End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +1186 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: mistral_7b_0-3_oh-dcft-v3.1-qwen-2.5-72b
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# mistral_7b_0-3_oh-dcft-v3.1-qwen-2.5-72b
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.3581
|
| 21 |
|
|
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: mistral_7b_0-3_oh-dcft-v3.1-qwen-2.5-72b
|
|
|
|
| 16 |
|
| 17 |
# mistral_7b_0-3_oh-dcft-v3.1-qwen-2.5-72b
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the mlfoundations-dev/oh-dcft-v3.1-qwen-2.5-72b dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.3581
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3581169545650482,
|
| 4 |
+
"eval_runtime": 55.9349,
|
| 5 |
+
"eval_samples_per_second": 257.049,
|
| 6 |
+
"eval_steps_per_second": 1.019,
|
| 7 |
+
"total_flos": 2683409667194880.0,
|
| 8 |
+
"train_loss": 0.2738360005445992,
|
| 9 |
+
"train_runtime": 9858.4557,
|
| 10 |
+
"train_samples_per_second": 83.126,
|
| 11 |
+
"train_steps_per_second": 0.163
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3581169545650482,
|
| 4 |
+
"eval_runtime": 55.9349,
|
| 5 |
+
"eval_samples_per_second": 257.049,
|
| 6 |
+
"eval_steps_per_second": 1.019
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2683409667194880.0,
|
| 4 |
+
"train_loss": 0.2738360005445992,
|
| 5 |
+
"train_runtime": 9858.4557,
|
| 6 |
+
"train_samples_per_second": 83.126,
|
| 7 |
+
"train_steps_per_second": 0.163
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1602,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.018726591760299626,
|
| 13 |
+
"grad_norm": 6.9583668884714545,
|
| 14 |
+
"learning_rate": 6.17283950617284e-07,
|
| 15 |
+
"loss": 0.5853,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03745318352059925,
|
| 20 |
+
"grad_norm": 2.459747968301057,
|
| 21 |
+
"learning_rate": 1.234567901234568e-06,
|
| 22 |
+
"loss": 0.4956,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.056179775280898875,
|
| 27 |
+
"grad_norm": 1.5518874081248877,
|
| 28 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 29 |
+
"loss": 0.4511,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0749063670411985,
|
| 34 |
+
"grad_norm": 1.327079400917728,
|
| 35 |
+
"learning_rate": 2.469135802469136e-06,
|
| 36 |
+
"loss": 0.4242,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09363295880149813,
|
| 41 |
+
"grad_norm": 1.6052721933898932,
|
| 42 |
+
"learning_rate": 3.08641975308642e-06,
|
| 43 |
+
"loss": 0.4112,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11235955056179775,
|
| 48 |
+
"grad_norm": 2.1363218656503236,
|
| 49 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 50 |
+
"loss": 0.4056,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.13108614232209737,
|
| 55 |
+
"grad_norm": 2.5025301508678246,
|
| 56 |
+
"learning_rate": 4.3209876543209875e-06,
|
| 57 |
+
"loss": 0.3934,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.149812734082397,
|
| 62 |
+
"grad_norm": 1.6968039665700831,
|
| 63 |
+
"learning_rate": 4.938271604938272e-06,
|
| 64 |
+
"loss": 0.3931,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16853932584269662,
|
| 69 |
+
"grad_norm": 1.710626024878931,
|
| 70 |
+
"learning_rate": 4.999611253625062e-06,
|
| 71 |
+
"loss": 0.3854,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.18726591760299627,
|
| 76 |
+
"grad_norm": 2.229340351997921,
|
| 77 |
+
"learning_rate": 4.99826761145702e-06,
|
| 78 |
+
"loss": 0.388,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.20599250936329588,
|
| 83 |
+
"grad_norm": 1.6554235103974217,
|
| 84 |
+
"learning_rate": 4.995964847225794e-06,
|
| 85 |
+
"loss": 0.3798,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2247191011235955,
|
| 90 |
+
"grad_norm": 1.5390529564927018,
|
| 91 |
+
"learning_rate": 4.992703943301973e-06,
|
| 92 |
+
"loss": 0.3788,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.24344569288389514,
|
| 97 |
+
"grad_norm": 1.686640147089175,
|
| 98 |
+
"learning_rate": 4.988486290803271e-06,
|
| 99 |
+
"loss": 0.3724,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.26217228464419473,
|
| 104 |
+
"grad_norm": 1.4586531417129365,
|
| 105 |
+
"learning_rate": 4.983313689001068e-06,
|
| 106 |
+
"loss": 0.3737,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2808988764044944,
|
| 111 |
+
"grad_norm": 1.383788345325366,
