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 +1053 -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-llama-3.1-70b
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-70b
|
| 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: 0.3472
|
| 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-llama-3.1-70b
|
|
|
|
| 16 |
|
| 17 |
# mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-70b
|
| 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-llama-3.1-70b dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.3472
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3471969962120056,
|
| 4 |
+
"eval_runtime": 49.0828,
|
| 5 |
+
"eval_samples_per_second": 259.643,
|
| 6 |
+
"eval_steps_per_second": 1.019,
|
| 7 |
+
"total_flos": 2376877851279360.0,
|
| 8 |
+
"train_loss": 0.25528628041493884,
|
| 9 |
+
"train_runtime": 8757.0427,
|
| 10 |
+
"train_samples_per_second": 82.948,
|
| 11 |
+
"train_steps_per_second": 0.162
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.3471969962120056,
|
| 4 |
+
"eval_runtime": 49.0828,
|
| 5 |
+
"eval_samples_per_second": 259.643,
|
| 6 |
+
"eval_steps_per_second": 1.019
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2376877851279360.0,
|
| 4 |
+
"train_loss": 0.25528628041493884,
|
| 5 |
+
"train_runtime": 8757.0427,
|
| 6 |
+
"train_samples_per_second": 82.948,
|
| 7 |
+
"train_steps_per_second": 0.162
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1053 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1419,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.021141649048625793,
|
| 13 |
+
"grad_norm": 6.24355232060165,
|
| 14 |
+
"learning_rate": 7.042253521126762e-07,
|
| 15 |
+
"loss": 0.5912,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.042283298097251586,
|
| 20 |
+
"grad_norm": 2.216096271272777,
|
| 21 |
+
"learning_rate": 1.4084507042253523e-06,
|
| 22 |
+
"loss": 0.5024,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06342494714587738,
|
| 27 |
+
"grad_norm": 1.449992996844573,
|
| 28 |
+
"learning_rate": 2.1126760563380285e-06,
|
| 29 |
+
"loss": 0.4488,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08456659619450317,
|
| 34 |
+
"grad_norm": 1.5865270130811833,
|
| 35 |
+
"learning_rate": 2.8169014084507046e-06,
|
| 36 |
+
"loss": 0.431,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10570824524312897,
|
| 41 |
+
"grad_norm": 1.6309436803659814,
|
| 42 |
+
"learning_rate": 3.5211267605633804e-06,
|
| 43 |
+
"loss": 0.4102,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12684989429175475,
|
| 48 |
+
"grad_norm": 1.3774805832619317,
|
| 49 |
+
"learning_rate": 4.225352112676057e-06,
|
| 50 |
+
"loss": 0.3953,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14799154334038056,
|
| 55 |
+
"grad_norm": 1.7489511540042915,
|
| 56 |
+
"learning_rate": 4.929577464788733e-06,
|
| 57 |
+
"loss": 0.392,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16913319238900634,
|
| 62 |
+
"grad_norm": 1.4907503413861205,
|
| 63 |
+
"learning_rate": 4.999505072506396e-06,
|
| 64 |
+
"loss": 0.3845,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.19027484143763213,
|
| 69 |
+
"grad_norm": 1.714741595800788,
|
| 70 |
+
"learning_rate": 4.997794491573208e-06,
|
| 71 |
+
"loss": 0.3754,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.21141649048625794,
|
| 76 |
+
"grad_norm": 1.6723995412498653,
|
| 77 |
+
"learning_rate": 4.994863075823568e-06,
|
| 78 |
+
"loss": 0.3731,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.23255813953488372,
|
| 83 |
+
"grad_norm": 1.559234855788911,
|
| 84 |
+
"learning_rate": 4.990712417384526e-06,
|
| 85 |
+
"loss": 0.3676,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2536997885835095,
|
| 90 |
+
"grad_norm": 1.57054985630799,
|
| 91 |
+
"learning_rate": 4.985344770585093e-06,
|
| 92 |
+
"loss": 0.3693,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2748414376321353,
