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 +2138 -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_mistral_bsz256
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# hp_ablations_mistral_bsz256
|
| 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: 0.1512
|
| 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_mistral_bsz256
|
|
|
|
| 16 |
|
| 17 |
# hp_ablations_mistral_bsz256
|
| 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-v3.1-gpt-4o-mini dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.1512
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9984810126582278,
|
| 3 |
+
"eval_loss": 0.15120381116867065,
|
| 4 |
+
"eval_runtime": 511.1809,
|
| 5 |
+
"eval_samples_per_second": 26.026,
|
| 6 |
+
"eval_steps_per_second": 0.407,
|
| 7 |
+
"total_flos": 2479683262218240.0,
|
| 8 |
+
"train_loss": 0.46617310421968466,
|
| 9 |
+
"train_runtime": 84499.4991,
|
| 10 |
+
"train_samples_per_second": 8.974,
|
| 11 |
+
"train_steps_per_second": 0.035
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9984810126582278,
|
| 3 |
+
"eval_loss": 0.15120381116867065,
|
| 4 |
+
"eval_runtime": 511.1809,
|
| 5 |
+
"eval_samples_per_second": 26.026,
|
| 6 |
+
"eval_steps_per_second": 0.407
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9984810126582278,
|
| 3 |
+
"total_flos": 2479683262218240.0,
|
| 4 |
+
"train_loss": 0.46617310421968466,
|
| 5 |
+
"train_runtime": 84499.4991,
|
| 6 |
+
"train_samples_per_second": 8.974,
|
| 7 |
+
"train_steps_per_second": 0.035
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2138 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9984810126582278,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2961,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010126582278481013,
|
| 13 |
+
"grad_norm": 2.9319232619463147,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.754,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.020253164556962026,
|
| 20 |
+
"grad_norm": 2.3527842138237443,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.6565,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.030379746835443037,
|
| 27 |
+
"grad_norm": 2.218097297827581,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6343,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04050632911392405,
|
| 34 |
+
"grad_norm": 1.728885416020883,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6266,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05063291139240506,
|
| 41 |
+
"grad_norm": 1.7600150721761239,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6225,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.060759493670886074,
|
| 48 |
+
"grad_norm": 2.036486573113631,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6185,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07088607594936709,
|
| 55 |
+
"grad_norm": 1.8472990337897077,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6137,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0810126582278481,
|
| 62 |
+
"grad_norm": 2.0268825324116926,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6035,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09113924050632911,
|
| 69 |
+
"grad_norm": 1.7502041363801772,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6063,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10126582278481013,
|
| 76 |
+
"grad_norm": 1.7706135514845762,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.5954,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11139240506329114,
|
| 83 |
+
"grad_norm": 1.6647485123629138,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.5933,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12151898734177215,
|
| 90 |
+
"grad_norm": 2.299944971561371,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.598,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.13164556962025317,
|
| 97 |
+
"grad_norm": 1.4323042297049071,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.597,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14177215189873418,
|
| 104 |
+