|
| 112 |
+
"learning_rate": 4.977188344552831e-06,
|
| 113 |
+
"loss": 0.3708,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.299625468164794,
|
| 118 |
+
"grad_norm": 2.344749126006063,
|
| 119 |
+
"learning_rate": 4.970112870560744e-06,
|
| 120 |
+
"loss": 0.3705,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.31835205992509363,
|
| 125 |
+
"grad_norm": 1.333313085188304,
|
| 126 |
+
"learning_rate": 4.962090285456943e-06,
|
| 127 |
+
"loss": 0.3726,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.33707865168539325,
|
| 132 |
+
"grad_norm": 1.2767396164379259,
|
| 133 |
+
"learning_rate": 4.953124011715844e-06,
|
| 134 |
+
"loss": 0.368,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.35580524344569286,
|
| 139 |
+
"grad_norm": 1.8939736466696433,
|
| 140 |
+
"learning_rate": 4.943217874394092e-06,
|
| 141 |
+
"loss": 0.3673,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.37453183520599254,
|
| 146 |
+
"grad_norm": 1.4076191572126282,
|
| 147 |
+
"learning_rate": 4.932376099498779e-06,
|
| 148 |
+
"loss": 0.3592,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.39325842696629215,
|
| 153 |
+
"grad_norm": 1.2393424297302382,
|
| 154 |
+
"learning_rate": 4.920603312184602e-06,
|
| 155 |
+
"loss": 0.3646,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.41198501872659177,
|
| 160 |
+
"grad_norm": 1.6712067045257082,
|
| 161 |
+
"learning_rate": 4.907904534780752e-06,
|
| 162 |
+
"loss": 0.3567,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4307116104868914,
|
| 167 |
+
"grad_norm": 1.612702425647588,
|
| 168 |
+
"learning_rate": 4.894285184648369e-06,
|
| 169 |
+
"loss": 0.3585,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.449438202247191,
|
| 174 |
+
"grad_norm": 2.4813858906223834,
|
| 175 |
+
"learning_rate": 4.879751071869461e-06,
|
| 176 |
+
"loss": 0.3623,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4681647940074906,
|
| 181 |
+
"grad_norm": 2.163116509485285,
|
| 182 |
+
"learning_rate": 4.864308396768294e-06,
|
| 183 |
+
"loss": 0.3586,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4868913857677903,
|
| 188 |
+
"grad_norm": 1.7525497737806786,
|
| 189 |
+
"learning_rate": 4.847963747266312e-06,
|
| 190 |
+
"loss": 0.3547,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5056179775280899,
|
| 195 |
+
"grad_norm": 1.3534898789991496,
|
| 196 |
+
"learning_rate": 4.83072409607169e-06,
|
| 197 |
+
"loss": 0.3534,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5243445692883895,
|
| 202 |
+
"grad_norm": 1.6829737229447668,
|
| 203 |
+
"learning_rate": 4.81259679770474e-06,
|
| 204 |
+
"loss": 0.3519,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5430711610486891,
|
| 209 |
+
"grad_norm": 1.3735075038975693,
|
| 210 |
+
"learning_rate": 4.7935895853604455e-06,
|
| 211 |
+
"loss": 0.353,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5617977528089888,
|
| 216 |
+
"grad_norm": 1.9414583619507375,
|
| 217 |
+
"learning_rate": 4.773710567609436e-06,
|
| 218 |
+
"loss": 0.3547,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5805243445692884,
|
| 223 |
+
"grad_norm": 1.4766636724732591,
|
| 224 |
+
"learning_rate": 4.7529682249388284e-06,
|
| 225 |
+
"loss": 0.3515,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.599250936329588,
|
| 230 |
+
"grad_norm": 1.420170786947651,
|
| 231 |
+
"learning_rate": 4.7313714061344165e-06,
|
| 232 |
+
"loss": 0.3559,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6179775280898876,
|
| 237 |
+
"grad_norm": 1.585688048778078,
|
| 238 |
+
"learning_rate": 4.708929324505729e-06,
|
| 239 |
+
"loss": 0.3474,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6367041198501873,
|
| 244 |
+
"grad_norm": 1.4660422960583963,
|
| 245 |
+
"learning_rate": 4.685651553955589e-06,
|
| 246 |
+
"loss": 0.3478,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.6554307116104869,
|
| 251 |
+
"grad_norm": 1.729952760499875,
|
| 252 |
+
"learning_rate": 4.661548024895843e-06,
|
| 253 |
+
"loss": 0.3502,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6741573033707865,
|
| 258 |
+
"grad_norm": 1.5477887192577897,
|
| 259 |
+
"learning_rate": 4.636629020010989e-06,
|
| 260 |
+
"loss": 0.3462,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6928838951310862,
|
| 265 |
+
"grad_norm": 1.3822726826009006,
|
| 266 |
+
"learning_rate": 4.6109051698715425e-06,
|
| 267 |
+