|
| 97 |
+
"grad_norm": 1.6051226014645448,
|
| 98 |
+
"learning_rate": 4.9787630507318594e-06,
|
| 99 |
+
"loss": 0.3627,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2959830866807611,
|
| 104 |
+
"grad_norm": 1.4668663144082323,
|
| 105 |
+
"learning_rate": 4.9709708325256166e-06,
|
| 106 |
+
"loss": 0.3613,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3171247357293869,
|
| 111 |
+
"grad_norm": 1.686344018429158,
|
| 112 |
+
"learning_rate": 4.961972348119846e-06,
|
| 113 |
+
"loss": 0.3602,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3382663847780127,
|
| 118 |
+
"grad_norm": 1.7636035584330838,
|
| 119 |
+
"learning_rate": 4.951772484822128e-06,
|
| 120 |
+
"loss": 0.3568,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3594080338266385,
|
| 125 |
+
"grad_norm": 2.5501698658025536,
|
| 126 |
+
"learning_rate": 4.940376782439721e-06,
|
| 127 |
+
"loss": 0.3596,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.38054968287526425,
|
| 132 |
+
"grad_norm": 1.3764768803172454,
|
| 133 |
+
"learning_rate": 4.9277914302707466e-06,
|
| 134 |
+
"loss": 0.3574,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.40169133192389006,
|
| 139 |
+
"grad_norm": 1.6197573568256791,
|
| 140 |
+
"learning_rate": 4.914023263742626e-06,
|
| 141 |
+
"loss": 0.3514,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.42283298097251587,
|
| 146 |
+
"grad_norm": 1.7518377059862333,
|
| 147 |
+
"learning_rate": 4.8990797606995845e-06,
|
| 148 |
+
"loss": 0.3468,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4439746300211416,
|
| 153 |
+
"grad_norm": 1.507687286038259,
|
| 154 |
+
"learning_rate": 4.882969037341239e-06,
|
| 155 |
+
"loss": 0.3455,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.46511627906976744,
|
| 160 |
+
"grad_norm": 1.5970818330074303,
|
| 161 |
+
"learning_rate": 4.865699843814485e-06,
|
| 162 |
+
"loss": 0.3428,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.48625792811839325,
|
| 167 |
+
"grad_norm": 1.4210922325164967,
|
| 168 |
+
"learning_rate": 4.847281559461071e-06,
|
| 169 |
+
"loss": 0.3461,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.507399577167019,
|
| 174 |
+
"grad_norm": 1.3912747321946244,
|
| 175 |
+
"learning_rate": 4.827724187723427e-06,
|
| 176 |
+
"loss": 0.3429,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5285412262156448,
|
| 181 |
+
"grad_norm": 1.3955734367464652,
|
| 182 |
+
"learning_rate": 4.80703835071155e-06,
|
| 183 |
+
"loss": 0.3437,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5496828752642706,
|
| 188 |
+
"grad_norm": 1.350657312273654,
|
| 189 |
+
"learning_rate": 4.785235283433857e-06,
|
| 190 |
+
"loss": 0.3393,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5708245243128964,
|
| 195 |
+
"grad_norm": 1.2717206677797315,
|
| 196 |
+
"learning_rate": 4.762326827695163e-06,
|
| 197 |
+
"loss": 0.3411,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5919661733615222,
|
| 202 |
+
"grad_norm": 1.191569222191925,
|
| 203 |
+
"learning_rate": 4.7383254256650964e-06,
|
| 204 |
+
"loss": 0.3385,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6131078224101479,
|
| 209 |
+
"grad_norm": 1.1928472569401323,
|
| 210 |
+
"learning_rate": 4.713244113120443e-06,
|
| 211 |
+
"loss": 0.34,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6342494714587738,
|
| 216 |
+
"grad_norm": 1.3276992158951393,
|
| 217 |
+
"learning_rate": 4.687096512365067e-06,
|
| 218 |
+
"loss": 0.3421,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6553911205073996,
|
| 223 |
+
"grad_norm": 1.1887821846438988,
|
| 224 |
+
"learning_rate": 4.659896824831302e-06,
|
| 225 |
+
"loss": 0.3356,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6765327695560254,
|
| 230 |
+
"grad_norm": 1.2621729084813325,
|
| 231 |
+
"learning_rate": 4.631659823366783e-06,
|
| 232 |
+
"loss": 0.3403,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6976744186046512,