"grad_norm": 1.5735470897186044,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.5859,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1518987341772152,
|
| 111 |
+
"grad_norm": 1.4396445107803788,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.5833,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.1620253164556962,
|
| 118 |
+
"grad_norm": 1.6048875360767698,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.5926,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17215189873417722,
|
| 125 |
+
"grad_norm": 1.7629611019629856,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.5878,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18227848101265823,
|
| 132 |
+
"grad_norm": 1.692513854306383,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.5875,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.19240506329113924,
|
| 139 |
+
"grad_norm": 2.4645583630800614,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.5908,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.20253164556962025,
|
| 146 |
+
"grad_norm": 1.6304964629686653,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.5887,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21265822784810126,
|
| 153 |
+
"grad_norm": 2.117270092101592,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.5872,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22278481012658227,
|
| 160 |
+
"grad_norm": 1.5677168838868496,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.575,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.23291139240506328,
|
| 167 |
+
"grad_norm": 1.6515647911636342,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.5747,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2430379746835443,
|
| 174 |
+
"grad_norm": 1.4834458457301771,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.5834,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25316455696202533,
|
| 181 |
+
"grad_norm": 2.090691823913493,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.5793,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.26329113924050634,
|
| 188 |
+
"grad_norm": 1.4520944372541345,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.5779,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.27341772151898736,
|
| 195 |
+
"grad_norm": 1.7905780272726646,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.5637,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.28354430379746837,
|
| 202 |
+
"grad_norm": 1.6753642665165924,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.5745,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2936708860759494,
|
| 209 |
+
"grad_norm": 1.8117100564747939,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.577,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3037974683544304,
|
| 216 |
+
"grad_norm": 1.5160896257954382,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.5821,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3139240506329114,
|
| 223 |
+
"grad_norm": 1.5338602444115943,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.574,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3240506329113924,
|
| 230 |
+
"grad_norm": 1.5768816348730115,
|
| 231 |
+
"learning_rate": 5e-06,
|
| 232 |
+
"loss": 0.5831,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3341772151898734,
|
| 237 |
+
"grad_norm": 1.5251244093370149,
|
| 238 |
+
"learning_rate": 5e-06,
|
| 239 |
+
"loss": 0.5793,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.34430379746835443,
|
| 244 |
+
"grad_norm": 1.5422210280401378,
|
| 245 |
+
"learning_rate": 5e-06,
|
| 246 |
+
"loss": 0.579,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35443037974683544,
|
| 251 |
+
"grad_norm": 1.4422891794630028,
|
| 252 |
+
"learning_rate": 5e-06,
|
| 253 |
+
"loss": 0.5794,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.36455696202531646,
|
| 258 |
+
"grad_norm": 1.778098375508798,
|
| 259 |
+
"learning_rate": 5e-06,
|
| 260 |
+
"loss": 0.5702,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.37468354430379747,
|
| 265 |
+
"grad_norm": 1.4276847831429058,
|
| 266 |
+
"learning_rate": 5e-06,
|
| 267 |
+
"loss": 0.5762,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.3848101265822785,