"loss": 0.3447,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7116104868913857,
|
| 272 |
+
"grad_norm": 1.4151446110625596,
|
| 273 |
+
"learning_rate": 4.5843874483989744e-06,
|
| 274 |
+
"loss": 0.3452,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.7303370786516854,
|
| 279 |
+
"grad_norm": 1.213585225271953,
|
| 280 |
+
"learning_rate": 4.5570871681841915e-06,
|
| 281 |
+
"loss": 0.3525,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7490636704119851,
|
| 286 |
+
"grad_norm": 1.2153292815838297,
|
| 287 |
+
"learning_rate": 4.529015975661518e-06,
|
| 288 |
+
"loss": 0.3497,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7677902621722846,
|
| 293 |
+
"grad_norm": 1.0767956653910644,
|
| 294 |
+
"learning_rate": 4.5001858461402765e-06,
|
| 295 |
+
"loss": 0.3469,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7865168539325843,
|
| 300 |
+
"grad_norm": 1.0725016047427482,
|
| 301 |
+
"learning_rate": 4.470609078696062e-06,
|
| 302 |
+
"loss": 0.3459,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8052434456928839,
|
| 307 |
+
"grad_norm": 1.14757314560185,
|
| 308 |
+
"learning_rate": 4.440298290923893e-06,
|
| 309 |
+
"loss": 0.3446,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8239700374531835,
|
| 314 |
+
"grad_norm": 1.435028611591542,
|
| 315 |
+
"learning_rate": 4.409266413555481e-06,
|
| 316 |
+
"loss": 0.3468,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.8426966292134831,
|
| 321 |
+
"grad_norm": 1.3563421475894943,
|
| 322 |
+
"learning_rate": 4.3775266849429245e-06,
|
| 323 |
+
"loss": 0.3471,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8614232209737828,
|
| 328 |
+
"grad_norm": 1.272968934220072,
|
| 329 |
+
"learning_rate": 4.345092645411154e-06,
|
| 330 |
+
"loss": 0.3459,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8801498127340824,
|
| 335 |
+
"grad_norm": 1.1988760260435716,
|
| 336 |
+
"learning_rate": 4.311978131481565e-06,
|
| 337 |
+
"loss": 0.3439,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.898876404494382,
|
| 342 |
+
"grad_norm": 1.2157009389799966,
|
| 343 |
+
"learning_rate": 4.2781972699692955e-06,
|
| 344 |
+
"loss": 0.341,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.9176029962546817,
|
| 349 |
+
"grad_norm": 1.2302691306481766,
|
| 350 |
+
"learning_rate": 4.243764471956648e-06,
|
| 351 |
+
"loss": 0.3436,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.9363295880149812,
|
| 356 |
+
"grad_norm": 1.1395524537917356,
|
| 357 |
+
"learning_rate": 4.208694426645257e-06,
|
| 358 |
+
"loss": 0.3442,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9550561797752809,
|
| 363 |
+
"grad_norm": 1.0502686640912897,
|
| 364 |
+
"learning_rate": 4.1730020950895985e-06,
|
| 365 |
+
"loss": 0.3406,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9737827715355806,
|
| 370 |
+
"grad_norm": 1.1788715508510565,
|
| 371 |
+
"learning_rate": 4.136702703814536e-06,
|
| 372 |
+
"loss": 0.3399,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.9925093632958801,
|
| 377 |
+
"grad_norm": 0.9961303841798358,
|
| 378 |
+
"learning_rate": 4.0998117383196035e-06,
|
| 379 |
+
"loss": 0.3426,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.0,
|
| 384 |
+
"eval_loss": 0.3397010564804077,
|
| 385 |
+
"eval_runtime": 49.3629,
|
| 386 |
+
"eval_samples_per_second": 291.271,
|
| 387 |
+
"eval_steps_per_second": 1.155,
|
| 388 |
+
"step": 534
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0112359550561798,
|
| 392 |
+
"grad_norm": 1.5016330309741552,
|
| 393 |
+
"learning_rate": 4.062344936472819e-06,
|
| 394 |
+
"loss": 0.3041,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.0299625468164795,
|
| 399 |
+
"grad_norm": 1.4105174151178423,
|
| 400 |
+
"learning_rate": 4.024318281796832e-06,
|
| 401 |
+
"loss": 0.2653,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.048689138576779,
|
| 406 |
+
"grad_norm": 1.1267842205009497,
|
| 407 |
+
"learning_rate": 3.985747996650271e-06,
|
| 408 |
+
"loss": 0.2635,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.0674157303370786,
|
| 413 |
+
"grad_norm": 1.4577741522220231,
|
| 414 |
+
"learning_rate": 3.94665053530721e-06,
|
| 415 |
+
"loss": 0.2661,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.0861423220973783,
|
| 420 |
+
"grad_norm": 1.1743755144218508,
|
| 421 |
+
"learning_rate": 3.907042576937689e-06,
|
| 422 |
+
