397e-06,
|
| 2365 |
+
"loss": 0.0048,
|
| 2366 |
+
"step": 2750
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 4.131455399061033,
|
| 2370 |
+
"eval_loss": 0.03982308879494667,
|
| 2371 |
+
"eval_runtime": 221.9797,
|
| 2372 |
+
"eval_samples_per_second": 5.334,
|
| 2373 |
+
"eval_steps_per_second": 5.334,
|
| 2374 |
+
"step": 2750
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 4.146478873239436,
|
| 2378 |
+
"grad_norm": 0.07863107323646545,
|
| 2379 |
+
"learning_rate": 8.543555193821634e-06,
|
| 2380 |
+
"loss": 0.0075,
|
| 2381 |
+
"step": 2760
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 4.161502347417841,
|
| 2385 |
+
"grad_norm": 0.25240081548690796,
|
| 2386 |
+
"learning_rate": 8.252341347488251e-06,
|
| 2387 |
+
"loss": 0.0043,
|
| 2388 |
+
"step": 2770
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 4.176525821596244,
|
| 2392 |
+
"grad_norm": 0.22023944556713104,
|
| 2393 |
+
"learning_rate": 7.965730116708681e-06,
|
| 2394 |
+
"loss": 0.008,
|
| 2395 |
+
"step": 2780
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 4.191549295774648,
|
| 2399 |
+
"grad_norm": 0.010785081423819065,
|
| 2400 |
+
"learning_rate": 7.68375309993304e-06,
|
| 2401 |
+
"loss": 0.0031,
|
| 2402 |
+
"step": 2790
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 4.206572769953052,
|
| 2406 |
+
"grad_norm": 0.11607765406370163,
|
| 2407 |
+
"learning_rate": 7.406441384696372e-06,
|
| 2408 |
+
"loss": 0.0065,
|
| 2409 |
+
"step": 2800
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 4.206572769953052,
|
| 2413 |
+
"eval_loss": 0.04066622629761696,
|
| 2414 |
+
"eval_runtime": 221.9129,
|
| 2415 |
+
"eval_samples_per_second": 5.335,
|
| 2416 |
+
"eval_steps_per_second": 5.335,
|
| 2417 |
+
"step": 2800
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 4.221596244131455,
|
| 2421 |
+
"grad_norm": 0.05429501459002495,
|
| 2422 |
+
"learning_rate": 7.133825544191464e-06,
|
| 2423 |
+
"loss": 0.0095,
|
| 2424 |
+
"step": 2810
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 4.236619718309859,
|
| 2428 |
+
"grad_norm": 0.09971319139003754,
|
| 2429 |
+
"learning_rate": 6.865935633897996e-06,
|
| 2430 |
+
"loss": 0.004,
|
| 2431 |
+
"step": 2820
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 4.251643192488263,
|
| 2435 |
+
"grad_norm": 0.09597835689783096,
|
| 2436 |
+
"learning_rate": 6.602801188269081e-06,
|
| 2437 |
+
"loss": 0.0064,
|
| 2438 |
+
"step": 2830
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 4.266666666666667,
|
| 2442 |
+
"grad_norm": 0.1952216774225235,
|
| 2443 |
+
"learning_rate": 6.344451217475183e-06,
|
| 2444 |
+
"loss": 0.0047,
|
| 2445 |
+
"step": 2840
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 4.28169014084507,
|
| 2449 |
+
"grad_norm": 0.0561475045979023,
|
| 2450 |
+
"learning_rate": 6.090914204205655e-06,
|
| 2451 |
+
"loss": 0.0031,
|
| 2452 |
+
"step": 2850
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 4.28169014084507,
|
| 2456 |
+
"eval_loss": 0.04169252887368202,
|
| 2457 |
+
"eval_runtime": 222.1513,
|
| 2458 |
+
"eval_samples_per_second": 5.33,
|
| 2459 |
+
"eval_steps_per_second": 5.33,
|
| 2460 |
+
"step": 2850
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 4.296713615023474,
|
| 2464 |
+
"grad_norm": 0.2749840021133423,
|
| 2465 |
+
"learning_rate": 5.842218100528679e-06,
|
| 2466 |
+
"loss": 0.0052,
|
| 2467 |
+
"step": 2860
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 4.311737089201878,
|
| 2471 |
+