|
| 272 |
+
"grad_norm": 1.8350692685441172,
|
| 273 |
+
"learning_rate": 5e-06,
|
| 274 |
+
"loss": 0.5742,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3949367088607595,
|
| 279 |
+
"grad_norm": 1.6654466906631886,
|
| 280 |
+
"learning_rate": 5e-06,
|
| 281 |
+
"loss": 0.57,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.4050632911392405,
|
| 286 |
+
"grad_norm": 1.5625914451532021,
|
| 287 |
+
"learning_rate": 5e-06,
|
| 288 |
+
"loss": 0.5688,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.4151898734177215,
|
| 293 |
+
"grad_norm": 1.5962831101550994,
|
| 294 |
+
"learning_rate": 5e-06,
|
| 295 |
+
"loss": 0.5567,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.4253164556962025,
|
| 300 |
+
"grad_norm": 1.4805817323596187,
|
| 301 |
+
"learning_rate": 5e-06,
|
| 302 |
+
"loss": 0.5723,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.43544303797468353,
|
| 307 |
+
"grad_norm": 1.4164796466458769,
|
| 308 |
+
"learning_rate": 5e-06,
|
| 309 |
+
"loss": 0.5712,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.44556962025316454,
|
| 314 |
+
"grad_norm": 1.5450127362729664,
|
| 315 |
+
"learning_rate": 5e-06,
|
| 316 |
+
"loss": 0.5704,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.45569620253164556,
|
| 321 |
+
"grad_norm": 1.6570421627781717,
|
| 322 |
+
"learning_rate": 5e-06,
|
| 323 |
+
"loss": 0.5729,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.46582278481012657,
|
| 328 |
+
"grad_norm": 1.623306340780734,
|
| 329 |
+
"learning_rate": 5e-06,
|
| 330 |
+
"loss": 0.5673,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4759493670886076,
|
| 335 |
+
"grad_norm": 1.747763610168544,
|
| 336 |
+
"learning_rate": 5e-06,
|
| 337 |
+
"loss": 0.5609,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4860759493670886,
|
| 342 |
+
"grad_norm": 1.295527401866004,
|
| 343 |
+
"learning_rate": 5e-06,
|
| 344 |
+
"loss": 0.5699,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4962025316455696,
|
| 349 |
+
"grad_norm": 1.3529876566902441,
|
| 350 |
+
"learning_rate": 5e-06,
|
| 351 |
+
"loss": 0.5783,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5063291139240507,
|
| 356 |
+
"grad_norm": 1.4541590535564763,
|
| 357 |
+
"learning_rate": 5e-06,
|
| 358 |
+
"loss": 0.5606,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5164556962025316,
|
| 363 |
+
"grad_norm": 1.5391153152763637,
|
| 364 |
+
"learning_rate": 5e-06,
|
| 365 |
+
"loss": 0.5598,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5265822784810127,
|
| 370 |
+
"grad_norm": 1.3816263219270335,
|
| 371 |
+
"learning_rate": 5e-06,
|
| 372 |
+
"loss": 0.552,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5367088607594936,
|
| 377 |
+
"grad_norm": 1.407148137331511,
|
| 378 |
+
"learning_rate": 5e-06,
|
| 379 |
+
"loss": 0.561,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5468354430379747,
|
| 384 |
+
"grad_norm": 1.4939662242734717,
|
| 385 |
+
"learning_rate": 5e-06,
|
| 386 |
+
"loss": 0.5691,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5569620253164557,
|
| 391 |
+
"grad_norm": 1.4769024813528058,
|
| 392 |
+
"learning_rate": 5e-06,
|
| 393 |
+
"loss": 0.5628,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5670886075949367,
|
| 398 |
+
"grad_norm": 1.5234796484078978,
|
| 399 |
+
"learning_rate": 5e-06,
|
| 400 |
+
"loss": 0.5571,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5772151898734177,
|
| 405 |
+
"grad_norm": 1.38748512162588,
|
| 406 |
+
"learning_rate": 5e-06,
|
| 407 |
+
"loss": 0.5588,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5873417721518988,
|
| 412 |
+
"grad_norm": 1.6616670500045834,
|
| 413 |
+
"learning_rate": 5e-06,
|
| 414 |
+
"loss": 0.5629,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5974683544303797,
|
| 419 |
+
"grad_norm": 1.3613770956091802,
|
| 420 |
+
"learning_rate": 5e-06,
|
| 421 |
+
"loss": 0.5597,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6075949367088608,
|
| 426 |
+
"grad_norm": 1.9645875965732293,
|
| 427 |
+
"learning_rate": 5e-06,
|
| 428 |
+
"loss": 0.5614,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6177215189873417,
