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 +1473 -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-v0.1
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: hp_ablations_grid_mistral_bsz512_lr5e-6_scheduler-cosine-warmup0.05-minlr5e-7-mistralv0.3
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_grid_mistral_bsz512_lr5e-6_scheduler-cosine-warmup0.05-minlr5e-7-mistralv0.3
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.0588
|
| 21 |
|
|
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.1
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: hp_ablations_grid_mistral_bsz512_lr5e-6_scheduler-cosine-warmup0.05-minlr5e-7-mistralv0.3
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_grid_mistral_bsz512_lr5e-6_scheduler-cosine-warmup0.05-minlr5e-7-mistralv0.3
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mlfoundations-dev/oh-dcft-v3-llama3.1-nemotron-70b_shareGPT_format dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.0588
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.997953488372093,
|
| 3 |
+
"eval_loss": 0.05878664180636406,
|
| 4 |
+
"eval_runtime": 450.0144,
|
| 5 |
+
"eval_samples_per_second": 40.228,
|
| 6 |
+
"eval_steps_per_second": 0.629,
|
| 7 |
+
"total_flos": 3371640595415040.0,
|
| 8 |
+
"train_loss": 0.382664002355982,
|
| 9 |
+
"train_runtime": 66379.5775,
|
| 10 |
+
"train_samples_per_second": 15.545,
|
| 11 |
+
"train_steps_per_second": 0.03
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.997953488372093,
|
| 3 |
+
"eval_loss": 0.05878664180636406,
|
| 4 |
+
"eval_runtime": 450.0144,
|
| 5 |
+
"eval_samples_per_second": 40.228,
|
| 6 |
+
"eval_steps_per_second": 0.629
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.997953488372093,
|
| 3 |
+
"total_flos": 3371640595415040.0,
|
| 4 |
+
"train_loss": 0.382664002355982,
|
| 5 |
+
"train_runtime": 66379.5775,
|
| 6 |
+
"train_samples_per_second": 15.545,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1473 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.997953488372093,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2013,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.014883720930232559,
|
| 13 |
+
"grad_norm": 8.902280630176788,
|
| 14 |
+
"learning_rate": 4.950495049504951e-07,
|
| 15 |
+
"loss": 0.8797,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.029767441860465118,
|
| 20 |
+
"grad_norm": 3.3029491558899418,
|
| 21 |
+
"learning_rate": 9.900990099009902e-07,
|
| 22 |
+
"loss": 0.774,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.044651162790697675,
|
| 27 |
+
"grad_norm": 1.9413977734691883,
|
| 28 |
+
"learning_rate": 1.4851485148514852e-06,
|
| 29 |
+
"loss": 0.6921,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.059534883720930236,
|
| 34 |
+
"grad_norm": 1.5937915147958766,
|
| 35 |
+
"learning_rate": 1.9801980198019803e-06,
|
| 36 |
+
"loss": 0.6479,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07441860465116279,
|
| 41 |
+
"grad_norm": 1.6800049101706125,
|
| 42 |
+
"learning_rate": 2.4752475247524753e-06,
|
| 43 |
+
"loss": 0.6181,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08930232558139535,
|
| 48 |
+
"grad_norm": 2.31629043986904,
|
| 49 |
+
"learning_rate": 2.9702970297029703e-06,
|
| 50 |
+
"loss": 0.598,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.10418604651162791,
|
| 55 |
+
"grad_norm": 1.813171169001016,
|
| 56 |
+
"learning_rate": 3.4653465346534653e-06,
|
| 57 |
+
"loss": 0.5869,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.11906976744186047,
|
| 62 |
+
"grad_norm": 2.4807755593084577,
|
| 63 |
+
"learning_rate": 3.960396039603961e-06,
|
| 64 |
+
"loss": 0.5731,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.13395348837209303,
|
| 69 |
+
"grad_norm": 2.627244335259517,
|
| 70 |
+
"learning_rate": 4.455445544554456e-06,
|
| 71 |
+
"loss": 0.5715,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.14883720930232558,
|
| 76 |
+
"grad_norm": 2.1854899385391042,
|
| 77 |
+
"learning_rate": 4.950495049504951e-06,
|
| 78 |
+
"loss": 0.5589,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16372093023255813,
|
| 83 |
+
"grad_norm": 2.1643320968352504,
|
| 84 |
+
"learning_rate": 4.999753989526703e-06,
|
| 85 |
+
"loss": 0.5591,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1786046511627907,
|
| 90 |
+
"grad_norm": 1.6906833870682114,
|
| 91 |
+
"learning_rate": 4.998903652018798e-06,
|
| 92 |
+
"loss": 0.5473,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19348837209302325,
|
| 97 |
+
"grad_norm": 1.6954868322135426,
|
| 98 |
+
"learning_rate": 4.997446179820209e-06,
|
| 99 |
+
"loss": 0.5418,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.20837209302325582,
|
| 104 |
+
"grad_norm": 3.1018130514636666,
|
| 105 |
+
"learning_rate": 4.995381966403521e-06,
|
| 106 |
+
"loss": 0.5392,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22325581395348837,
|
| 111 |
+
"grad_norm": 2.249001753991247,
|
| 112 |
+
"learning_rate": 4.9927115690427536e-06,
|
| 113 |
+
"loss": 0.538,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.23813953488372094,
|
| 118 |
+
"grad_norm": 2.5823816443281173,
|
| 119 |
+
"learning_rate": 4.989435708662909e-06,
|
| 120 |
+
"loss": 0.5337,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.25302325581395346,
|
| 125 |
+
"grad_norm": 2.271030426623902,
|
| 126 |
+
"learning_rate": 4.985555269645351e-06,
|
| 127 |
+
"loss": 0.5236,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.26790697674418606,
|
| 132 |
+
"grad_norm": 2.318489677318734,
|
| 133 |
+
"learning_rate": 4.981071299589047e-06,
|
| 134 |
+
"loss": 0.5245,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2827906976744186,
|
| 139 |
+