"loss": 0.2614,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.104868913857678,
|
| 427 |
+
"grad_norm": 1.1077587707464613,
|
| 428 |
+
"learning_rate": 3.8669410184923e-06,
|
| 429 |
+
"loss": 0.2638,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.1235955056179776,
|
| 434 |
+
"grad_norm": 1.499213012424364,
|
| 435 |
+
"learning_rate": 3.826362967493867e-06,
|
| 436 |
+
"loss": 0.2649,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.142322097378277,
|
| 441 |
+
"grad_norm": 1.2186830879218926,
|
| 442 |
+
"learning_rate": 3.7853257347392865e-06,
|
| 443 |
+
"loss": 0.2664,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.1610486891385767,
|
| 448 |
+
"grad_norm": 1.5365523984386804,
|
| 449 |
+
"learning_rate": 3.74384682691466e-06,
|
| 450 |
+
"loss": 0.2656,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.1797752808988764,
|
| 455 |
+
"grad_norm": 1.2824215690021454,
|
| 456 |
+
"learning_rate": 3.701943939126856e-06,
|
| 457 |
+
"loss": 0.264,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.198501872659176,
|
| 462 |
+
"grad_norm": 1.1855864127248261,
|
| 463 |
+
"learning_rate": 3.659634947354686e-06,
|
| 464 |
+
"loss": 0.2612,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.2172284644194757,
|
| 469 |
+
"grad_norm": 1.0357586593608685,
|
| 470 |
+
"learning_rate": 3.61693790082293e-06,
|
| 471 |
+
"loss": 0.2609,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.2359550561797752,
|
| 476 |
+
"grad_norm": 1.2298112784157318,
|
| 477 |
+
"learning_rate": 3.5738710143024406e-06,
|
| 478 |
+
"loss": 0.2648,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.2546816479400749,
|
| 483 |
+
"grad_norm": 1.1873771143182965,
|
| 484 |
+
"learning_rate": 3.530452660339638e-06,
|
| 485 |
+
"loss": 0.2648,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.2734082397003745,
|
| 490 |
+
"grad_norm": 1.1104387940773652,
|
| 491 |
+
"learning_rate": 3.486701361418686e-06,
|
| 492 |
+
"loss": 0.2669,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.2921348314606742,
|
| 497 |
+
"grad_norm": 1.0981466202062755,
|
| 498 |
+
"learning_rate": 3.4426357820597144e-06,
|
| 499 |
+
"loss": 0.2656,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.3108614232209739,
|
| 504 |
+
"grad_norm": 1.0622356793414978,
|
| 505 |
+
"learning_rate": 3.3982747208564383e-06,
|
| 506 |
+
"loss": 0.2684,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.3295880149812733,
|
| 511 |
+
"grad_norm": 1.1204181746461044,
|
| 512 |
+
"learning_rate": 3.353637102456585e-06,
|
| 513 |
+
"loss": 0.2632,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.348314606741573,
|
| 518 |
+
"grad_norm": 1.1561988340566498,
|
| 519 |
+
"learning_rate": 3.30874196948855e-06,
|
| 520 |
+
"loss": 0.2655,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.3670411985018727,
|
| 525 |
+
"grad_norm": 1.1634759808618433,
|
| 526 |
+
"learning_rate": 3.2636084744377123e-06,
|
| 527 |
+
"loss": 0.2662,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.3857677902621723,
|
| 532 |
+
"grad_norm": 1.183622746360506,
|
| 533 |
+
"learning_rate": 3.218255871475894e-06,
|
| 534 |
+
"loss": 0.2665,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.404494382022472,
|
| 539 |
+
"grad_norm": 1.3035644321187594,
|
| 540 |
+
"learning_rate": 3.172703508247433e-06,
|
| 541 |
+
"loss": 0.2641,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.4232209737827715,
|
| 546 |
+
"grad_norm": 1.2520080394828148,
|
| 547 |
+
"learning_rate": 3.126970817615384e-06,
|
| 548 |
+
"loss": 0.2641,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.4419475655430711,
|
| 553 |
+
"grad_norm": 1.1658528659149325,
|
| 554 |
+
"learning_rate": 3.0810773093713615e-06,
|
| 555 |
+
"loss": 0.2672,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.4606741573033708,
|
| 560 |
+
"grad_norm": 1.1183168156658745,
|
| 561 |
+
"learning_rate": 3.0350425619125678e-06,
|
| 562 |
+
"loss": 0.2679,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.4794007490636705,
|
| 567 |
+
"grad_norm": 1.2572209531909289,
|
| 568 |
+
"learning_rate": 2.988886213889551e-06,
|
| 569 |
+
"loss": 0.2673,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.4981273408239701,
|
| 574 |
+
"grad_norm": 1.1500933028632634,
|
| 575 |
+
"learning_rate": 2.9426279558282617e-06,
|
| 576 |
+
"loss": 0.2661,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.5168539325842696,
|
| 581 |
+