|
| 237 |
+
"grad_norm": 1.4771395957284508,
|
| 238 |
+
"learning_rate": 4.6024008442109335e-06,
|
| 239 |
+
"loss": 0.3346,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.718816067653277,
|
| 244 |
+
"grad_norm": 1.0838005637440695,
|
| 245 |
+
"learning_rate": 4.572135778665464e-06,
|
| 246 |
+
"loss": 0.3361,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7399577167019028,
|
| 251 |
+
"grad_norm": 1.1405549612151125,
|
| 252 |
+
"learning_rate": 4.5408810644633956e-06,
|
| 253 |
+
"loss": 0.3308,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7610993657505285,
|
| 258 |
+
"grad_norm": 1.111444530105349,
|
| 259 |
+
"learning_rate": 4.508653676841308e-06,
|
| 260 |
+
"loss": 0.3337,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7822410147991543,
|
| 265 |
+
"grad_norm": 1.1516361189573001,
|
| 266 |
+
"learning_rate": 4.475471119319651e-06,
|
| 267 |
+
"loss": 0.3279,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8033826638477801,
|
| 272 |
+
"grad_norm": 1.1394562536949298,
|
| 273 |
+
"learning_rate": 4.441351414196139e-06,
|
| 274 |
+
"loss": 0.329,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8245243128964059,
|
| 279 |
+
"grad_norm": 1.2624616377095017,
|
| 280 |
+
"learning_rate": 4.406313092757369e-06,
|
| 281 |
+
"loss": 0.3307,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8456659619450317,
|
| 286 |
+
"grad_norm": 1.3399500811254008,
|
| 287 |
+
"learning_rate": 4.370375185214014e-06,
|
| 288 |
+
"loss": 0.3308,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8668076109936576,
|
| 293 |
+
"grad_norm": 1.2917888536055562,
|
| 294 |
+
"learning_rate": 4.333557210365023e-06,
|
| 295 |
+
"loss": 0.3301,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8879492600422833,
|
| 300 |
+
"grad_norm": 1.1769502964733598,
|
| 301 |
+
"learning_rate": 4.295879164996462e-06,
|
| 302 |
+
"loss": 0.3271,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9090909090909091,
|
| 307 |
+
"grad_norm": 1.2781090385858154,
|
| 308 |
+
"learning_rate": 4.257361513020745e-06,
|
| 309 |
+
"loss": 0.3242,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9302325581395349,
|
| 314 |
+
"grad_norm": 1.2394021660894758,
|
| 315 |
+
"learning_rate": 4.218025174362161e-06,
|
| 316 |
+
"loss": 0.3236,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9513742071881607,
|
| 321 |
+
"grad_norm": 1.2387248179248829,
|
| 322 |
+
"learning_rate": 4.177891513594724e-06,
|
| 323 |
+
"loss": 0.3249,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9725158562367865,
|
| 328 |
+
"grad_norm": 1.2158882947276164,
|
| 329 |
+
"learning_rate": 4.136982328338531e-06,
|
| 330 |
+
"loss": 0.3245,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9936575052854123,
|
| 335 |
+
"grad_norm": 1.0357890278389636,
|
| 336 |
+
"learning_rate": 4.0953198374209045e-06,
|
| 337 |
+
"loss": 0.3209,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0,
|
| 342 |
+
"eval_loss": 0.3238590359687805,
|
| 343 |
+
"eval_runtime": 42.8216,
|
| 344 |
+
"eval_samples_per_second": 297.607,
|
| 345 |
+
"eval_steps_per_second": 1.168,
|
| 346 |
+
"step": 473
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.014799154334038,
|
| 350 |
+
"grad_norm": 1.7559135479092256,
|
| 351 |
+
"learning_rate": 4.052926668808791e-06,
|
| 352 |
+
"loss": 0.2656,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.0359408033826638,
|
| 357 |
+
"grad_norm": 1.5639568698805122,
|
| 358 |
+
"learning_rate": 4.009825847318922e-06,
|
| 359 |
+
"loss": 0.2439,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.0570824524312896,
|
| 364 |
+
"grad_norm": 1.1876279950755768,
|
| 365 |
+
"learning_rate": 3.966040782112451e-06,
|
| 366 |
+
"loss": 0.2422,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0782241014799154,
|
| 371 |
+
"grad_norm": 1.1202067555602988,
|
| 372 |
+
"learning_rate": 3.921595253980836e-06,