"grad_norm": 0.22904397547245026,
|
| 2472 |
+
"learning_rate": 5.598390324809555e-06,
|
| 2473 |
+
"loss": 0.0046,
|
| 2474 |
+
"step": 2870
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 4.326760563380281,
|
| 2478 |
+
"grad_norm": 0.26261669397354126,
|
| 2479 |
+
"learning_rate": 5.359457758687841e-06,
|
| 2480 |
+
"loss": 0.0089,
|
| 2481 |
+
"step": 2880
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 4.341784037558686,
|
| 2485 |
+
"grad_norm": 0.21348492801189423,
|
| 2486 |
+
"learning_rate": 5.125446744113743e-06,
|
| 2487 |
+
"loss": 0.0057,
|
| 2488 |
+
"step": 2890
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 4.356807511737089,
|
| 2492 |
+
"grad_norm": 0.0984061136841774,
|
| 2493 |
+
"learning_rate": 4.896383080443934e-06,
|
| 2494 |
+
"loss": 0.0028,
|
| 2495 |
+
"step": 2900
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 4.356807511737089,
|
| 2499 |
+
"eval_loss": 0.041960135102272034,
|
| 2500 |
+
"eval_runtime": 222.6261,
|
| 2501 |
+
"eval_samples_per_second": 5.318,
|
| 2502 |
+
"eval_steps_per_second": 5.318,
|
| 2503 |
+
"step": 2900
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 4.371830985915493,
|
| 2507 |
+
"grad_norm": 0.2619991898536682,
|
| 2508 |
+
"learning_rate": 4.672292021597174e-06,
|
| 2509 |
+
"loss": 0.0056,
|
| 2510 |
+
"step": 2910
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 4.386854460093897,
|
| 2514 |
+
"grad_norm": 0.16088174283504486,
|
| 2515 |
+
"learning_rate": 4.4531982732702145e-06,
|
| 2516 |
+
"loss": 0.0064,
|
| 2517 |
+
"step": 2920
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 4.4018779342723,
|
| 2521 |
+
"grad_norm": 0.3044176995754242,
|
| 2522 |
+
"learning_rate": 4.239125990213883e-06,
|
| 2523 |
+
"loss": 0.0039,
|
| 2524 |
+
"step": 2930
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 4.416901408450705,
|
| 2528 |
+
"grad_norm": 0.06394259631633759,
|
| 2529 |
+
"learning_rate": 4.030098773570174e-06,
|
| 2530 |
+
"loss": 0.0016,
|
| 2531 |
+
"step": 2940
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 4.431924882629108,
|
| 2535 |
+
"grad_norm": 0.15494537353515625,
|
| 2536 |
+
"learning_rate": 3.826139668270234e-06,
|
| 2537 |
+
"loss": 0.0043,
|
| 2538 |
+
"step": 2950
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 4.431924882629108,
|
| 2542 |
+
"eval_loss": 0.042107198387384415,
|
| 2543 |
+
"eval_runtime": 222.671,
|
| 2544 |
+
"eval_samples_per_second": 5.317,
|
| 2545 |
+
"eval_steps_per_second": 5.317,
|
| 2546 |
+
"step": 2950
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 4.446948356807511,
|
| 2550 |
+
"grad_norm": 0.1741316169500351,
|
| 2551 |
+
"learning_rate": 3.6272711604936504e-06,
|
| 2552 |
+
"loss": 0.0049,
|
| 2553 |
+
"step": 2960
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 4.461971830985916,
|
| 2557 |
+
"grad_norm": 0.08207106590270996,
|
| 2558 |
+
"learning_rate": 3.433515175189428e-06,
|
| 2559 |
+
"loss": 0.0053,
|
| 2560 |
+
"step": 2970
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 4.476995305164319,
|
| 2564 |
+
"grad_norm": 0.09723722189664841,
|
| 2565 |
+
"learning_rate": 3.2448930736588e-06,
|
| 2566 |
+
"loss": 0.0077,
|
| 2567 |
+
"step": 2980
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 4.492018779342723,
|
| 2571 |
+
"grad_norm": 0.03407386317849159,
|
| 2572 |
+
"learning_rate": 3.061425651200117e-06,
|
| 2573 |
+
"loss": 0.0038,
|
| 2574 |
+
"step": 2990
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 4.507042253521127,
|
| 2578 |