"grad_norm": 1.090606866883075,
|
| 582 |
+
"learning_rate": 2.896287521729974e-06,
|
| 583 |
+
"loss": 0.2616,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.5355805243445693,
|
| 588 |
+
"grad_norm": 1.1320316226013576,
|
| 589 |
+
"learning_rate": 2.849884680652666e-06,
|
| 590 |
+
"loss": 0.27,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.554307116104869,
|
| 595 |
+
"grad_norm": 1.103244276726941,
|
| 596 |
+
"learning_rate": 2.8034392282774415e-06,
|
| 597 |
+
"loss": 0.2635,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.5730337078651684,
|
| 602 |
+
"grad_norm": 1.099593203599622,
|
| 603 |
+
"learning_rate": 2.7569709784635934e-06,
|
| 604 |
+
"loss": 0.2619,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.5917602996254683,
|
| 609 |
+
"grad_norm": 1.0707708462874654,
|
| 610 |
+
"learning_rate": 2.710499754795916e-06,
|
| 611 |
+
"loss": 0.2636,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.6104868913857677,
|
| 616 |
+
"grad_norm": 1.1077681383186893,
|
| 617 |
+
"learning_rate": 2.664045382127863e-06,
|
| 618 |
+
"loss": 0.2635,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.6292134831460674,
|
| 623 |
+
"grad_norm": 1.0523839107618693,
|
| 624 |
+
"learning_rate": 2.6176276781241695e-06,
|
| 625 |
+
"loss": 0.2671,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.647940074906367,
|
| 630 |
+
"grad_norm": 1.1453676425960706,
|
| 631 |
+
"learning_rate": 2.571266444806535e-06,
|
| 632 |
+
"loss": 0.2636,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.6666666666666665,
|
| 637 |
+
"grad_norm": 1.0877845955489727,
|
| 638 |
+
"learning_rate": 2.5249814601059897e-06,
|
| 639 |
+
"loss": 0.2622,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.6853932584269664,
|
| 644 |
+
"grad_norm": 1.109206643426623,
|
| 645 |
+
"learning_rate": 2.4787924694255235e-06,
|
| 646 |
+
"loss": 0.2674,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.7041198501872659,
|
| 651 |
+
"grad_norm": 1.148003058058869,
|
| 652 |
+
"learning_rate": 2.4327191772166046e-06,
|
| 653 |
+
"loss": 0.2666,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.7228464419475655,
|
| 658 |
+
"grad_norm": 1.0373837405536894,
|
| 659 |
+
"learning_rate": 2.3867812385731627e-06,
|
| 660 |
+
"loss": 0.2624,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.7415730337078652,
|
| 665 |
+
"grad_norm": 1.0452119354423188,
|
| 666 |
+
"learning_rate": 2.340998250846629e-06,
|
| 667 |
+
"loss": 0.2672,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.7602996254681647,
|
| 672 |
+
"grad_norm": 1.0839517512429684,
|
| 673 |
+
"learning_rate": 2.2953897452856035e-06,
|
| 674 |
+
"loss": 0.2627,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.7790262172284645,
|
| 679 |
+
"grad_norm": 1.1357572847216206,
|
| 680 |
+
"learning_rate": 2.2499751787037404e-06,
|
| 681 |
+
"loss": 0.2658,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.797752808988764,
|
| 686 |
+
"grad_norm": 1.049597976731673,
|
| 687 |
+
"learning_rate": 2.2047739251793624e-06,
|
| 688 |
+
"loss": 0.2634,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.8164794007490637,
|
| 693 |
+
"grad_norm": 1.1411990916035244,
|
| 694 |
+
"learning_rate": 2.1598052677903904e-06,
|
| 695 |
+
"loss": 0.261,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.8352059925093633,
|
| 700 |
+
"grad_norm": 1.1008798760396603,
|
| 701 |
+
"learning_rate": 2.1150883903880863e-06,
|
| 702 |
+
"loss": 0.2622,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.8539325842696628,
|
| 707 |
+
"grad_norm": 1.0556909678728987,
|
| 708 |
+
"learning_rate": 2.0706423694131246e-06,
|
| 709 |
+
"loss": 0.2589,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.8726591760299627,
|
| 714 |
+
"grad_norm": 1.01122078883521,
|
| 715 |
+
"learning_rate": 2.0264861657574936e-06,
|
| 716 |
+
"loss": 0.2602,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.8913857677902621,
|
| 721 |
+
"grad_norm": 1.030111452172985,
|
| 722 |
+
"learning_rate": 1.9826386166756796e-06,
|
| 723 |
+
"loss": 0.2625,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.9101123595505618,
|
| 728 |
+
"grad_norm": 1.0304981637914168,
|
| 729 |
+
"learning_rate": 1.9391184277486046e-06,
|
| 730 |
+
"loss": 0.2616,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.9288389513108615,
|
| 735 |
+
"grad_norm": 1.1002315606948327,
|
| 736 |
+
"learning_rate": 1.8959441649037334e-06,