|
| 373 |
+
"loss": 0.2425,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.0993657505285412,
|
| 378 |
+
"grad_norm": 1.1539051669533806,
|
| 379 |
+
"learning_rate": 3.876513402429883e-06,
|
| 380 |
+
"loss": 0.2431,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.120507399577167,
|
| 385 |
+
"grad_norm": 1.199763607338339,
|
| 386 |
+
"learning_rate": 3.83081971256896e-06,
|
| 387 |
+
"loss": 0.2413,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.1416490486257929,
|
| 392 |
+
"grad_norm": 1.1190658369259245,
|
| 393 |
+
"learning_rate": 3.7845390018125065e-06,
|
| 394 |
+
"loss": 0.2454,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.1627906976744187,
|
| 399 |
+
"grad_norm": 1.116921299899802,
|
| 400 |
+
"learning_rate": 3.737696406401062e-06,
|
| 401 |
+
"loss": 0.2394,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.1839323467230445,
|
| 406 |
+
"grad_norm": 1.1477296149656728,
|
| 407 |
+
"learning_rate": 3.6903173677491266e-06,
|
| 408 |
+
"loss": 0.2448,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.20507399577167,
|
| 413 |
+
"grad_norm": 1.1454465891047556,
|
| 414 |
+
"learning_rate": 3.642427618627277e-06,
|
| 415 |
+
"loss": 0.2437,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.226215644820296,
|
| 420 |
+
"grad_norm": 1.1642513264879382,
|
| 421 |
+
"learning_rate": 3.5940531691860405e-06,
|
| 422 |
+
"loss": 0.2428,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.2473572938689217,
|
| 427 |
+
"grad_norm": 1.3763131648109288,
|
| 428 |
+
"learning_rate": 3.545220292829113e-06,
|
| 429 |
+
"loss": 0.2403,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.2684989429175475,
|
| 434 |
+
"grad_norm": 1.291772297216049,
|
| 435 |
+
"learning_rate": 3.4959555119436033e-06,
|
| 436 |
+
"loss": 0.2425,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.2896405919661733,
|
| 441 |
+
"grad_norm": 1.4742209496360983,
|
| 442 |
+
"learning_rate": 3.446285583495041e-06,
|
| 443 |
+
"loss": 0.2419,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.3107822410147991,
|
| 448 |
+
"grad_norm": 1.2162630598081021,
|
| 449 |
+
"learning_rate": 3.396237484494985e-06,
|
| 450 |
+
"loss": 0.2434,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.331923890063425,
|
| 455 |
+
"grad_norm": 1.1668893819388124,
|
| 456 |
+
"learning_rate": 3.345838397349115e-06,
|
| 457 |
+
"loss": 0.2413,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.3530655391120507,
|
| 462 |
+
"grad_norm": 1.1883780283974819,
|
| 463 |
+
"learning_rate": 3.2951156950937728e-06,
|
| 464 |
+
"loss": 0.243,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.3742071881606766,
|
| 469 |
+
"grad_norm": 1.2805764821322394,
|
| 470 |
+
"learning_rate": 3.2440969265289624e-06,
|
| 471 |
+
"loss": 0.2421,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.3953488372093024,
|
| 476 |
+
"grad_norm": 1.0727847278011404,
|
| 477 |
+
"learning_rate": 3.1928098012558895e-06,
|
| 478 |
+
"loss": 0.244,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.4164904862579282,
|
| 483 |
+
"grad_norm": 1.1764000750965518,
|
| 484 |
+
"learning_rate": 3.1412821746271693e-06,
|
| 485 |
+
"loss": 0.2439,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.437632135306554,
|
| 490 |
+
"grad_norm": 1.1953048969350113,
|
| 491 |
+
"learning_rate": 3.0895420326178705e-06,
|
| 492 |
+
"loss": 0.2408,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.4587737843551798,
|
| 497 |
+
"grad_norm": 1.2778095821404496,
|
| 498 |
+
"learning_rate": 3.03761747662561e-06,
|
| 499 |
+
"loss": 0.2423,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.4799154334038054,
|
| 504 |
+
"grad_norm": 1.2694623156337455,
|
| 505 |
+
"learning_rate": 2.985536708207971e-06,
|
| 506 |
+
"loss": 0.2401,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.5010570824524314,
|
| 511 |
+
"grad_norm": 1.1185495126584732,