"grad_norm": 1.9206844645280905,
|
| 140 |
+
"learning_rate": 4.975985009027748e-06,
|
| 141 |
+
"loss": 0.5218,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.29767441860465116,
|
| 146 |
+
"grad_norm": 2.9110235751406983,
|
| 147 |
+
"learning_rate": 4.970297771103183e-06,
|
| 148 |
+
"loss": 0.5225,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3125581395348837,
|
| 153 |
+
"grad_norm": 2.4648347140615376,
|
| 154 |
+
"learning_rate": 4.964011121194349e-06,
|
| 155 |
+
"loss": 0.5125,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.32744186046511625,
|
| 160 |
+
"grad_norm": 1.8762938609852826,
|
| 161 |
+
"learning_rate": 4.957126756503014e-06,
|
| 162 |
+
"loss": 0.5119,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.34232558139534885,
|
| 167 |
+
"grad_norm": 2.634519062843505,
|
| 168 |
+
"learning_rate": 4.949646535595514e-06,
|
| 169 |
+
"loss": 0.5089,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3572093023255814,
|
| 174 |
+
"grad_norm": 2.7133170705876712,
|
| 175 |
+
"learning_rate": 4.941572477901008e-06,
|
| 176 |
+
"loss": 0.5028,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.37209302325581395,
|
| 181 |
+
"grad_norm": 1.772078056634877,
|
| 182 |
+
"learning_rate": 4.932906763166286e-06,
|
| 183 |
+
"loss": 0.5036,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3869767441860465,
|
| 188 |
+
"grad_norm": 2.3772772619843856,
|
| 189 |
+
"learning_rate": 4.9236517308673135e-06,
|
| 190 |
+
"loss": 0.5051,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.4018604651162791,
|
| 195 |
+
"grad_norm": 1.7586683386687694,
|
| 196 |
+
"learning_rate": 4.9138098795776335e-06,
|
| 197 |
+
"loss": 0.4996,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.41674418604651164,
|
| 202 |
+
"grad_norm": 2.441721431571048,
|
| 203 |
+
"learning_rate": 4.903383866293839e-06,
|
| 204 |
+
"loss": 0.5003,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4316279069767442,
|
| 209 |
+
"grad_norm": 1.6335890264070398,
|
| 210 |
+
"learning_rate": 4.89237650571826e-06,
|
| 211 |
+
"loss": 0.5009,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.44651162790697674,
|
| 216 |
+
"grad_norm": 1.8052385644508027,
|
| 217 |
+
"learning_rate": 4.880790769499083e-06,
|
| 218 |
+
"loss": 0.4983,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4613953488372093,
|
| 223 |
+
"grad_norm": 1.800037297627897,
|
| 224 |
+
"learning_rate": 4.868629785428096e-06,
|
| 225 |
+
"loss": 0.497,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4762790697674419,
|
| 230 |
+
"grad_norm": 1.605966469734369,
|
| 231 |
+
"learning_rate": 4.855896836596282e-06,
|
| 232 |
+
"loss": 0.4929,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.49116279069767443,
|
| 237 |
+
"grad_norm": 1.994957648259566,
|
| 238 |
+
"learning_rate": 4.842595360507486e-06,
|
| 239 |
+
"loss": 0.4966,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5060465116279069,
|
| 244 |
+
"grad_norm": 1.7903983897538154,
|
| 245 |
+
"learning_rate": 4.828728948150395e-06,
|
| 246 |
+
"loss": 0.4948,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5209302325581395,
|
| 251 |
+
"grad_norm": 1.5566561390721225,
|
| 252 |
+
"learning_rate": 4.8143013430290805e-06,
|
| 253 |
+
"loss": 0.4918,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5358139534883721,
|
| 258 |
+
"grad_norm": 1.449953579251504,
|
| 259 |
+
"learning_rate": 4.799316440152367e-06,
|
| 260 |
+
"loss": 0.4899,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5506976744186046,
|
| 265 |
+
"grad_norm": 1.945851570880214,
|
| 266 |
+
"learning_rate": 4.783778284982303e-06,
|
| 267 |
+
"loss": 0.4859,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5655813953488372,
|
| 272 |
+
"grad_norm": 1.6223980803827658,
|
| 273 |
+
"learning_rate": 4.767691072342006e-06,
|
| 274 |
+
"loss": 0.4884,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5804651162790697,
|
| 279 |
+
"grad_norm": 1.8375917251155103,
|
| 280 |
+
"learning_rate": 4.7510591452831975e-06,
|
| 281 |
+
"loss": 0.4809,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5953488372093023,
|
| 286 |
+
"grad_norm": 1.742322917798526,
|
| 287 |
+
"learning_rate": 4.733886993913704e-06,
|
| 288 |
+
"loss": 0.4857,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6102325581395349,
|
| 293 |
+
"grad_norm": 1.6356424508667176,
|
| 294 |
+
"learning_rate": 4.7161792541852675e-06,
|
| 295 |
+
"loss": 0.4872,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6251162790697674,
|
| 300 |
+
"grad_norm": 1.5526421294000088,
|
| 301 |
+
"learning_rate": 4.69794070664199e-06,
|
| 302 |
+
"loss": 0.4829,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.64,
|
| 307 |
+
"grad_norm": 1.4232942412821412,
|
| 308 |
+
"learning_rate": 4.6791762751297236e-06,
|
| 309 |
+
"loss": 0.481,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6548837209302325,
|
| 314 |
+
"grad_norm": 1.483668935893611,
|
| 315 |
+
"learning_rate": 4.65989102546679e-06,
|
| 316 |
+
"loss": 0.4821,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6697674418604651,
|
| 321 |
+
"grad_norm": 2.1539928952738405,
|
| 322 |
+
"learning_rate": 4.640090164076361e-06,
|
| 323 |
+
"loss": 0.4749,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6846511627906977,
|
| 328 |
+
"grad_norm": 2.1025050973956767,
|
| 329 |
+
"learning_rate": 4.61977903658089e-06,
|
| 330 |
+
"loss": 0.4804,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6995348837209302,
|
| 335 |
+
"grad_norm": 1.7519293478931726,
|
| 336 |
+
"learning_rate": 4.5989631263589546e-06,
|
| 337 |
+
"loss": 0.4743,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7144186046511628,
|
| 342 |
+
"grad_norm": 2.0423876723189243,
|
| 343 |
+