|
| 737 |
+
"loss": 0.2618,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.947565543071161,
|
| 742 |
+
"grad_norm": 1.0682674034690947,
|
| 743 |
+
"learning_rate": 1.8531342464947551e-06,
|
| 744 |
+
"loss": 0.2604,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.9662921348314608,
|
| 749 |
+
"grad_norm": 1.0285099387800412,
|
| 750 |
+
"learning_rate": 1.810706935444223e-06,
|
| 751 |
+
"loss": 0.2623,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.9850187265917603,
|
| 756 |
+
"grad_norm": 1.054687396839439,
|
| 757 |
+
"learning_rate": 1.7686803314525056e-06,
|
| 758 |
+
"loss": 0.264,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.0,
|
| 763 |
+
"eval_loss": 0.3330671787261963,
|
| 764 |
+
"eval_runtime": 50.16,
|
| 765 |
+
"eval_samples_per_second": 286.643,
|
| 766 |
+
"eval_steps_per_second": 1.136,
|
| 767 |
+
"step": 1068
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.0037453183520597,
|
| 771 |
+
"grad_norm": 1.7923902161169425,
|
| 772 |
+
"learning_rate": 1.7270723632763563e-06,
|
| 773 |
+
"loss": 0.2446,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.0224719101123596,
|
| 778 |
+
"grad_norm": 1.345237927622502,
|
| 779 |
+
"learning_rate": 1.685900781080428e-06,
|
| 780 |
+
"loss": 0.1907,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.041198501872659,
|
| 785 |
+
"grad_norm": 1.1496616768510104,
|
| 786 |
+
"learning_rate": 1.6451831488649562e-06,
|
| 787 |
+
"loss": 0.1874,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.059925093632959,
|
| 792 |
+
"grad_norm": 1.1107541934632752,
|
| 793 |
+
"learning_rate": 1.6049368369728748e-06,
|
| 794 |
+
"loss": 0.187,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.0786516853932584,
|
| 799 |
+
"grad_norm": 1.1294609243658884,
|
| 800 |
+
"learning_rate": 1.5651790146795434e-06,
|
| 801 |
+
"loss": 0.1861,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.097378277153558,
|
| 806 |
+
"grad_norm": 1.1064659786749327,
|
| 807 |
+
"learning_rate": 1.525926642868242e-06,
|
| 808 |
+
"loss": 0.188,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.1161048689138577,
|
| 813 |
+
"grad_norm": 1.1288942946219733,
|
| 814 |
+
"learning_rate": 1.4871964667945815e-06,
|
| 815 |
+
"loss": 0.1889,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.134831460674157,
|
| 820 |
+
"grad_norm": 1.1095107607344008,
|
| 821 |
+
"learning_rate": 1.4490050089428965e-06,
|
| 822 |
+
"loss": 0.1878,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.153558052434457,
|
| 827 |
+
"grad_norm": 1.1688843518004304,
|
| 828 |
+
"learning_rate": 1.411368561977662e-06,
|
| 829 |
+
"loss": 0.1873,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.1722846441947565,
|
| 834 |
+
"grad_norm": 1.1746086833216787,
|
| 835 |
+
"learning_rate": 1.3743031817929725e-06,
|
| 836 |
+
"loss": 0.1885,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.191011235955056,
|
| 841 |
+
"grad_norm": 1.0953314936943879,
|
| 842 |
+
"learning_rate": 1.337824680663016e-06,
|
| 843 |
+
"loss": 0.1918,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.209737827715356,
|
| 848 |
+
"grad_norm": 1.1591899466119513,
|
| 849 |
+
"learning_rate": 1.3019486204964737e-06,
|
| 850 |
+
"loss": 0.1883,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.2284644194756553,
|
| 855 |
+
"grad_norm": 1.1558485332053892,
|
| 856 |
+
"learning_rate": 1.2666903061977397e-06,
|
| 857 |
+
"loss": 0.1901,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.247191011235955,
|
| 862 |
+
"grad_norm": 1.2117954838055114,
|
| 863 |
+
"learning_rate": 1.2320647791377731e-06,
|
| 864 |
+
"loss": 0.1895,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.2659176029962547,
|
| 869 |
+
"grad_norm": 1.0645164422876823,
|
| 870 |
+
"learning_rate": 1.1980868107373687e-06,
|
| 871 |
+
"loss": 0.1863,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.284644194756554,
|
| 876 |
+
"grad_norm": 1.1541528576196018,
|
| 877 |
+
"learning_rate": 1.1647708961656036e-06,
|
| 878 |
+
"loss": 0.1901,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.303370786516854,
|
| 883 |
+
"grad_norm": 1.116106510953403,
|
| 884 |
+
"learning_rate": 1.1321312481561198e-06,
|
| 885 |
+
"loss": 0.1892,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 2.3220973782771535,
|
| 890 |
+
"grad_norm": 1.0597155426906926,
|
| 891 |
+
"learning_rate": 1.1001817909439065e-06,
|