|
| 512 |
+
"learning_rate": 2.933328013765505e-06,
|
| 513 |
+
"loss": 0.2449,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.522198731501057,
|
| 518 |
+
"grad_norm": 1.168267493603954,
|
| 519 |
+
"learning_rate": 2.8810197491786714e-06,
|
| 520 |
+
"loss": 0.2431,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.543340380549683,
|
| 525 |
+
"grad_norm": 1.1355953180673453,
|
| 526 |
+
"learning_rate": 2.8286403244070252e-06,
|
| 527 |
+
"loss": 0.2414,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.5644820295983086,
|
| 532 |
+
"grad_norm": 1.0758782490829244,
|
| 533 |
+
"learning_rate": 2.7762181880590442e-06,
|
| 534 |
+
"loss": 0.2415,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.5856236786469344,
|
| 539 |
+
"grad_norm": 1.1428398992616515,
|
| 540 |
+
"learning_rate": 2.723781811940956e-06,
|
| 541 |
+
"loss": 0.2431,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.6067653276955602,
|
| 546 |
+
"grad_norm": 1.0783252394855087,
|
| 547 |
+
"learning_rate": 2.6713596755929755e-06,
|
| 548 |
+
"loss": 0.2417,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.627906976744186,
|
| 553 |
+
"grad_norm": 1.0764260279790863,
|
| 554 |
+
"learning_rate": 2.6189802508213297e-06,
|
| 555 |
+
"loss": 0.2391,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.6490486257928119,
|
| 560 |
+
"grad_norm": 1.0866425685585581,
|
| 561 |
+
"learning_rate": 2.5666719862344957e-06,
|
| 562 |
+
"loss": 0.2445,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.6701902748414377,
|
| 567 |
+
"grad_norm": 1.0529013290697624,
|
| 568 |
+
"learning_rate": 2.51446329179203e-06,
|
| 569 |
+
"loss": 0.2409,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.6913319238900635,
|
| 574 |
+
"grad_norm": 1.1356539893062763,
|
| 575 |
+
"learning_rate": 2.4623825233743896e-06,
|
| 576 |
+
"loss": 0.241,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.712473572938689,
|
| 581 |
+
"grad_norm": 1.1111512278149636,
|
| 582 |
+
"learning_rate": 2.4104579673821297e-06,
|
| 583 |
+
"loss": 0.2434,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.733615221987315,
|
| 588 |
+
"grad_norm": 1.102236029943368,
|
| 589 |
+
"learning_rate": 2.358717825372831e-06,
|
| 590 |
+
"loss": 0.2413,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.7547568710359407,
|
| 595 |
+
"grad_norm": 1.124459202450407,
|
| 596 |
+
"learning_rate": 2.3071901987441116e-06,
|
| 597 |
+
"loss": 0.2385,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.7758985200845667,
|
| 602 |
+
"grad_norm": 1.2179881959913421,
|
| 603 |
+
"learning_rate": 2.2559030734710396e-06,
|
| 604 |
+
"loss": 0.2386,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.7970401691331923,
|
| 609 |
+
"grad_norm": 1.0767260671283274,
|
| 610 |
+
"learning_rate": 2.2048843049062275e-06,
|
| 611 |
+
"loss": 0.2398,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.8181818181818183,
|
| 616 |
+
"grad_norm": 1.1119843325482643,
|
| 617 |
+
"learning_rate": 2.1541616026508854e-06,
|
| 618 |
+
"loss": 0.2413,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.839323467230444,
|
| 623 |
+
"grad_norm": 1.0955361377120234,
|
| 624 |
+
"learning_rate": 2.103762515505016e-06,
|
| 625 |
+
"loss": 0.2417,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.8604651162790697,
|
| 630 |
+
"grad_norm": 1.0754774699879763,
|
| 631 |
+
"learning_rate": 2.0537144165049597e-06,
|
| 632 |
+
"loss": 0.236,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.8816067653276956,
|
| 637 |
+
"grad_norm": 1.116290127791819,
|
| 638 |
+
"learning_rate": 2.0040444880563974e-06,
|
| 639 |
+
"loss": 0.2396,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.9027484143763214,
|
| 644 |
+
"grad_norm": 1.0587161825711324,
|
| 645 |
+
"learning_rate": 1.9547797071708873e-06,
|
| 646 |
+
"loss": 0.2402,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.9238900634249472,
|