"learning_rate": 4.5776480530649155e-06,
|
| 344 |
+
"loss": 0.4726,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7293023255813953,
|
| 349 |
+
"grad_norm": 1.771995353632526,
|
| 350 |
+
"learning_rate": 4.555839571111782e-06,
|
| 351 |
+
"loss": 0.4728,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7441860465116279,
|
| 356 |
+
"grad_norm": 1.7344799986510258,
|
| 357 |
+
"learning_rate": 4.533543568117697e-06,
|
| 358 |
+
"loss": 0.4725,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.7590697674418605,
|
| 363 |
+
"grad_norm": 1.3702958835212744,
|
| 364 |
+
"learning_rate": 4.5107660633164645e-06,
|
| 365 |
+
"loss": 0.475,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.773953488372093,
|
| 370 |
+
"grad_norm": 1.5331038199927272,
|
| 371 |
+
"learning_rate": 4.487513205932537e-06,
|
| 372 |
+
"loss": 0.4758,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7888372093023256,
|
| 377 |
+
"grad_norm": 1.4415751132180146,
|
| 378 |
+
"learning_rate": 4.46379127352092e-06,
|
| 379 |
+
"loss": 0.4651,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.8037209302325582,
|
| 384 |
+
"grad_norm": 1.5387474607067084,
|
| 385 |
+
"learning_rate": 4.439606670272421e-06,
|
| 386 |
+
"loss": 0.4717,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8186046511627907,
|
| 391 |
+
"grad_norm": 1.3930637138971305,
|
| 392 |
+
"learning_rate": 4.414965925284719e-06,
|
| 393 |
+
"loss": 0.4683,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8334883720930233,
|
| 398 |
+
"grad_norm": 1.4357407382473695,
|
| 399 |
+
"learning_rate": 4.389875690799706e-06,
|
| 400 |
+
"loss": 0.4705,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8483720930232558,
|
| 405 |
+
"grad_norm": 1.867160837622047,
|
| 406 |
+
"learning_rate": 4.364342740407589e-06,
|
| 407 |
+
"loss": 0.4684,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8632558139534884,
|
| 412 |
+
"grad_norm": 2.079586326625279,
|
| 413 |
+
"learning_rate": 4.338373967218229e-06,
|
| 414 |
+
"loss": 0.4629,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.878139534883721,
|
| 419 |
+
"grad_norm": 1.6049505438520633,
|
| 420 |
+
"learning_rate": 4.3119763820002105e-06,
|
| 421 |
+
"loss": 0.4643,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8930232558139535,
|
| 426 |
+
"grad_norm": 1.2874061526414027,
|
| 427 |
+
"learning_rate": 4.285157111288156e-06,
|
| 428 |
+
"loss": 0.4642,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9079069767441861,
|
| 433 |
+
"grad_norm": 1.545108445973155,
|
| 434 |
+
"learning_rate": 4.257923395458778e-06,
|
| 435 |
+
"loss": 0.4606,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9227906976744186,
|
| 440 |
+
"grad_norm": 1.6451186302266136,
|
| 441 |
+
"learning_rate": 4.230282586776198e-06,
|
| 442 |
+
"loss": 0.4584,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.9376744186046512,
|
| 447 |
+
"grad_norm": 2.1325124535031263,
|
| 448 |
+
"learning_rate": 4.202242147407065e-06,
|
| 449 |
+
"loss": 0.4621,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9525581395348838,
|
| 454 |
+
"grad_norm": 2.0011178228576862,
|
| 455 |
+
"learning_rate": 4.173809647406001e-06,
|
| 456 |
+
"loss": 0.4601,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9674418604651163,
|
| 461 |
+
"grad_norm": 1.5742244911614134,
|
| 462 |
+
"learning_rate": 4.1449927626719164e-06,
|
| 463 |
+
"loss": 0.456,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.9823255813953489,
|
| 468 |
+
"grad_norm": 1.7386928439484917,
|
| 469 |
+
"learning_rate": 4.115799272875756e-06,
|
| 470 |
+
"loss": 0.4548,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.9972093023255814,
|
| 475 |
+
"grad_norm": 1.5311577866040798,
|
| 476 |
+
"learning_rate": 4.086237059360228e-06,
|
| 477 |
+
"loss": 0.4624,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.9986976744186047,
|
| 482 |
+
"eval_loss": 0.05731714889407158,
|
| 483 |
+
"eval_runtime": 455.161,
|
| 484 |
+
"eval_samples_per_second": 39.773,
|
| 485 |
+
"eval_steps_per_second": 0.622,
|
| 486 |
+
"step": 671
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.0130232558139536,
|
| 490 |
+
"grad_norm": 1.9657672125173562,
|
| 491 |
+
"learning_rate": 4.056314103012081e-06,
|
| 492 |
+
"loss": 0.3827,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.027906976744186,
|
| 497 |
+
"grad_norm": 1.6977560049208136,
|
| 498 |
+
"learning_rate": 4.026038482107515e-06,
|
| 499 |
+
"loss": 0.3642,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.0427906976744186,
|
| 504 |
+
"grad_norm": 1.7125394386865773,
|
| 505 |
+
"learning_rate": 3.995418370131294e-06,
|
| 506 |
+
"loss": 0.3649,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.0576744186046512,
|
| 511 |
+
"grad_norm": 1.8561895728990827,
|
| 512 |
+
"learning_rate": 3.964462033570154e-06,
|
| 513 |
+
"loss": 0.3662,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.0725581395348838,
|
| 518 |
+
"grad_norm": 1.5618111186332968,
|
| 519 |
+
"learning_rate": 3.9331778296811126e-06,
|
| 520 |
+
"loss": 0.3658,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.0874418604651164,
|
| 525 |
+
"grad_norm": 1.6589684668519427,
|
| 526 |
+
"learning_rate": 3.9015742042352575e-06,
|
| 527 |
+
"loss": 0.3633,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.1023255813953488,
|
| 532 |
+
"grad_norm": 1.6737692269771278,
|
| 533 |
+
"learning_rate": 3.8696596892376615e-06,
|
| 534 |
+
"loss": 0.3683,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.1172093023255814,
|
| 539 |
+
"grad_norm": 1.9826343984663213,
|
| 540 |
+
"learning_rate": 3.8374429006239915e-06,
|
| 541 |
+
"loss": 0.366,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.132093023255814,
|
| 546 |