| 892 |
+
"loss": 0.1894,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.3408239700374533,
|
| 897 |
+
"grad_norm": 1.1196874887948434,
|
| 898 |
+
"learning_rate": 1.0689361543251532e-06,
|
| 899 |
+
"loss": 0.1887,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.359550561797753,
|
| 904 |
+
"grad_norm": 1.1714322905126089,
|
| 905 |
+
"learning_rate": 1.038407667842705e-06,
|
| 906 |
+
"loss": 0.1864,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.3782771535580522,
|
| 911 |
+
"grad_norm": 1.0763451314249386,
|
| 912 |
+
"learning_rate": 1.008609355099621e-06,
|
| 913 |
+
"loss": 0.1864,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.397003745318352,
|
| 918 |
+
"grad_norm": 1.1473816461596407,
|
| 919 |
+
"learning_rate": 9.795539282032386e-07,
|
| 920 |
+
"loss": 0.1875,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.4157303370786516,
|
| 925 |
+
"grad_norm": 1.098908279814066,
|
| 926 |
+
"learning_rate": 9.512537823421229e-07,
|
| 927 |
+
"loss": 0.1864,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.4344569288389515,
|
| 932 |
+
"grad_norm": 1.1578303830285432,
|
| 933 |
+
"learning_rate": 9.237209904982213e-07,
|
| 934 |
+
"loss": 0.1904,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.453183520599251,
|
| 939 |
+
"grad_norm": 1.1404639700697865,
|
| 940 |
+
"learning_rate": 8.969672982964711e-07,
|
| 941 |
+
"loss": 0.1907,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.4719101123595504,
|
| 946 |
+
"grad_norm": 1.0825550585492165,
|
| 947 |
+
"learning_rate": 8.710041189940513e-07,
|
| 948 |
+
"loss": 0.1868,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.4906367041198503,
|
| 953 |
+
"grad_norm": 1.1503042083334585,
|
| 954 |
+
"learning_rate": 8.458425286114396e-07,
|
| 955 |
+
"loss": 0.189,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.5093632958801497,
|
| 960 |
+
"grad_norm": 1.1289267850787381,
|
| 961 |
+
"learning_rate": 8.214932612073207e-07,
|
| 962 |
+
"loss": 0.1891,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.5280898876404496,
|
| 967 |
+
"grad_norm": 1.0922707099906583,
|
| 968 |
+
"learning_rate": 7.979667042993847e-07,
|
| 969 |
+
"loss": 0.1882,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 2.546816479400749,
|
| 974 |
+
"grad_norm": 1.159472662881807,
|
| 975 |
+
"learning_rate": 7.752728944329658e-07,
|
| 976 |
+
"loss": 0.1884,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.5655430711610485,
|
| 981 |
+
"grad_norm": 1.096656880156706,
|
| 982 |
+
"learning_rate": 7.534215128993953e-07,
|
| 983 |
+
"loss": 0.1912,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.5842696629213484,
|
| 988 |
+
"grad_norm": 1.177067424925165,
|
| 989 |
+
"learning_rate": 7.324218816059202e-07,
|
| 990 |
+
"loss": 0.1869,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.602996254681648,
|
| 995 |
+
"grad_norm": 1.1433902131098264,
|
| 996 |
+
"learning_rate": 7.122829590989305e-07,
|
| 997 |
+
"loss": 0.1859,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.6217228464419478,
|
| 1002 |
+
"grad_norm": 1.0780282296557087,
|
| 1003 |
+
"learning_rate": 6.930133367421965e-07,
|
| 1004 |
+
"loss": 0.1903,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.640449438202247,
|
| 1009 |
+
"grad_norm": 1.1302075316778486,
|
| 1010 |
+
"learning_rate": 6.746212350517535e-07,
|
| 1011 |
+
"loss": 0.187,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 2.6591760299625467,
|
| 1016 |
+
"grad_norm": 1.1073095304447773,
|
| 1017 |
+
"learning_rate": 6.57114500188989e-07,
|
| 1018 |
+
"loss": 0.1886,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.6779026217228465,
|
| 1023 |
+
"grad_norm": 1.0420564202777807,
|
| 1024 |
+
"learning_rate": 6.405006006134285e-07,
|
| 1025 |
+
"loss": 0.1879,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 2.696629213483146,
|
| 1030 |
+
"grad_norm": 1.0777824438327903,
|
| 1031 |
+
"learning_rate": 6.247866238966579e-07,
|
| 1032 |
+
"loss": 0.1904,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.715355805243446,
|
| 1037 |
+
"grad_norm": 1.1121227385024828,
|
| 1038 |
+
"learning_rate": 6.099792736987272e-07,
|
| 1039 |
+
"loss": 0.1873,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.7340823970037453,
|
| 1044 |
+
"grad_norm": 1.109347730952044,
|
| 1045 |
+
"learning_rate": 5.96084866908337e-07,
|
| 1046 |
+
"loss": 0.1897,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 2.752808988764045,