| 651 |
+
"grad_norm": 1.0735022958430327,
|
| 652 |
+
"learning_rate": 1.90594683081396e-06,
|
| 653 |
+
"loss": 0.2454,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.945031712473573,
|
| 658 |
+
"grad_norm": 1.0625353794584775,
|
| 659 |
+
"learning_rate": 1.8575723813727234e-06,
|
| 660 |
+
"loss": 0.241,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.9661733615221988,
|
| 665 |
+
"grad_norm": 1.0540753292992686,
|
| 666 |
+
"learning_rate": 1.8096826322508745e-06,
|
| 667 |
+
"loss": 0.2377,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.9873150105708244,
|
| 672 |
+
"grad_norm": 1.1469855369418602,
|
| 673 |
+
"learning_rate": 1.7623035935989388e-06,
|
| 674 |
+
"loss": 0.2363,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.0,
|
| 679 |
+
"eval_loss": 0.3176339268684387,
|
| 680 |
+
"eval_runtime": 46.7927,
|
| 681 |
+
"eval_samples_per_second": 272.35,
|
| 682 |
+
"eval_steps_per_second": 1.069,
|
| 683 |
+
"step": 946
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.0084566596194504,
|
| 687 |
+
"grad_norm": 1.442414950660575,
|
| 688 |
+
"learning_rate": 1.7154609981874945e-06,
|
| 689 |
+
"loss": 0.2124,
|
| 690 |
+
"step": 950
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.029598308668076,
|
| 694 |
+
"grad_norm": 1.3377936728948667,
|
| 695 |
+
"learning_rate": 1.6691802874310402e-06,
|
| 696 |
+
"loss": 0.1655,
|
| 697 |
+
"step": 960
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 2.050739957716702,
|
| 701 |
+
"grad_norm": 1.3199120792646895,
|
| 702 |
+
"learning_rate": 1.6234865975701169e-06,
|
| 703 |
+
"loss": 0.1695,
|
| 704 |
+
"step": 970
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 2.0718816067653276,
|
| 708 |
+
"grad_norm": 1.1515710499363918,
|
| 709 |
+
"learning_rate": 1.5784047460191638e-06,
|
| 710 |
+
"loss": 0.1628,
|
| 711 |
+
"step": 980
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.0930232558139537,
|
| 715 |
+
"grad_norm": 1.0930247346459778,
|
| 716 |
+
"learning_rate": 1.5339592178875502e-06,
|
| 717 |
+
"loss": 0.1629,
|
| 718 |
+
"step": 990
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.1141649048625792,
|
| 722 |
+
"grad_norm": 1.1196935797630683,
|
| 723 |
+
"learning_rate": 1.4901741526810793e-06,
|
| 724 |
+
"loss": 0.1632,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.1353065539112053,
|
| 729 |
+
"grad_norm": 1.266477815888195,
|
| 730 |
+
"learning_rate": 1.4470733311912094e-06,
|
| 731 |
+
"loss": 0.1648,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.156448202959831,
|
| 736 |
+
"grad_norm": 1.2032366464614765,
|
| 737 |
+
"learning_rate": 1.4046801625790954e-06,
|
| 738 |
+
"loss": 0.1635,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.177589852008457,
|
| 743 |
+
"grad_norm": 1.1118635376846853,
|
| 744 |
+
"learning_rate": 1.3630176716614696e-06,
|
| 745 |
+
"loss": 0.1645,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.1987315010570825,
|
| 750 |
+
"grad_norm": 1.1762161662179742,
|
| 751 |
+
"learning_rate": 1.3221084864052755e-06,
|
| 752 |
+
"loss": 0.167,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.219873150105708,
|
| 757 |
+
"grad_norm": 1.148476335813781,
|
| 758 |
+
"learning_rate": 1.2819748256378395e-06,
|
| 759 |
+
"loss": 0.1621,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.241014799154334,
|
| 764 |
+
"grad_norm": 1.1550786530748482,
|
| 765 |
+
"learning_rate": 1.2426384869792552e-06,
|
| 766 |
+
"loss": 0.1626,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.2621564482029597,
|
| 771 |
+
"grad_norm": 1.1566446795682568,
|
| 772 |
+
"learning_rate": 1.2041208350035386e-06,
|
| 773 |
+
"loss": 0.1642,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.2832980972515857,
|
| 778 |
+
"grad_norm": 1.1410532150519583,
|
| 779 |
+
"learning_rate": 1.1664427896349775e-06,
|
| 780 |
+
"loss": 0.1642,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.3044397463002113,