+
"grad_norm": 1.7884778638246734,
|
| 547 |
+
"learning_rate": 3.8049325359344804e-06,
|
| 548 |
+
"loss": 0.3692,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1469767441860466,
|
| 553 |
+
"grad_norm": 2.611566598798586,
|
| 554 |
+
"learning_rate": 3.7721373719658526e-06,
|
| 555 |
+
"loss": 0.3712,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.1618604651162792,
|
| 560 |
+
"grad_norm": 1.7457709366563403,
|
| 561 |
+
"learning_rate": 3.7390662624018648e-06,
|
| 562 |
+
"loss": 0.3693,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.1767441860465115,
|
| 567 |
+
"grad_norm": 1.9569418242995398,
|
| 568 |
+
"learning_rate": 3.7057281354230794e-06,
|
| 569 |
+
"loss": 0.3653,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.1916279069767441,
|
| 574 |
+
"grad_norm": 2.173634628655407,
|
| 575 |
+
"learning_rate": 3.6721319912965366e-06,
|
| 576 |
+
"loss": 0.3649,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.2065116279069767,
|
| 581 |
+
"grad_norm": 2.2801578585083324,
|
| 582 |
+
"learning_rate": 3.6382868999459524e-06,
|
| 583 |
+
"loss": 0.3685,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.2213953488372093,
|
| 588 |
+
"grad_norm": 2.3692062383608197,
|
| 589 |
+
"learning_rate": 3.6042019985031244e-06,
|
| 590 |
+
"loss": 0.3712,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.236279069767442,
|
| 595 |
+
"grad_norm": 2.3703941397431945,
|
| 596 |
+
"learning_rate": 3.569886488841187e-06,
|
| 597 |
+
"loss": 0.3659,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.2511627906976743,
|
| 602 |
+
"grad_norm": 1.8705318246188423,
|
| 603 |
+
"learning_rate": 3.535349635090386e-06,
|
| 604 |
+
"loss": 0.3682,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.266046511627907,
|
| 609 |
+
"grad_norm": 1.404469365060906,
|
| 610 |
+
"learning_rate": 3.5006007611370513e-06,
|
| 611 |
+
"loss": 0.3662,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.2809302325581395,
|
| 616 |
+
"grad_norm": 1.797601704761619,
|
| 617 |
+
"learning_rate": 3.465649248106435e-06,
|
| 618 |
+
"loss": 0.3661,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.2958139534883721,
|
| 623 |
+
"grad_norm": 1.8585619220995928,
|
| 624 |
+
"learning_rate": 3.4305045318300974e-06,
|
| 625 |
+
"loss": 0.3647,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.3106976744186047,
|
| 630 |
+
"grad_norm": 1.4765933716397583,
|
| 631 |
+
"learning_rate": 3.3951761002985184e-06,
|
| 632 |
+
"loss": 0.3673,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.3255813953488373,
|
| 637 |
+
"grad_norm": 1.4125161732202964,
|
| 638 |
+
"learning_rate": 3.3596734910996397e-06,
|
| 639 |
+
"loss": 0.3642,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.3404651162790697,
|
| 644 |
+
"grad_norm": 1.4190852276081691,
|
| 645 |
+
"learning_rate": 3.3240062888440046e-06,
|
| 646 |
+
"loss": 0.3688,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.3553488372093023,
|
| 651 |
+
"grad_norm": 1.7022894617337727,
|
| 652 |
+
"learning_rate": 3.2881841225772097e-06,
|
| 653 |
+
"loss": 0.3697,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.370232558139535,
|
| 658 |
+
"grad_norm": 1.4899520481456272,
|
| 659 |
+
"learning_rate": 3.2522166631803616e-06,
|
| 660 |
+
"loss": 0.371,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.3851162790697673,
|
| 665 |
+
"grad_norm": 1.5320087948001035,
|
| 666 |
+
"learning_rate": 3.2161136207592323e-06,
|
| 667 |
+
"loss": 0.3661,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.4,
|
| 672 |
+
"grad_norm": 1.5107565046345355,
|
| 673 |
+
"learning_rate": 3.1798847420228358e-06,
|
| 674 |
+
"loss": 0.3716,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.4148837209302325,
|
| 679 |
+
"grad_norm": 1.4964498805556543,
|
| 680 |
+
"learning_rate": 3.14353980765211e-06,
|
| 681 |
+
"loss": 0.3681,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.4297674418604651,
|
| 686 |
+
"grad_norm": 1.4602567799649149,
|
| 687 |
+
"learning_rate": 3.1070886296594427e-06,
|
| 688 |
+
"loss": 0.367,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.4446511627906977,
|
| 693 |
+
"grad_norm": 1.5151588768049005,
|
| 694 |
+
"learning_rate": 3.0705410487397214e-06,
|
| 695 |
+
"loss": 0.3634,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.4595348837209303,
|
| 700 |
+
"grad_norm": 1.4429762741417749,
|
| 701 |
+
"learning_rate": 3.0339069316136573e-06,
|
| 702 |
+
"loss": 0.3692,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.474418604651163,
|
| 707 |
+
"grad_norm": 1.795013045267996,
|
| 708 |
+
"learning_rate": 2.9971961683640683e-06,
|
| 709 |
+
"loss": 0.3677,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.4893023255813953,
|
| 714 |
+
"grad_norm": 1.772230394205441,
|
| 715 |
+
"learning_rate": 2.9604186697658642e-06,
|
| 716 |
+
"loss": 0.3653,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.504186046511628,
|
| 721 |
+
"grad_norm": 1.517966913164345,
|
| 722 |
+
"learning_rate": 2.923584364610444e-06,
|
| 723 |
+
"loss": 0.3674,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.5190697674418605,
|
| 728 |
+
"grad_norm": 1.3320746245185962,
|
| 729 |
+
"learning_rate": 2.8867031970252262e-06,
|
| 730 |
+
"loss": 0.3654,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.5339534883720929,
|
| 735 |
+
"grad_norm": 1.4422364510340309,
|
| 736 |
+
"learning_rate": 2.84978512378904e-06,
|
| 737 |
+
"loss": 0.3649,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.5488372093023255,
|
| 742 |
+
"grad_norm": 1.422279942394409,
|
| 743 |
+
"learning_rate": 2.8128401116441058e-06,
|
| 744 |
+
"loss": 0.3637,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.563720930232558,
|
| 749 |
+