+
"grad_norm": 0.21065545082092285,
|
| 2579 |
+
"learning_rate": 2.883133134816296e-06,
|
| 2580 |
+
"loss": 0.0048,
|
| 2581 |
+
"step": 3000
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 4.507042253521127,
|
| 2585 |
+
"eval_loss": 0.04240112751722336,
|
| 2586 |
+
"eval_runtime": 222.0883,
|
| 2587 |
+
"eval_samples_per_second": 5.331,
|
| 2588 |
+
"eval_steps_per_second": 5.331,
|
| 2589 |
+
"step": 3000
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 4.52206572769953,
|
| 2593 |
+
"grad_norm": 0.06989271938800812,
|
| 2594 |
+
"learning_rate": 2.7100351809847326e-06,
|
| 2595 |
+
"loss": 0.0046,
|
| 2596 |
+
"step": 3010
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 4.537089201877935,
|
| 2600 |
+
"grad_norm": 0.07983122766017914,
|
| 2601 |
+
"learning_rate": 2.542150873490251e-06,
|
| 2602 |
+
"loss": 0.0028,
|
| 2603 |
+
"step": 3020
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 4.552112676056338,
|
| 2607 |
+
"grad_norm": 0.27960672974586487,
|
| 2608 |
+
"learning_rate": 2.3794987213211383e-06,
|
| 2609 |
+
"loss": 0.0055,
|
| 2610 |
+
"step": 3030
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 4.567136150234742,
|
| 2614 |
+
"grad_norm": 0.2628881633281708,
|
| 2615 |
+
"learning_rate": 2.222096656628547e-06,
|
| 2616 |
+
"loss": 0.0035,
|
| 2617 |
+
"step": 3040
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 4.582159624413146,
|
| 2621 |
+
"grad_norm": 0.13972419500350952,
|
| 2622 |
+
"learning_rate": 2.0699620327495174e-06,
|
| 2623 |
+
"loss": 0.0038,
|
| 2624 |
+
"step": 3050
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 4.582159624413146,
|
| 2628 |
+
"eval_loss": 0.04280637204647064,
|
| 2629 |
+
"eval_runtime": 222.2913,
|
| 2630 |
+
"eval_samples_per_second": 5.326,
|
| 2631 |
+
"eval_steps_per_second": 5.326,
|
| 2632 |
+
"step": 3050
|
| 2633 |
+
},
|
| 2634 |
+
{
|
| 2635 |
+
"epoch": 4.597183098591549,
|
| 2636 |
+
"grad_norm": 0.017372630536556244,
|
| 2637 |
+
"learning_rate": 1.9231116222937996e-06,
|
| 2638 |
+
"loss": 0.0025,
|
| 2639 |
+
"step": 3060
|
| 2640 |
+
},
|
| 2641 |
+
{
|
| 2642 |
+
"epoch": 4.612206572769953,
|
| 2643 |
+
"grad_norm": 0.14479027688503265,
|
| 2644 |
+
"learning_rate": 1.7815616152946523e-06,
|
| 2645 |
+
"loss": 0.0045,
|
| 2646 |
+
"step": 3070
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 4.627230046948357,
|
| 2650 |
+
"grad_norm": 0.03770596161484718,
|
| 2651 |
+
"learning_rate": 1.6453276174240195e-06,
|
| 2652 |
+
"loss": 0.004,
|
| 2653 |
+
"step": 3080
|
| 2654 |
+
},
|
| 2655 |
+
{
|
| 2656 |
+
"epoch": 4.642253521126761,
|
| 2657 |
+
"grad_norm": 0.19395627081394196,
|
| 2658 |
+
"learning_rate": 1.5144246482719114e-06,
|
| 2659 |
+
"loss": 0.0043,
|
| 2660 |
+
"step": 3090
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 4.657276995305164,
|
| 2664 |
+
"grad_norm": 0.05098709091544151,
|
| 2665 |
+
"learning_rate": 1.3888671396905805e-06,
|
| 2666 |
+
"loss": 0.0041,
|
| 2667 |
+
"step": 3100
|
| 2668 |
+
},
|
| 2669 |
+
{
|
| 2670 |
+
"epoch": 4.657276995305164,
|
| 2671 |
+
"eval_loss": 0.0430205836892128,
|
| 2672 |
+
"eval_runtime": 222.6399,
|
| 2673 |
+
"eval_samples_per_second": 5.318,
|
| 2674 |
+
"eval_steps_per_second": 5.318,
|
| 2675 |
+
"step": 3100
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 4.672300469483568,
|
| 2679 |
+
"grad_norm": 0.3063965439796448,
|
| 2680 |
+
"learning_rate": 1.2686689342034431e-06,
|
| 2681 |
+
"loss": 0.0082,
|
| 2682 |
+
"step": 3110