|
| 433 |
+
"grad_norm": 1.3122962556953037,
|
| 434 |
+
"learning_rate": 5e-06,
|
| 435 |
+
"loss": 0.57,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6278481012658228,
|
| 440 |
+
"grad_norm": 1.5024516266743335,
|
| 441 |
+
"learning_rate": 5e-06,
|
| 442 |
+
"loss": 0.5629,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6379746835443038,
|
| 447 |
+
"grad_norm": 1.3599801460050238,
|
| 448 |
+
"learning_rate": 5e-06,
|
| 449 |
+
"loss": 0.5679,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6481012658227848,
|
| 454 |
+
"grad_norm": 1.3697118174446543,
|
| 455 |
+
"learning_rate": 5e-06,
|
| 456 |
+
"loss": 0.5567,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6582278481012658,
|
| 461 |
+
"grad_norm": 1.447467777446668,
|
| 462 |
+
"learning_rate": 5e-06,
|
| 463 |
+
"loss": 0.5614,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6683544303797468,
|
| 468 |
+
"grad_norm": 1.4332946464750023,
|
| 469 |
+
"learning_rate": 5e-06,
|
| 470 |
+
"loss": 0.5629,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6784810126582278,
|
| 475 |
+
"grad_norm": 1.4245518863907969,
|
| 476 |
+
"learning_rate": 5e-06,
|
| 477 |
+
"loss": 0.5607,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6886075949367089,
|
| 482 |
+
"grad_norm": 1.392960913673911,
|
| 483 |
+
"learning_rate": 5e-06,
|
| 484 |
+
"loss": 0.5631,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6987341772151898,
|
| 489 |
+
"grad_norm": 1.3851040024314174,
|
| 490 |
+
"learning_rate": 5e-06,
|
| 491 |
+
"loss": 0.5547,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7088607594936709,
|
| 496 |
+
"grad_norm": 1.5665731498555628,
|
| 497 |
+
"learning_rate": 5e-06,
|
| 498 |
+
"loss": 0.5729,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7189873417721518,
|
| 503 |
+
"grad_norm": 1.3541133459928674,
|
| 504 |
+
"learning_rate": 5e-06,
|
| 505 |
+
"loss": 0.567,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7291139240506329,
|
| 510 |
+
"grad_norm": 1.3050422174455243,
|
| 511 |
+
"learning_rate": 5e-06,
|
| 512 |
+
"loss": 0.5649,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.739240506329114,
|
| 517 |
+
"grad_norm": 1.4474203168962876,
|
| 518 |
+
"learning_rate": 5e-06,
|
| 519 |
+
"loss": 0.5587,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7493670886075949,
|
| 524 |
+
"grad_norm": 1.2849691486389958,
|
| 525 |
+
"learning_rate": 5e-06,
|
| 526 |
+
"loss": 0.5584,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.759493670886076,
|
| 531 |
+
"grad_norm": 1.4374182399089128,
|
| 532 |
+
"learning_rate": 5e-06,
|
| 533 |
+
"loss": 0.5485,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.769620253164557,
|
| 538 |
+
"grad_norm": 1.2784166011759992,
|
| 539 |
+
"learning_rate": 5e-06,
|
| 540 |
+
"loss": 0.5619,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.779746835443038,
|
| 545 |
+
"grad_norm": 1.338741085455322,
|
| 546 |
+
"learning_rate": 5e-06,
|
| 547 |
+
"loss": 0.5625,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.789873417721519,
|
| 552 |
+
"grad_norm": 1.3830345955439087,
|
| 553 |
+
"learning_rate": 5e-06,
|
| 554 |
+
"loss": 0.5532,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8,
|
| 559 |
+
"grad_norm": 1.2560229071250701,
|
| 560 |
+
"learning_rate": 5e-06,
|
| 561 |
+
"loss": 0.557,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.810126582278481,
|
| 566 |
+
"grad_norm": 1.3094948207554686,
|
| 567 |
+
"learning_rate": 5e-06,
|
| 568 |
+
"loss": 0.5606,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8202531645569621,
|
| 573 |
+
"grad_norm": 1.2980406537156193,
|
| 574 |
+
"learning_rate": 5e-06,
|
| 575 |
+
"loss": 0.5435,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.830379746835443,
|
| 580 |
+
"grad_norm": 1.3336678028790407,
|
| 581 |
+
"learning_rate": 5e-06,
|
| 582 |
+
"loss": 0.5644,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.8405063291139241,
|
| 587 |
+
"grad_norm": 1.4362113888678505,
|
| 588 |
+
"learning_rate": 5e-06,
|
| 589 |
+
"loss": 0.5615,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.850632911392405,
|
| 594 |