|
| 1051 |
+
"grad_norm": 1.131219956309559,
|
| 1052 |
+
"learning_rate": 5.831093309480244e-07,
|
| 1053 |
+
"loss": 0.1848,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 2.7715355805243447,
|
| 1058 |
+
"grad_norm": 1.0563106661530495,
|
| 1059 |
+
"learning_rate": 5.710582012454928e-07,
|
| 1060 |
+
"loss": 0.1862,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 2.790262172284644,
|
| 1065 |
+
"grad_norm": 1.1800848608197825,
|
| 1066 |
+
"learning_rate": 5.599366188721713e-07,
|
| 1067 |
+
"loss": 0.1878,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 2.808988764044944,
|
| 1072 |
+
"grad_norm": 1.1143488700250004,
|
| 1073 |
+
"learning_rate": 5.497493283500105e-07,
|
| 1074 |
+
"loss": 0.1875,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 2.8277153558052435,
|
| 1079 |
+
"grad_norm": 1.0433720861256908,
|
| 1080 |
+
"learning_rate": 5.405006756274435e-07,
|
| 1081 |
+
"loss": 0.1872,
|
| 1082 |
+
"step": 1510
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.846441947565543,
|
| 1086 |
+
"grad_norm": 1.1343560438226763,
|
| 1087 |
+
"learning_rate": 5.321946062253841e-07,
|
| 1088 |
+
"loss": 0.1894,
|
| 1089 |
+
"step": 1520
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.865168539325843,
|
| 1093 |
+
"grad_norm": 1.1296113946372626,
|
| 1094 |
+
"learning_rate": 5.248346635540485e-07,
|
| 1095 |
+
"loss": 0.1864,
|
| 1096 |
+
"step": 1530
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.8838951310861423,
|
| 1100 |
+
"grad_norm": 1.1286259876244906,
|
| 1101 |
+
"learning_rate": 5.184239874013167e-07,
|
| 1102 |
+
"loss": 0.1893,
|
| 1103 |
+
"step": 1540
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.902621722846442,
|
| 1107 |
+
"grad_norm": 1.079795528012858,
|
| 1108 |
+
"learning_rate": 5.129653125932872e-07,
|
| 1109 |
+
"loss": 0.1867,
|
| 1110 |
+
"step": 1550
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 2.9213483146067416,
|
| 1114 |
+
"grad_norm": 1.1367094539661884,
|
| 1115 |
+
"learning_rate": 5.084609678275833e-07,
|
| 1116 |
+
"loss": 0.1913,
|
| 1117 |
+
"step": 1560
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 2.940074906367041,
|
| 1121 |
+
"grad_norm": 1.1317799697421957,
|
| 1122 |
+
"learning_rate": 5.049128746799206e-07,
|
| 1123 |
+
"loss": 0.1874,
|
| 1124 |
+
"step": 1570
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.958801498127341,
|
| 1128 |
+
"grad_norm": 1.0575396751944557,
|
| 1129 |
+
"learning_rate": 5.023225467843537e-07,
|
| 1130 |
+
"loss": 0.1868,
|
| 1131 |
+
"step": 1580
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.9775280898876404,
|
| 1135 |
+
"grad_norm": 1.0774002831939866,
|
| 1136 |
+
"learning_rate": 5.006910891875522e-07,
|
| 1137 |
+
"loss": 0.1875,
|
| 1138 |
+
"step": 1590
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.9962546816479403,
|
| 1142 |
+
"grad_norm": 1.0987879465354378,
|
| 1143 |
+
"learning_rate": 5.00019197877381e-07,
|
| 1144 |
+
"loss": 0.1874,
|
| 1145 |
+
"step": 1600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 3.0,
|
| 1149 |
+
"eval_loss": 0.3581169545650482,
|
| 1150 |
+
"eval_runtime": 55.9282,
|
| 1151 |
+
"eval_samples_per_second": 257.08,
|
| 1152 |
+
"eval_steps_per_second": 1.019,
|
| 1153 |
+
"step": 1602
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 3.0,
|
| 1157 |
+
"step": 1602,
|
| 1158 |
+
"total_flos": 2683409667194880.0,
|
| 1159 |
+
"train_loss": 0.2738360005445992,
|
| 1160 |
+
"train_runtime": 9858.4557,
|
| 1161 |
+
"train_samples_per_second": 83.126,
|
| 1162 |
+
"train_steps_per_second": 0.163
|
| 1163 |
+
}
|
| 1164 |
+
],
|
| 1165 |
+
"logging_steps": 10,
|
| 1166 |
+
"max_steps": 1602,
|
| 1167 |
+
"num_input_tokens_seen": 0,
|
| 1168 |
+
"num_train_epochs": 3,
|
| 1169 |
+
"save_steps": 500,
|
| 1170 |
+
"stateful_callbacks": {
|
| 1171 |
+
"TrainerControl": {
|
| 1172 |
+
"args": {
|
| 1173 |
+
"should_epoch_stop": false,
|
| 1174 |
+
"should_evaluate": false,
|
| 1175 |
+
"should_log": false,
|
| 1176 |
+
"should_save": true,
|
| 1177 |
+
"should_training_stop": true
|
| 1178 |
+
},
|
| 1179 |
+
"attributes": {}
|
| 1180 |
+
}
|
| 1181 |
+
},
|
| 1182 |
+
"total_flos": 2683409667194880.0,
|
| 1183 |
+
"train_batch_size": 16,
|
| 1184 |
+
"trial_name": null,
|
| 1185 |
+
"trial_params": null
|
| 1186 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|