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 4.687323943661972,
|
| 2686 |
+
"grad_norm": 0.21486127376556396,
|
| 2687 |
+
"learning_rate": 1.1538432834789227e-06,
|
| 2688 |
+
"loss": 0.0054,
|
| 2689 |
+
"step": 3120
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 4.702347417840375,
|
| 2693 |
+
"grad_norm": 0.3423399329185486,
|
| 2694 |
+
"learning_rate": 1.044402846869491e-06,
|
| 2695 |
+
"loss": 0.0049,
|
| 2696 |
+
"step": 3130
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 4.71737089201878,
|
| 2700 |
+
"grad_norm": 0.15926554799079895,
|
| 2701 |
+
"learning_rate": 9.403596900160073e-07,
|
| 2702 |
+
"loss": 0.0034,
|
| 2703 |
+
"step": 3140
|
| 2704 |
+
},
|
| 2705 |
+
{
|
| 2706 |
+
"epoch": 4.732394366197183,
|
| 2707 |
+
"grad_norm": 0.24265912175178528,
|
| 2708 |
+
"learning_rate": 8.417252835174749e-07,
|
| 2709 |
+
"loss": 0.0066,
|
| 2710 |
+
"step": 3150
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 4.732394366197183,
|
| 2714 |
+
"eval_loss": 0.043101657181978226,
|
| 2715 |
+
"eval_runtime": 223.1112,
|
| 2716 |
+
"eval_samples_per_second": 5.307,
|
| 2717 |
+
"eval_steps_per_second": 5.307,
|
| 2718 |
+
"step": 3150
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 4.7474178403755865,
|
| 2722 |
+
"grad_norm": 0.13980786502361298,
|
| 2723 |
+
"learning_rate": 7.48510501666455e-07,
|
| 2724 |
+
"loss": 0.0073,
|
| 2725 |
+
"step": 3160
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 4.762441314553991,
|
| 2729 |
+
"grad_norm": 0.14822153747081757,
|
| 2730 |
+
"learning_rate": 6.607256212501578e-07,
|
| 2731 |
+
"loss": 0.0067,
|
| 2732 |
+
"step": 3170
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 4.777464788732394,
|
| 2736 |
+
"grad_norm": 0.2598550021648407,
|
| 2737 |
+
"learning_rate": 5.783803204174654e-07,
|
| 2738 |
+
"loss": 0.0054,
|
| 2739 |
+
"step": 3180
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 4.792488262910798,
|
| 2743 |
+
"grad_norm": 0.145918071269989,
|
| 2744 |
+
"learning_rate": 5.014836776119358e-07,
|
| 2745 |
+
"loss": 0.0051,
|
| 2746 |
+
"step": 3190
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 4.807511737089202,
|
| 2750 |
+
"grad_norm": 0.152046799659729,
|
| 2751 |
+
"learning_rate": 4.300441705708924e-07,
|
| 2752 |
+
"loss": 0.003,
|
| 2753 |
+
"step": 3200
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 4.807511737089202,
|
| 2757 |
+
"eval_loss": 0.043016187846660614,
|
| 2758 |
+
"eval_runtime": 222.4587,
|
| 2759 |
+
"eval_samples_per_second": 5.322,
|
| 2760 |
+
"eval_steps_per_second": 5.322,
|
| 2761 |
+
"step": 3200
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 4.822535211267605,
|
| 2765 |
+
"grad_norm": 0.1859809011220932,
|
| 2766 |
+
"learning_rate": 3.6406967539078796e-07,
|
| 2767 |
+
"loss": 0.0042,
|
| 2768 |
+
"step": 3210
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 4.83755868544601,
|
| 2772 |
+
"grad_norm": 0.10558822751045227,
|
| 2773 |
+
"learning_rate": 3.0356746565887715e-07,
|
| 2774 |
+
"loss": 0.0056,
|
| 2775 |
+
"step": 3220
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 4.852582159624413,
|
| 2779 |
+
"grad_norm": 0.3504423201084137,
|
| 2780 |
+
"learning_rate": 2.485442116513026e-07,
|
| 2781 |
+
"loss": 0.0048,
|
| 2782 |
+
"step": 3230
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 4.867605633802817,
|
| 2786 |
+
"grad_norm": 0.21484680473804474,
|
| 2787 |
+
"learning_rate": 1.9900597959770507e-07,
|
| 2788 |
+
"loss": 0.0079,
|
| 2789 |
+
"step": 3240