|
| 785 |
+
"grad_norm": 1.1400713733613699,
|
| 786 |
+
"learning_rate": 1.129624814785987e-06,
|
| 787 |
+
"loss": 0.1647,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.3255813953488373,
|
| 792 |
+
"grad_norm": 1.0699436865402963,
|
| 793 |
+
"learning_rate": 1.0936869072426324e-06,
|
| 794 |
+
"loss": 0.163,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.346723044397463,
|
| 799 |
+
"grad_norm": 1.1541436695499816,
|
| 800 |
+
"learning_rate": 1.0586485858038623e-06,
|
| 801 |
+
"loss": 0.1626,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.367864693446089,
|
| 806 |
+
"grad_norm": 1.125483674986556,
|
| 807 |
+
"learning_rate": 1.0245288806803492e-06,
|
| 808 |
+
"loss": 0.1669,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.3890063424947146,
|
| 813 |
+
"grad_norm": 1.1026248670403718,
|
| 814 |
+
"learning_rate": 9.913463231586928e-07,
|
| 815 |
+
"loss": 0.1629,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.41014799154334,
|
| 820 |
+
"grad_norm": 1.159203695893513,
|
| 821 |
+
"learning_rate": 9.591189355366054e-07,
|
| 822 |
+
"loss": 0.1644,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.431289640591966,
|
| 827 |
+
"grad_norm": 1.1077004937863641,
|
| 828 |
+
"learning_rate": 9.278642213345369e-07,
|
| 829 |
+
"loss": 0.167,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.452431289640592,
|
| 834 |
+
"grad_norm": 1.11182862218811,
|
| 835 |
+
"learning_rate": 8.975991557890668e-07,
|
| 836 |
+
"loss": 0.1638,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.473572938689218,
|
| 841 |
+
"grad_norm": 1.1755110000283455,
|
| 842 |
+
"learning_rate": 8.683401766332171e-07,
|
| 843 |
+
"loss": 0.1642,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.4947145877378434,
|
| 848 |
+
"grad_norm": 1.1475624161633082,
|
| 849 |
+
"learning_rate": 8.401031751686978e-07,
|
| 850 |
+
"loss": 0.1676,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.5158562367864694,
|
| 855 |
+
"grad_norm": 1.148787879127806,
|
| 856 |
+
"learning_rate": 8.129034876349334e-07,
|
| 857 |
+
"loss": 0.1644,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.536997885835095,
|
| 862 |
+
"grad_norm": 1.190372203322852,
|
| 863 |
+
"learning_rate": 7.867558868795578e-07,
|
| 864 |
+
"loss": 0.165,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.558139534883721,
|
| 869 |
+
"grad_norm": 1.1595867121561987,
|
| 870 |
+
"learning_rate": 7.616745743349038e-07,
|
| 871 |
+
"loss": 0.1638,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.5792811839323466,
|
| 876 |
+
"grad_norm": 1.1432214797090672,
|
| 877 |
+
"learning_rate": 7.376731723048383e-07,
|
| 878 |
+
"loss": 0.1652,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.6004228329809727,
|
| 883 |
+
"grad_norm": 1.1142453334609799,
|
| 884 |
+
"learning_rate": 7.147647165661439e-07,
|
| 885 |
+
"loss": 0.1636,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 2.6215644820295982,
|
| 890 |
+
"grad_norm": 1.0990743293486216,
|
| 891 |
+
"learning_rate": 6.929616492884497e-07,
|
| 892 |
+
"loss": 0.1625,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.6427061310782243,
|
| 897 |
+
"grad_norm": 1.0991366585298548,
|
| 898 |
+
"learning_rate": 6.722758122765728e-07,
|
| 899 |
+
"loss": 0.1638,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.66384778012685,
|
| 904 |
+
"grad_norm": 1.1273149039763855,
|
| 905 |
+
"learning_rate": 6.527184405389298e-07,
|
| 906 |
+
"loss": 0.1611,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.6849894291754755,
|
| 911 |
+
"grad_norm": 1.101167246100859,
|
| 912 |
+
"learning_rate": 6.343001561855149e-07,
|
| 913 |
+
"loss": 0.1605,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.7061310782241015,
|
| 918 |
+
"grad_norm": 1.0867484989228724,
|
| 919 |
+
"learning_rate": 6.17030962658762e-07,
|
| 920 |
+