"grad_norm": 1.325658675407642,
|
| 750 |
+
"learning_rate": 2.7758781346053165e-06,
|
| 751 |
+
"loss": 0.365,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5786046511627907,
|
| 756 |
+
"grad_norm": 1.3384490232214399,
|
| 757 |
+
"learning_rate": 2.738909171267566e-06,
|
| 758 |
+
"loss": 0.3673,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5934883720930233,
|
| 763 |
+
"grad_norm": 1.4369227508461269,
|
| 764 |
+
"learning_rate": 2.7019432021118314e-06,
|
| 765 |
+
"loss": 0.3667,
|
| 766 |
+
"step": 1070
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.608372093023256,
|
| 770 |
+
"grad_norm": 1.550458179933586,
|
| 771 |
+
"learning_rate": 2.664990206810755e-06,
|
| 772 |
+
"loss": 0.3609,
|
| 773 |
+
"step": 1080
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.6232558139534885,
|
| 777 |
+
"grad_norm": 1.4665966921755653,
|
| 778 |
+
"learning_rate": 2.628060161534437e-06,
|
| 779 |
+
"loss": 0.3677,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.6381395348837209,
|
| 784 |
+
"grad_norm": 1.4478720169649395,
|
| 785 |
+
"learning_rate": 2.5911630362571787e-06,
|
| 786 |
+
"loss": 0.3663,
|
| 787 |
+
"step": 1100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.6530232558139535,
|
| 791 |
+
"grad_norm": 1.427086254069164,
|
| 792 |
+
"learning_rate": 2.5543087920658945e-06,
|
| 793 |
+
"loss": 0.3639,
|
| 794 |
+
"step": 1110
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.667906976744186,
|
| 798 |
+
"grad_norm": 1.354522173953887,
|
| 799 |
+
"learning_rate": 2.517507378470929e-06,
|
| 800 |
+
"loss": 0.3611,
|
| 801 |
+
"step": 1120
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.6827906976744185,
|
| 805 |
+
"grad_norm": 1.3693977150384091,
|
| 806 |
+
"learning_rate": 2.480768730719992e-06,
|
| 807 |
+
"loss": 0.3652,
|
| 808 |
+
"step": 1130
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.697674418604651,
|
| 812 |
+
"grad_norm": 1.2855988884013045,
|
| 813 |
+
"learning_rate": 2.4441027671159503e-06,
|
| 814 |
+
"loss": 0.3639,
|
| 815 |
+
"step": 1140
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.7125581395348837,
|
| 819 |
+
"grad_norm": 1.403080844114371,
|
| 820 |
+
"learning_rate": 2.4075193863391906e-06,
|
| 821 |
+
"loss": 0.3647,
|
| 822 |
+
"step": 1150
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.7274418604651163,
|
| 826 |
+
"grad_norm": 1.4103414505961822,
|
| 827 |
+
"learning_rate": 2.3710284647752805e-06,
|
| 828 |
+
"loss": 0.3656,
|
| 829 |
+
"step": 1160
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.7423255813953489,
|
| 833 |
+
"grad_norm": 1.328279125442394,
|
| 834 |
+
"learning_rate": 2.3346398538486488e-06,
|
| 835 |
+
"loss": 0.3601,
|
| 836 |
+
"step": 1170
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.7572093023255815,
|
| 840 |
+
"grad_norm": 1.4237569279155826,
|
| 841 |
+
"learning_rate": 2.2983633773630056e-06,
|
| 842 |
+
"loss": 0.3648,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.772093023255814,
|
| 847 |
+
"grad_norm": 1.3716090849409592,
|
| 848 |
+
"learning_rate": 2.2622088288492166e-06,
|
| 849 |
+
"loss": 0.3608,
|
| 850 |
+
"step": 1190
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.7869767441860465,
|
| 854 |
+
"grad_norm": 1.3654603526464721,
|
| 855 |
+
"learning_rate": 2.2261859689213523e-06,
|
| 856 |
+
"loss": 0.3597,
|
| 857 |
+
"step": 1200
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.801860465116279,
|
| 861 |
+
"grad_norm": 1.2691881731891166,
|
| 862 |
+
"learning_rate": 2.1903045226416216e-06,
|
| 863 |
+
"loss": 0.362,
|
| 864 |
+
"step": 1210
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.8167441860465117,
|
| 868 |
+
"grad_norm": 1.3157985277229747,
|
| 869 |
+
"learning_rate": 2.1545741768949085e-06,
|
| 870 |
+
"loss": 0.3611,
|
| 871 |
+
"step": 1220
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.831627906976744,
|
| 875 |
+
"grad_norm": 1.3583216702788425,
|
| 876 |
+
"learning_rate": 2.1190045777736057e-06,
|
| 877 |
+
"loss": 0.3613,
|
| 878 |
+
"step": 1230
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.8465116279069766,
|
| 882 |
+
"grad_norm": 1.2745346937557744,
|
| 883 |
+
"learning_rate": 2.0836053279734723e-06,
|
| 884 |
+
"loss": 0.3569,
|
| 885 |
+
"step": 1240
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.8613953488372093,
|
| 889 |
+
"grad_norm": 1.3036190956031755,
|
| 890 |
+
"learning_rate": 2.0483859842011976e-06,
|
| 891 |
+
"loss": 0.3597,
|
| 892 |
+
"step": 1250
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.8762790697674419,
|
| 896 |
+
"grad_norm": 1.4119443086776595,
|
| 897 |
+
"learning_rate": 2.0133560545943902e-06,
|
| 898 |
+
"loss": 0.3636,
|
| 899 |
+
"step": 1260
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.8911627906976745,
|
| 903 |
+
"grad_norm": 1.2897404474298366,
|
| 904 |
+
"learning_rate": 1.9785249961546668e-06,
|
| 905 |
+
"loss": 0.3575,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.906046511627907,
|
| 910 |
+
"grad_norm": 1.3639990026448434,
|
| 911 |
+
"learning_rate": 1.94390221219456e-06,
|
| 912 |
+
"loss": 0.3592,
|
| 913 |
+
"step": 1280
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.9209302325581397,
|
| 917 |
+
"grad_norm": 1.2546167591452697,
|
| 918 |
+
"learning_rate": 1.909497049798906e-06,
|
| 919 |
+
"loss": 0.3584,
|
| 920 |
+
"step": 1290
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.935813953488372,
|
| 924 |
+
"grad_norm": 1.3438380702724106,
|
| 925 |
+
"learning_rate": 1.8753187973014302e-06,
|
| 926 |
+
"loss": 0.3623,
|
| 927 |
+
"step": 1300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.9506976744186046,
|
| 931 |
+
"grad_norm": 1.3452056759551596,
|
| 932 |
+