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 4.882629107981221,
|
| 2793 |
+
"grad_norm": 0.07226985692977905,
|
| 2794 |
+
"learning_rate": 1.5495823101245866e-07,
|
| 2795 |
+
"loss": 0.0031,
|
| 2796 |
+
"step": 3250
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 4.882629107981221,
|
| 2800 |
+
"eval_loss": 0.042939841747283936,
|
| 2801 |
+
"eval_runtime": 222.1162,
|
| 2802 |
+
"eval_samples_per_second": 5.331,
|
| 2803 |
+
"eval_steps_per_second": 5.331,
|
| 2804 |
+
"step": 3250
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 4.897652582159624,
|
| 2808 |
+
"grad_norm": 0.0429568886756897,
|
| 2809 |
+
"learning_rate": 1.164058220925135e-07,
|
| 2810 |
+
"loss": 0.0043,
|
| 2811 |
+
"step": 3260
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 4.912676056338028,
|
| 2815 |
+
"grad_norm": 0.3218914568424225,
|
| 2816 |
+
"learning_rate": 8.335300318201844e-08,
|
| 2817 |
+
"loss": 0.0074,
|
| 2818 |
+
"step": 3270
|
| 2819 |
+
},
|
| 2820 |
+
{
|
| 2821 |
+
"epoch": 4.927699530516432,
|
| 2822 |
+
"grad_norm": 0.10779738426208496,
|
| 2823 |
+
"learning_rate": 5.5803418303745917e-08,
|
| 2824 |
+
"loss": 0.005,
|
| 2825 |
+
"step": 3280
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 4.942723004694836,
|
| 2829 |
+
"grad_norm": 0.0708744004368782,
|
| 2830 |
+
"learning_rate": 3.3760104757313284e-08,
|
| 2831 |
+
"loss": 0.003,
|
| 2832 |
+
"step": 3290
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 4.957746478873239,
|
| 2836 |
+
"grad_norm": 0.19295717775821686,
|
| 2837 |
+
"learning_rate": 1.7225492784345156e-08,
|
| 2838 |
+
"loss": 0.0046,
|
| 2839 |
+
"step": 3300
|
| 2840 |
+
},
|
| 2841 |
+
{
|
| 2842 |
+
"epoch": 4.957746478873239,
|
| 2843 |
+
"eval_loss": 0.042976681143045425,
|
| 2844 |
+
"eval_runtime": 222.2614,
|
| 2845 |
+
"eval_samples_per_second": 5.327,
|
| 2846 |
+
"eval_steps_per_second": 5.327,
|
| 2847 |
+
"step": 3300
|
| 2848 |
+
},
|
| 2849 |
+
{
|
| 2850 |
+
"epoch": 4.972769953051643,
|
| 2851 |
+
"grad_norm": 0.10675567388534546,
|
| 2852 |
+
"learning_rate": 6.201405300532148e-09,
|
| 2853 |
+
"loss": 0.0058,
|
| 2854 |
+
"step": 3310
|
| 2855 |
+
},
|
| 2856 |
+
{
|
| 2857 |
+
"epoch": 4.987793427230047,
|
| 2858 |
+
"grad_norm": 0.09640572220087051,
|
| 2859 |
+
"learning_rate": 6.890576946805282e-10,
|
| 2860 |
+
"loss": 0.0043,
|
| 2861 |
+
"step": 3320
|
| 2862 |
+
},
|
| 2863 |
+
{
|
| 2864 |
+
"epoch": 4.995305164319249,
|
| 2865 |
+
"step": 3325,
|
| 2866 |
+
"total_flos": 8.987524405535048e+17,
|
| 2867 |
+
"train_loss": 0.049058449685125426,
|
| 2868 |
+
"train_runtime": 45792.1032,
|
| 2869 |
+
"train_samples_per_second": 1.163,
|
| 2870 |
+
"train_steps_per_second": 0.073
|
| 2871 |
+
}
|
| 2872 |
+
],
|
| 2873 |
+
"logging_steps": 10,
|
| 2874 |
+
"max_steps": 3325,
|
| 2875 |
+
"num_input_tokens_seen": 0,
|
| 2876 |
+
"num_train_epochs": 5,
|
| 2877 |
+
"save_steps": 50,
|
| 2878 |
+
"stateful_callbacks": {
|
| 2879 |
+
"TrainerControl": {
|
| 2880 |
+
"args": {
|
| 2881 |
+
"should_epoch_stop": false,
|
| 2882 |
+
"should_evaluate": false,
|
| 2883 |
+
"should_log": false,
|
| 2884 |
+
"should_save": true,
|
| 2885 |
+
"should_training_stop": true
|
| 2886 |
+
},
|
| 2887 |
+
"attributes": {}
|
| 2888 |
+
}
|
| 2889 |
+
},
|
| 2890 |
+
"total_flos": 8.987524405535048e+17,
|
| 2891 |
+
"train_batch_size": 1,
|
| 2892 |
+
"trial_name": null,
|
| 2893 |
+
"trial_params": null
|
| 2894 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|