+
"grad_norm": 1.5496595209680522,
|
| 595 |
+
"learning_rate": 5e-06,
|
| 596 |
+
"loss": 0.5619,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8607594936708861,
|
| 601 |
+
"grad_norm": 1.4648845925364151,
|
| 602 |
+
"learning_rate": 5e-06,
|
| 603 |
+
"loss": 0.5556,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8708860759493671,
|
| 608 |
+
"grad_norm": 1.3830147030582034,
|
| 609 |
+
"learning_rate": 5e-06,
|
| 610 |
+
"loss": 0.5534,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8810126582278481,
|
| 615 |
+
"grad_norm": 1.3802375047462647,
|
| 616 |
+
"learning_rate": 5e-06,
|
| 617 |
+
"loss": 0.5654,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.8911392405063291,
|
| 622 |
+
"grad_norm": 1.665356350061181,
|
| 623 |
+
"learning_rate": 5e-06,
|
| 624 |
+
"loss": 0.5542,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9012658227848102,
|
| 629 |
+
"grad_norm": 1.5275969827966436,
|
| 630 |
+
"learning_rate": 5e-06,
|
| 631 |
+
"loss": 0.5492,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9113924050632911,
|
| 636 |
+
"grad_norm": 1.5257937868842386,
|
| 637 |
+
"learning_rate": 5e-06,
|
| 638 |
+
"loss": 0.5509,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9215189873417722,
|
| 643 |
+
"grad_norm": 1.3059469433496236,
|
| 644 |
+
"learning_rate": 5e-06,
|
| 645 |
+
"loss": 0.5563,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9316455696202531,
|
| 650 |
+
"grad_norm": 1.3687394115237168,
|
| 651 |
+
"learning_rate": 5e-06,
|
| 652 |
+
"loss": 0.5527,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9417721518987342,
|
| 657 |
+
"grad_norm": 1.3468586235595001,
|
| 658 |
+
"learning_rate": 5e-06,
|
| 659 |
+
"loss": 0.5594,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9518987341772152,
|
| 664 |
+
"grad_norm": 1.4377274430310882,
|
| 665 |
+
"learning_rate": 5e-06,
|
| 666 |
+
"loss": 0.5512,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9620253164556962,
|
| 671 |
+
"grad_norm": 1.2680685522134791,
|
| 672 |
+
"learning_rate": 5e-06,
|
| 673 |
+
"loss": 0.5486,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9721518987341772,
|
| 678 |
+
"grad_norm": 1.2807954706357672,
|
| 679 |
+
"learning_rate": 5e-06,
|
| 680 |
+
"loss": 0.5566,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9822784810126582,
|
| 685 |
+
"grad_norm": 1.3423829666439555,
|
| 686 |
+
"learning_rate": 5e-06,
|
| 687 |
+
"loss": 0.5485,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.9924050632911392,
|
| 692 |
+
"grad_norm": 1.4283880670670543,
|
| 693 |
+
"learning_rate": 5e-06,
|
| 694 |
+
"loss": 0.5515,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.999493670886076,
|
| 699 |
+
"eval_loss": 0.13834980130195618,
|
| 700 |
+
"eval_runtime": 507.9163,
|
| 701 |
+
"eval_samples_per_second": 26.193,
|
| 702 |
+
"eval_steps_per_second": 0.41,
|
| 703 |
+
"step": 987
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.0025316455696203,
|
| 707 |
+
"grad_norm": 1.6161728864680243,
|
| 708 |
+
"learning_rate": 5e-06,
|
| 709 |
+
"loss": 0.5308,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.0126582278481013,
|
| 714 |
+
"grad_norm": 1.5737881952906483,
|
| 715 |
+
"learning_rate": 5e-06,
|
| 716 |
+
"loss": 0.4633,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.0227848101265822,
|
| 721 |
+
"grad_norm": 1.5616941170427905,
|
| 722 |
+
"learning_rate": 5e-06,
|
| 723 |
+
"loss": 0.4634,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.0329113924050632,
|
| 728 |
+
"grad_norm": 1.5346519713859017,
|
| 729 |
+
"learning_rate": 5e-06,
|
| 730 |
+
"loss": 0.4435,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.0430379746835443,
|
| 735 |
+
"grad_norm": 1.3492138788127321,
|
| 736 |
+
"learning_rate": 5e-06,
|
| 737 |
+
"loss": 0.4503,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.0531645569620254,
|
| 742 |
+
"grad_norm": 1.4763337790140822,
|
| 743 |
+
"learning_rate": 5e-06,
|
| 744 |
+
"loss": 0.4506,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.0632911392405062,
|
| 749 |
+
"grad_norm": 1.4225225244349122,
|
| 750 |
+
"learning_rate": 5e-06,
|
| 751 |
+