"loss": 0.1655,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.7272727272727275,
|
| 925 |
+
"grad_norm": 1.154566269426987,
|
| 926 |
+
"learning_rate": 6.009202393004164e-07,
|
| 927 |
+
"loss": 0.1636,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.748414376321353,
|
| 932 |
+
"grad_norm": 1.1187041716958608,
|
| 933 |
+
"learning_rate": 5.859767362573741e-07,
|
| 934 |
+
"loss": 0.1644,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.7695560253699787,
|
| 939 |
+
"grad_norm": 1.1267150964681174,
|
| 940 |
+
"learning_rate": 5.722085697292536e-07,
|
| 941 |
+
"loss": 0.1659,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.7906976744186047,
|
| 946 |
+
"grad_norm": 1.0795591650211436,
|
| 947 |
+
"learning_rate": 5.596232175602791e-07,
|
| 948 |
+
"loss": 0.1667,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.8118393234672303,
|
| 953 |
+
"grad_norm": 1.060071714533849,
|
| 954 |
+
"learning_rate": 5.482275151778719e-07,
|
| 955 |
+
"loss": 0.164,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.8329809725158563,
|
| 960 |
+
"grad_norm": 1.10306526157592,
|
| 961 |
+
"learning_rate": 5.380276518801548e-07,
|
| 962 |
+
"loss": 0.1632,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.854122621564482,
|
| 967 |
+
"grad_norm": 1.0906207707452118,
|
| 968 |
+
"learning_rate": 5.290291674743844e-07,
|
| 969 |
+
"loss": 0.1642,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 2.875264270613108,
|
| 974 |
+
"grad_norm": 1.0845536549774597,
|
| 975 |
+
"learning_rate": 5.212369492681416e-07,
|
| 976 |
+
"loss": 0.1641,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.8964059196617336,
|
| 981 |
+
"grad_norm": 1.1569339067127147,
|
| 982 |
+
"learning_rate": 5.146552294149075e-07,
|
| 983 |
+
"loss": 0.1629,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.9175475687103596,
|
| 988 |
+
"grad_norm": 1.064398221843151,
|
| 989 |
+
"learning_rate": 5.092875826154744e-07,
|
| 990 |
+
"loss": 0.1648,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.938689217758985,
|
| 995 |
+
"grad_norm": 1.093531486342711,
|
| 996 |
+
"learning_rate": 5.051369241764327e-07,
|
| 997 |
+
"loss": 0.1621,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.9598308668076108,
|
| 1002 |
+
"grad_norm": 1.0796661303223472,
|
| 1003 |
+
"learning_rate": 5.022055084267933e-07,
|
| 1004 |
+
"loss": 0.1623,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.980972515856237,
|
| 1009 |
+
"grad_norm": 1.0990164115583458,
|
| 1010 |
+
"learning_rate": 5.004949274936044e-07,
|
| 1011 |
+
"loss": 0.1634,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 3.0,
|
| 1016 |
+
"eval_loss": 0.3471969962120056,
|
| 1017 |
+
"eval_runtime": 49.2029,
|
| 1018 |
+
"eval_samples_per_second": 259.009,
|
| 1019 |
+
"eval_steps_per_second": 1.016,
|
| 1020 |
+
"step": 1419
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 3.0,
|
| 1024 |
+
"step": 1419,
|
| 1025 |
+
"total_flos": 2376877851279360.0,
|
| 1026 |
+
"train_loss": 0.25528628041493884,
|
| 1027 |
+
"train_runtime": 8757.0427,
|
| 1028 |
+
"train_samples_per_second": 82.948,
|
| 1029 |
+
"train_steps_per_second": 0.162
|
| 1030 |
+
}
|
| 1031 |
+
],
|
| 1032 |
+
"logging_steps": 10,
|
| 1033 |
+
"max_steps": 1419,
|
| 1034 |
+
"num_input_tokens_seen": 0,
|
| 1035 |
+
"num_train_epochs": 3,
|
| 1036 |
+
"save_steps": 500,
|
| 1037 |
+
"stateful_callbacks": {
|
| 1038 |
+
"TrainerControl": {
|
| 1039 |
+
"args": {
|
| 1040 |
+
"should_epoch_stop": false,
|
| 1041 |
+
"should_evaluate": false,
|
| 1042 |
+
"should_log": false,
|
| 1043 |
+
"should_save": true,
|
| 1044 |
+
"should_training_stop": true
|
| 1045 |
+
},
|
| 1046 |
+
"attributes": {}
|
| 1047 |
+
}
|
| 1048 |
+
},
|
| 1049 |
+
"total_flos": 2376877851279360.0,
|
| 1050 |
+
"train_batch_size": 16,
|
| 1051 |
+
"trial_name": null,
|
| 1052 |
+
"trial_params": null
|
| 1053 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|