"learning_rate": 1.8413766817771716e-06,
|
| 933 |
+
"loss": 0.3597,
|
| 934 |
+
"step": 1310
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.9655813953488372,
|
| 938 |
+
"grad_norm": 1.3834284182053571,
|
| 939 |
+
"learning_rate": 1.8076798665514672e-06,
|
| 940 |
+
"loss": 0.3586,
|
| 941 |
+
"step": 1320
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.9804651162790696,
|
| 945 |
+
"grad_norm": 1.275093148045728,
|
| 946 |
+
"learning_rate": 1.7742374487261275e-06,
|
| 947 |
+
"loss": 0.3556,
|
| 948 |
+
"step": 1330
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.9953488372093022,
|
| 952 |
+
"grad_norm": 1.3336419955281094,
|
| 953 |
+
"learning_rate": 1.7410584567235063e-06,
|
| 954 |
+
"loss": 0.3593,
|
| 955 |
+
"step": 1340
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.9983255813953489,
|
| 959 |
+
"eval_loss": 0.05562544986605644,
|
| 960 |
+
"eval_runtime": 454.3217,
|
| 961 |
+
"eval_samples_per_second": 39.846,
|
| 962 |
+
"eval_steps_per_second": 0.623,
|
| 963 |
+
"step": 1342
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.0111627906976746,
|
| 967 |
+
"grad_norm": 1.8801844172458124,
|
| 968 |
+
"learning_rate": 1.7081518478491024e-06,
|
| 969 |
+
"loss": 0.2974,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 2.026046511627907,
|
| 974 |
+
"grad_norm": 1.5466963455442697,
|
| 975 |
+
"learning_rate": 1.6755265058733625e-06,
|
| 976 |
+
"loss": 0.2705,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0409302325581393,
|
| 981 |
+
"grad_norm": 1.6491531999787248,
|
| 982 |
+
"learning_rate": 1.6431912386333337e-06,
|
| 983 |
+
"loss": 0.2715,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.055813953488372,
|
| 988 |
+
"grad_norm": 1.3549718677985036,
|
| 989 |
+
"learning_rate": 1.61115477565483e-06,
|
| 990 |
+
"loss": 0.2711,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.0706976744186045,
|
| 995 |
+
"grad_norm": 1.4210689732513506,
|
| 996 |
+
"learning_rate": 1.5794257657957149e-06,
|
| 997 |
+
"loss": 0.2678,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.085581395348837,
|
| 1002 |
+
"grad_norm": 1.5199527949561047,
|
| 1003 |
+
"learning_rate": 1.5480127749109867e-06,
|
| 1004 |
+
"loss": 0.2715,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.1004651162790697,
|
| 1009 |
+
"grad_norm": 1.3921480720538892,
|
| 1010 |
+
"learning_rate": 1.516924283540257e-06,
|
| 1011 |
+
"loss": 0.268,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 2.1153488372093023,
|
| 1016 |
+
"grad_norm": 1.5217951800344998,
|
| 1017 |
+
"learning_rate": 1.486168684618268e-06,
|
| 1018 |
+
"loss": 0.2666,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.130232558139535,
|
| 1023 |
+
"grad_norm": 1.4410321423758377,
|
| 1024 |
+
"learning_rate": 1.4557542812090574e-06,
|
| 1025 |
+
"loss": 0.2698,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 2.1451162790697675,
|
| 1030 |
+
"grad_norm": 1.4576711215393965,
|
| 1031 |
+
"learning_rate": 1.4256892842643893e-06,
|
| 1032 |
+
"loss": 0.2675,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.16,
|
| 1037 |
+
"grad_norm": 1.40927027905477,
|
| 1038 |
+
"learning_rate": 1.3959818104070452e-06,
|
| 1039 |
+
"loss": 0.2687,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.1748837209302327,
|
| 1044 |
+
"grad_norm": 1.4616727679236827,
|
| 1045 |
+
"learning_rate": 1.3666398797395948e-06,
|
| 1046 |
+
"loss": 0.2707,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 2.1897674418604653,
|
| 1051 |
+
"grad_norm": 1.3795657326148565,
|
| 1052 |
+
"learning_rate": 1.3376714136792034e-06,
|
| 1053 |
+
"loss": 0.2665,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 2.2046511627906975,
|
| 1058 |
+
"grad_norm": 1.5330397219497653,
|
| 1059 |
+
"learning_rate": 1.3090842328191053e-06,
|
| 1060 |
+
"loss": 0.2699,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 2.21953488372093,
|
| 1065 |
+
"grad_norm": 1.4083642956808657,
|
| 1066 |
+
"learning_rate": 1.280886054817277e-06,
|
| 1067 |
+
"loss": 0.2709,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 2.2344186046511627,
|
| 1072 |
+
"grad_norm": 1.4908783738210238,
|
| 1073 |
+
"learning_rate": 1.2530844923129096e-06,
|
| 1074 |
+
"loss": 0.2712,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 2.2493023255813953,
|
| 1079 |
+
"grad_norm": 1.3954198116255683,
|
| 1080 |
+
"learning_rate": 1.225687050871231e-06,
|
| 1081 |
+
"loss": 0.2705,
|
| 1082 |
+
"step": 1510
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.264186046511628,
|
| 1086 |
+
"grad_norm": 1.38515024685356,
|
| 1087 |
+
"learning_rate": 1.1987011269572357e-06,
|
| 1088 |
+
"loss": 0.2701,
|
| 1089 |
+
"step": 1520
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2790697674418605,
|
| 1093 |
+
"grad_norm": 1.3876830410349839,
|
| 1094 |
+
"learning_rate": 1.1721340059388617e-06,
|
| 1095 |
+
"loss": 0.2672,
|
| 1096 |
+
"step": 1530
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.293953488372093,
|
| 1100 |
+
"grad_norm": 1.5349649821163696,
|
| 1101 |
+
"learning_rate": 1.1459928601201756e-06,
|
| 1102 |
+
"loss": 0.2696,
|
| 1103 |
+
"step": 1540
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.3088372093023257,
|
| 1107 |
+
"grad_norm": 1.6222404726063406,
|
| 1108 |
+
"learning_rate": 1.1202847468050597e-06,
|
| 1109 |
+
"loss": 0.2689,
|
| 1110 |
+
"step": 1550
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 2.3237209302325583,
|
| 1114 |
+
"grad_norm": 1.4018802032895803,
|
| 1115 |
+
"learning_rate": 1.0950166063919694e-06,
|
| 1116 |
+
"loss": 0.2701,
|
| 1117 |
+
"step": 1560
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 2.3386046511627905,
|
| 1121 |
+
"grad_norm": 1.3526044361279066,
|
| 1122 |
+