"loss": 0.4591,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.0734177215189873,
|
| 756 |
+
"grad_norm": 1.377686472346139,
|
| 757 |
+
"learning_rate": 5e-06,
|
| 758 |
+
"loss": 0.4498,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.0835443037974684,
|
| 763 |
+
"grad_norm": 1.524899956605274,
|
| 764 |
+
"learning_rate": 5e-06,
|
| 765 |
+
"loss": 0.4521,
|
| 766 |
+
"step": 1070
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.0936708860759494,
|
| 770 |
+
"grad_norm": 1.4555671547809872,
|
| 771 |
+
"learning_rate": 5e-06,
|
| 772 |
+
"loss": 0.4537,
|
| 773 |
+
"step": 1080
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.1037974683544305,
|
| 777 |
+
"grad_norm": 1.352652609202294,
|
| 778 |
+
"learning_rate": 5e-06,
|
| 779 |
+
"loss": 0.4591,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.1139240506329113,
|
| 784 |
+
"grad_norm": 1.389780851883711,
|
| 785 |
+
"learning_rate": 5e-06,
|
| 786 |
+
"loss": 0.451,
|
| 787 |
+
"step": 1100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.1240506329113924,
|
| 791 |
+
"grad_norm": 1.5493151434769157,
|
| 792 |
+
"learning_rate": 5e-06,
|
| 793 |
+
"loss": 0.4545,
|
| 794 |
+
"step": 1110
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.1341772151898735,
|
| 798 |
+
"grad_norm": 1.4921163467682323,
|
| 799 |
+
"learning_rate": 5e-06,
|
| 800 |
+
"loss": 0.4511,
|
| 801 |
+
"step": 1120
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.1443037974683543,
|
| 805 |
+
"grad_norm": 1.3424106662166821,
|
| 806 |
+
"learning_rate": 5e-06,
|
| 807 |
+
"loss": 0.4597,
|
| 808 |
+
"step": 1130
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.1544303797468354,
|
| 812 |
+
"grad_norm": 1.4205445966678738,
|
| 813 |
+
"learning_rate": 5e-06,
|
| 814 |
+
"loss": 0.4587,
|
| 815 |
+
"step": 1140
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.1645569620253164,
|
| 819 |
+
"grad_norm": 1.3812137606947525,
|
| 820 |
+
"learning_rate": 5e-06,
|
| 821 |
+
"loss": 0.457,
|
| 822 |
+
"step": 1150
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.1746835443037975,
|
| 826 |
+
"grad_norm": 1.504094206656827,
|
| 827 |
+
"learning_rate": 5e-06,
|
| 828 |
+
"loss": 0.4582,
|
| 829 |
+
"step": 1160
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.1848101265822786,
|
| 833 |
+
"grad_norm": 1.4384269712621547,
|
| 834 |
+
"learning_rate": 5e-06,
|
| 835 |
+
"loss": 0.4478,
|
| 836 |
+
"step": 1170
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.1949367088607594,
|
| 840 |
+
"grad_norm": 1.4604436825013358,
|
| 841 |
+
"learning_rate": 5e-06,
|
| 842 |
+
"loss": 0.4561,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.2050632911392405,
|
| 847 |
+
"grad_norm": 1.484357252258588,
|
| 848 |
+
"learning_rate": 5e-06,
|
| 849 |
+
"loss": 0.463,
|
| 850 |
+
"step": 1190
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.2151898734177216,
|
| 854 |
+
"grad_norm": 1.3507024353159274,
|
| 855 |
+
"learning_rate": 5e-06,
|
| 856 |
+
"loss": 0.4557,
|
| 857 |
+
"step": 1200
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.2253164556962026,
|
| 861 |
+
"grad_norm": 1.5356768758566504,
|
| 862 |
+
"learning_rate": 5e-06,
|
| 863 |
+
"loss": 0.4616,
|
| 864 |
+
"step": 1210
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.2354430379746835,
|
| 868 |
+
"grad_norm": 1.507777364560312,
|
| 869 |
+
"learning_rate": 5e-06,
|
| 870 |
+
"loss": 0.4492,
|
| 871 |
+
"step": 1220
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.2455696202531645,
|
| 875 |
+
"grad_norm": 1.402742621286163,
|
| 876 |
+
"learning_rate": 5e-06,
|
| 877 |
+
"loss": 0.4623,
|
| 878 |
+
"step": 1230
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.2556962025316456,
|
| 882 |
+
"grad_norm": 1.4914361590080172,
|
| 883 |
+
"learning_rate": 5e-06,
|
| 884 |
+
"loss": 0.4576,
|
| 885 |
+
"step": 1240
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.2658227848101267,
|
| 889 |
+
"grad_norm": 1.4449809880641054,
|
| 890 |
+
"learning_rate": 5e-06,
|
| 891 |
+
"loss": 0.4581,
|
| 892 |
+
"step": 1250
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.2759493670886077,
|
| 896 |
+
"grad_norm": 1.401311614162818,
|
| 897 |
+
"learning_rate": 5e-06,
|
| 898 |
+
"loss": 0.4613,
|
| 899 |
+
"step": 1260
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.2860759493670886,
|
| 903 |
+
"grad_norm": 1.4035116916924748,
|
| 904 |
+
"learning_rate": 5e-06,
|
| 905 |
+
"loss": 0.4546,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.2962025316455696,
|
| 910 |
+
"grad_norm": 1.5360414127344846,
|
| 911 |
+
"learning_rate": 5e-06,
|
| 912 |
+
"loss": 0.4638,
|
| 913 |
+
"step": 1280
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.3063291139240507,
|
| 917 |
+
"grad_norm": 1.4721064580495236,
|
| 918 |
+
"learning_rate": 5e-06,
|
| 919 |
+
"loss": 0.4605,
|
| 920 |
+
"step": 1290
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.3164556962025316,
|
| 924 |
+
"grad_norm": 1.5018662441760837,
|
| 925 |
+
"learning_rate": 5e-06,
|
| 926 |
+
"loss": 0.4656,
|
| 927 |
+
"step": 1300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.3265822784810126,
|
| 931 |
+
"grad_norm": 1.4057750381381058,
|
| 932 |
+
"learning_rate": 5e-06,
|
| 933 |
+
"loss": 0.4546,
|
| 934 |
+
"step": 1310
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.3367088607594937,
|
| 938 |
+
"grad_norm": 1.472547127671551,
|
| 939 |
+
"learning_rate": 5e-06,
|
| 940 |
+
"loss": 0.4509,
|
| 941 |
+
"step": 1320
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.3468354430379748,
|
| 945 |
+
"grad_norm": 1.378222631334251,
|
| 946 |
+
"learning_rate": 5e-06,
|
| 947 |
+
"loss": 0.4592,
|
| 948 |
+
"step": 1330
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.3569620253164558,
|
| 952 |
+
"grad_norm": 1.477645235480499,
|
| 953 |
+
"learning_rate": 5e-06,
|
| 954 |
+
"loss": 0.4731,
|
| 955 |
+
"step": 1340
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.3670886075949367,
|
| 959 |
+
"grad_norm": 1.4190131074856087,
|
| 960 |
+
"learning_rate": 5e-06,
|
| 961 |
+
"loss": 0.4614,
|
| 962 |
+
"step": 1350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.3772151898734177,
|
| 966 |
+
"grad_norm": 1.598773468818255,
|
| 967 |
+
"learning_rate": 5e-06,
|
| 968 |
+
"loss": 0.4642,
|
| 969 |
+
"step": 1360
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.3873417721518988,
|
| 973 |
+
"grad_norm": 1.401701470107724,
|
| 974 |
+
"learning_rate": 5e-06,
|
| 975 |
+
"loss": 0.4599,
|
| 976 |
+
"step": 1370
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.3974683544303796,
|
| 980 |
+
"grad_norm": 1.5834771319486265,
|
| 981 |
+
"learning_rate": 5e-06,
|
| 982 |
+
"loss": 0.4635,
|
| 983 |
+
"step": 1380
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.4075949367088607,
|
| 987 |
+
"grad_norm": 1.398082340799174,
|
| 988 |
+
"learning_rate": 5e-06,
|
| 989 |
+
"loss": 0.4581,
|
| 990 |
+
"step": 1390
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.4177215189873418,
|
| 994 |
+
"grad_norm": 1.4006340071029044,
|
| 995 |
+
"learning_rate": 5e-06,
|
| 996 |
+
"loss": 0.4602,
|
| 997 |
+
"step": 1400
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.4278481012658228,
|
| 1001 |
+
"grad_norm": 1.3336328229398797,
|
| 1002 |
+
"learning_rate": 5e-06,
|
| 1003 |
+
"loss": 0.4602,
|
| 1004 |
+
"step": 1410
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.437974683544304,
|
| 1008 |
+
"grad_norm": 1.3492817883075563,
|
| 1009 |
+
"learning_rate": 5e-06,
|
| 1010 |
+
"loss": 0.4687,
|
| 1011 |
+
"step": 1420
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.4481012658227848,
|
| 1015 |
+
"grad_norm": 1.342687219512621,
|
| 1016 |
+
"learning_rate": 5e-06,
|
| 1017 |
+
"loss": 0.4651,
|
| 1018 |
+
"step": 1430
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.4582278481012658,
|
| 1022 |
+
"grad_norm": 1.4950318280118493,
|
| 1023 |
+
"learning_rate": 5e-06,
|
| 1024 |
+
"loss": 0.4643,
|
| 1025 |
+
"step": 1440
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.4683544303797469,
|
| 1029 |
+
"grad_norm": 1.5341419482481558,
|
| 1030 |
+
"learning_rate": 5e-06,
|
| 1031 |
+
"loss": 0.467,
|
| 1032 |
+
"step": 1450
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.4784810126582277,
|
| 1036 |
+