"learning_rate": 1.0701952605002275e-06,
|
| 1123 |
+
"loss": 0.2676,
|
| 1124 |
+
"step": 1570
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.353488372093023,
|
| 1128 |
+
"grad_norm": 1.4442038534402912,
|
| 1129 |
+
"learning_rate": 1.045827410128407e-06,
|
| 1130 |
+
"loss": 0.2661,
|
| 1131 |
+
"step": 1580
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.3683720930232557,
|
| 1135 |
+
"grad_norm": 1.4163720971352327,
|
| 1136 |
+
"learning_rate": 1.0219196338452623e-06,
|
| 1137 |
+
"loss": 0.2689,
|
| 1138 |
+
"step": 1590
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.3832558139534883,
|
| 1142 |
+
"grad_norm": 1.356177063119025,
|
| 1143 |
+
"learning_rate": 9.984783860137213e-07,
|
| 1144 |
+
"loss": 0.2676,
|
| 1145 |
+
"step": 1600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.398139534883721,
|
| 1149 |
+
"grad_norm": 1.3502869776338615,
|
| 1150 |
+
"learning_rate": 9.75509995048404e-07,
|
| 1151 |
+
"loss": 0.2681,
|
| 1152 |
+
"step": 1610
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.4130232558139535,
|
| 1156 |
+
"grad_norm": 1.337863106664254,
|
| 1157 |
+
"learning_rate": 9.53020661707148e-07,
|
| 1158 |
+
"loss": 0.2695,
|
| 1159 |
+
"step": 1620
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 2.427906976744186,
|
| 1163 |
+
"grad_norm": 1.3624966178122666,
|
| 1164 |
+
"learning_rate": 9.310164574169911e-07,
|
| 1165 |
+
"loss": 0.2661,
|
| 1166 |
+
"step": 1630
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.4427906976744187,
|
| 1170 |
+
"grad_norm": 1.4427762559619615,
|
| 1171 |
+
"learning_rate": 9.095033226350787e-07,
|
| 1172 |
+
"loss": 0.2682,
|
| 1173 |
+
"step": 1640
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 2.4576744186046513,
|
| 1177 |
+
"grad_norm": 1.3824477379757505,
|
| 1178 |
+
"learning_rate": 8.884870652449176e-07,
|
| 1179 |
+
"loss": 0.2683,
|
| 1180 |
+
"step": 1650
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.472558139534884,
|
| 1184 |
+
"grad_norm": 1.3968898578295752,
|
| 1185 |
+
"learning_rate": 8.679733589884308e-07,
|
| 1186 |
+
"loss": 0.2676,
|
| 1187 |
+
"step": 1660
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.4874418604651165,
|
| 1191 |
+
"grad_norm": 1.403396644885307,
|
| 1192 |
+
"learning_rate": 8.479677419342195e-07,
|
| 1193 |
+
"loss": 0.2675,
|
| 1194 |
+
"step": 1670
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.5023255813953487,
|
| 1198 |
+
"grad_norm": 1.4635100367305935,
|
| 1199 |
+
"learning_rate": 8.284756149824561e-07,
|
| 1200 |
+
"loss": 0.2691,
|
| 1201 |
+
"step": 1680
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.5172093023255813,
|
| 1205 |
+
"grad_norm": 1.4243622651693753,
|
| 1206 |
+
"learning_rate": 8.095022404068078e-07,
|
| 1207 |
+
"loss": 0.269,
|
| 1208 |
+
"step": 1690
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.532093023255814,
|
| 1212 |
+
"grad_norm": 1.379314761648038,
|
| 1213 |
+
"learning_rate": 7.910527404337846e-07,
|
| 1214 |
+
"loss": 0.2687,
|
| 1215 |
+
"step": 1700
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 2.5469767441860465,
|
| 1219 |
+
"grad_norm": 1.3800285755689508,
|
| 1220 |
+
"learning_rate": 7.731320958598944e-07,
|
| 1221 |
+
"loss": 0.2687,
|
| 1222 |
+
"step": 1710
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.561860465116279,
|
| 1226 |
+
"grad_norm": 1.4168008157356147,
|
| 1227 |
+
"learning_rate": 7.557451447069862e-07,
|
| 1228 |
+
"loss": 0.2686,
|
| 1229 |
+
"step": 1720
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.5767441860465117,
|
| 1233 |
+
"grad_norm": 1.405042195482243,
|
| 1234 |
+
"learning_rate": 7.388965809161264e-07,
|
| 1235 |
+
"loss": 0.273,
|
| 1236 |
+
"step": 1730
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 2.5916279069767443,
|
| 1240 |
+
"grad_norm": 1.339874655650546,
|
| 1241 |
+
"learning_rate": 7.225909530803849e-07,
|
| 1242 |
+
"loss": 0.2702,
|
| 1243 |
+
"step": 1740
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 2.606511627906977,
|
| 1247 |
+
"grad_norm": 1.376936805811691,
|
| 1248 |
+
"learning_rate": 7.068326632168529e-07,
|
| 1249 |
+
"loss": 0.2682,
|
| 1250 |
+
"step": 1750
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.6213953488372095,
|
| 1254 |
+
"grad_norm": 1.370179020476883,
|
| 1255 |
+
"learning_rate": 6.91625965578234e-07,
|
| 1256 |
+
"loss": 0.268,
|
| 1257 |
+
"step": 1760
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 2.6362790697674416,
|
| 1261 |
+
"grad_norm": 1.3701048711478194,
|
| 1262 |
+
"learning_rate": 6.769749655043278e-07,
|
| 1263 |
+
"loss": 0.2678,
|
| 1264 |
+
"step": 1770
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.6511627906976747,
|
| 1268 |
+
"grad_norm": 1.4206830049678218,
|
| 1269 |
+
"learning_rate": 6.628836183137136e-07,
|
| 1270 |
+
"loss": 0.2701,
|
| 1271 |
+
"step": 1780
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.666046511627907,
|
| 1275 |
+
"grad_norm": 1.393374140302477,
|
| 1276 |
+
"learning_rate": 6.493557282359362e-07,
|
| 1277 |
+
"loss": 0.2687,
|
| 1278 |
+
"step": 1790
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 2.6809302325581394,
|
| 1282 |
+
"grad_norm": 1.461263759783157,
|
| 1283 |
+
"learning_rate": 6.363949473844831e-07,
|
| 1284 |
+
"loss": 0.268,
|
| 1285 |
+
"step": 1800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 2.695813953488372,
|
| 1289 |
+
"grad_norm": 1.4734006960029564,
|
| 1290 |
+
"learning_rate": 6.240047747708234e-07,
|
| 1291 |
+
"loss": 0.2677,
|
| 1292 |
+
"step": 1810
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.7106976744186047,
|
| 1296 |
+
"grad_norm": 1.3366661643587023,
|
| 1297 |
+
"learning_rate": 6.121885553597864e-07,
|
| 1298 |
+
"loss": 0.2681,
|
| 1299 |
+
"step": 1820
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.7255813953488373,
|
| 1303 |
+
"grad_norm": 1.422096088237659,
|
| 1304 |
+
"learning_rate": 6.009494791665193e-07,
|
| 1305 |
+
"loss": 0.2696,
|
| 1306 |
+
"step": 1830
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.74046511627907,
|
| 1310 |
+
"grad_norm": 1.34042220211478,
|
| 1311 |
+
"learning_rate": 5.902905803952853e-07,
|
| 1312 |
+
"loss": 0.2707,
|
| 1313 |
+
"step": 1840
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.7553488372093025,
|
| 1317 |
+
"grad_norm": 1.43033974764486,
|
| 1318 |
+
"learning_rate": 5.802147366203209e-07,
|
| 1319 |
+
"loss": 0.2682,
|
| 1320 |
+
"step": 1850
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.7702325581395346,
|
| 1324 |
+
"grad_norm": 1.3377632425473605,
|
| 1325 |
+
"learning_rate": 5.707246680089786e-07,
|
| 1326 |
+
"loss": 0.2682,
|
| 1327 |
+
"step": 1860
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.7851162790697677,
|
| 1331 |
+
"grad_norm": 1.3552116715763636,
|
| 1332 |
+
"learning_rate": 5.618229365873664e-07,
|
| 1333 |
+
"loss": 0.2679,
|
| 1334 |
+
"step": 1870
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.8,
|
| 1338 |
+
"grad_norm": 1.407612240704182,
|
| 1339 |
+
"learning_rate": 5.535119455486798e-07,
|
| 1340 |
+
"loss": 0.2671,
|
| 1341 |
+
"step": 1880
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 2.8148837209302324,
|
| 1345 |
+
"grad_norm": 1.3307680080669904,
|
| 1346 |
+
"learning_rate": 5.457939386044124e-07,
|
| 1347 |
+
"loss": 0.2691,
|
| 1348 |
+
"step": 1890
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.829767441860465,
|
| 1352 |
+
"grad_norm": 1.33966014097135,
|
| 1353 |
+
"learning_rate": 5.386709993786254e-07,
|
| 1354 |
+
"loss": 0.2691,
|
| 1355 |
+
"step": 1900
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 2.8446511627906976,
|
| 1359 |
+
"grad_norm": 1.3388112782785693,
|
| 1360 |
+
"learning_rate": 5.321450508454304e-07,
|
| 1361 |
+
"loss": 0.2678,
|
| 1362 |
+
"step": 1910
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 2.8595348837209302,
|
| 1366 |
+
"grad_norm": 1.3186034865675909,
|
| 1367 |
+
"learning_rate": 5.262178548098479e-07,
|
| 1368 |
+
"loss": 0.2668,
|
| 1369 |
+
"step": 1920
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 2.874418604651163,
|
| 1373 |
+
"grad_norm": 1.3351972694593512,
|
| 1374 |
+
"learning_rate": 5.208910114321729e-07,
|
| 1375 |
+
"loss": 0.2662,
|
| 1376 |
+
"step": 1930
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.8893023255813954,
|
| 1380 |
+
"grad_norm": 1.373685192971363,
|
| 1381 |
+
"learning_rate": 5.161659587959818e-07,
|
| 1382 |
+
"loss": 0.2669,
|
| 1383 |
+
"step": 1940
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 2.904186046511628,
|
| 1387 |
+
"grad_norm": 1.3931156997045835,
|
| 1388 |
+
"learning_rate": 5.120439725198932e-07,
|
| 1389 |
+
"loss": 0.2663,
|
| 1390 |
+
"step": 1950
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 2.9190697674418606,
|
| 1394 |
+
"grad_norm": 1.2915260316130637,
|
| 1395 |
+
"learning_rate": 5.085261654131918e-07,
|
| 1396 |
+
"loss": 0.2632,
|
| 1397 |
+
"step": 1960
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.933953488372093,
|
| 1401 |
+
"grad_norm": 1.3421708919414022,
|
| 1402 |
+
"learning_rate": 5.056134871754014e-07,
|
| 1403 |
+
"loss": 0.2705,
|
| 1404 |
+
"step": 1970
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.948837209302326,
|
| 1408 |
+
"grad_norm": 1.3219790823176543,
|
| 1409 |
+
"learning_rate": 5.03306724139899e-07,
|
| 1410 |
+
"loss": 0.2675,
|
| 1411 |
+
"step": 1980
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.963720930232558,
|
| 1415 |
+
"grad_norm": 1.3110383511595092,
|
| 1416 |
+
"learning_rate": 5.016064990616251e-07,
|
| 1417 |
+
"loss": 0.2652,
|
| 1418 |
+
"step": 1990
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.9786046511627906,
|
| 1422 |
+
"grad_norm": 1.312152439222843,
|
| 1423 |
+
"learning_rate": 5.005132709489625e-07,
|
| 1424 |
+
"loss": 0.2666,
|
| 1425 |
+
"step": 2000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 2.993488372093023,
|
| 1429 |
+
"grad_norm": 1.385415982308588,
|
| 1430 |
+
"learning_rate": 5.000273349398159e-07,
|
| 1431 |
+
"loss": 0.2664,
|
| 1432 |
+
"step": 2010
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 2.997953488372093,
|
| 1436 |
+
"eval_loss": 0.05878664180636406,
|
| 1437 |
+
"eval_runtime": 453.0631,
|
| 1438 |
+
"eval_samples_per_second": 39.957,
|
| 1439 |
+
"eval_steps_per_second": 0.625,
|
| 1440 |
+
"step": 2013
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 2.997953488372093,
|
| 1444 |
+
"step": 2013,
|
| 1445 |
+
"total_flos": 3371640595415040.0,
|
| 1446 |
+
"train_loss": 0.382664002355982,
|
| 1447 |
+
"train_runtime": 66379.5775,
|
| 1448 |
+
"train_samples_per_second": 15.545,
|
| 1449 |
+
"train_steps_per_second": 0.03
|
| 1450 |
+
}
|
| 1451 |
+
],
|
| 1452 |
+
"logging_steps": 10,
|
| 1453 |
+
"max_steps": 2013,
|
| 1454 |
+
"num_input_tokens_seen": 0,
|
| 1455 |
+
"num_train_epochs": 3,
|
| 1456 |
+
"save_steps": 500,
|
| 1457 |
+
"stateful_callbacks": {
|
| 1458 |
+
"TrainerControl": {
|
| 1459 |
+
"args": {
|
| 1460 |
+
"should_epoch_stop": false,
|
| 1461 |
+
"should_evaluate": false,
|
| 1462 |
+
"should_log": false,
|
| 1463 |
+
"should_save": true,
|
| 1464 |
+
"should_training_stop": true
|
| 1465 |
+
},
|
| 1466 |
+
"attributes": {}
|
| 1467 |
+
}
|
| 1468 |
+
},
|
| 1469 |
+
"total_flos": 3371640595415040.0,
|
| 1470 |
+
"train_batch_size": 8,
|
| 1471 |
+
"trial_name": null,
|
| 1472 |
+
"trial_params": null
|
| 1473 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|