"grad_norm": 1.3873503711966886,
|
| 1037 |
+
"learning_rate": 5e-06,
|
| 1038 |
+
"loss": 0.4692,
|
| 1039 |
+
"step": 1460
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.4886075949367088,
|
| 1043 |
+
"grad_norm": 1.3838670814377412,
|
| 1044 |
+
"learning_rate": 5e-06,
|
| 1045 |
+
"loss": 0.4606,
|
| 1046 |
+
"step": 1470
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.4987341772151899,
|
| 1050 |
+
"grad_norm": 1.435159074036311,
|
| 1051 |
+
"learning_rate": 5e-06,
|
| 1052 |
+
"loss": 0.4607,
|
| 1053 |
+
"step": 1480
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.508860759493671,
|
| 1057 |
+
"grad_norm": 1.416449038632833,
|
| 1058 |
+
"learning_rate": 5e-06,
|
| 1059 |
+
"loss": 0.4603,
|
| 1060 |
+
"step": 1490
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.518987341772152,
|
| 1064 |
+
"grad_norm": 1.349053423409347,
|
| 1065 |
+
"learning_rate": 5e-06,
|
| 1066 |
+
"loss": 0.4671,
|
| 1067 |
+
"step": 1500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.529113924050633,
|
| 1071 |
+
"grad_norm": 1.5418750945784405,
|
| 1072 |
+
"learning_rate": 5e-06,
|
| 1073 |
+
"loss": 0.4642,
|
| 1074 |
+
"step": 1510
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.539240506329114,
|
| 1078 |
+
"grad_norm": 1.5350888472278188,
|
| 1079 |
+
"learning_rate": 5e-06,
|
| 1080 |
+
"loss": 0.4654,
|
| 1081 |
+
"step": 1520
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.549367088607595,
|
| 1085 |
+
"grad_norm": 1.5918459039624637,
|
| 1086 |
+
"learning_rate": 5e-06,
|
| 1087 |
+
"loss": 0.4623,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.5594936708860758,
|
| 1092 |
+
"grad_norm": 1.4044783253886717,
|
| 1093 |
+
"learning_rate": 5e-06,
|
| 1094 |
+
"loss": 0.4644,
|
| 1095 |
+
"step": 1540
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.5696202531645569,
|
| 1099 |
+
"grad_norm": 1.4733622287724428,
|
| 1100 |
+
"learning_rate": 5e-06,
|
| 1101 |
+
"loss": 0.4699,
|
| 1102 |
+
"step": 1550
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.579746835443038,
|
| 1106 |
+
"grad_norm": 1.4190534275279456,
|
| 1107 |
+
"learning_rate": 5e-06,
|
| 1108 |
+
"loss": 0.4585,
|
| 1109 |
+
"step": 1560
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.589873417721519,
|
| 1113 |
+
"grad_norm": 1.588541120324257,
|
| 1114 |
+
"learning_rate": 5e-06,
|
| 1115 |
+
"loss": 0.4674,
|
| 1116 |
+
"step": 1570
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.6,
|
| 1120 |
+
"grad_norm": 1.4489487698353771,
|
| 1121 |
+
"learning_rate": 5e-06,
|
| 1122 |
+
"loss": 0.4659,
|
| 1123 |
+
"step": 1580
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.6101265822784812,
|
| 1127 |
+
"grad_norm": 1.3499923545579962,
|
| 1128 |
+
"learning_rate": 5e-06,
|
| 1129 |
+
"loss": 0.4582,
|
| 1130 |
+
"step": 1590
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.620253164556962,
|
| 1134 |
+
"grad_norm": 1.3640430679010667,
|
| 1135 |
+
"learning_rate": 5e-06,
|
| 1136 |
+
"loss": 0.4676,
|
| 1137 |
+
"step": 1600
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.630379746835443,
|
| 1141 |
+
"grad_norm": 1.3690059243296218,
|
| 1142 |
+
"learning_rate": 5e-06,
|
| 1143 |
+
"loss": 0.4612,
|
| 1144 |
+
"step": 1610
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.640506329113924,
|
| 1148 |
+
"grad_norm": 1.2447374210211029,
|
| 1149 |
+
"learning_rate": 5e-06,
|
| 1150 |
+
"loss": 0.4506,
|
| 1151 |
+
"step": 1620
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.650632911392405,
|
| 1155 |
+
"grad_norm": 1.351219366349358,
|
| 1156 |
+
"learning_rate": 5e-06,
|
| 1157 |
+
"loss": 0.4611,
|
| 1158 |
+
"step": 1630
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.660759493670886,
|
| 1162 |
+
"grad_norm": 1.335122455907824,
|
| 1163 |
+
"learning_rate": 5e-06,
|
| 1164 |
+
"loss": 0.4699,
|
| 1165 |
+
"step": 1640
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.6708860759493671,
|
| 1169 |
+
"grad_norm": 1.5625860035007904,
|
| 1170 |
+
"learning_rate": 5e-06,
|
| 1171 |
+
"loss": 0.4612,
|
| 1172 |
+
"step": 1650
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.6810126582278482,
|
| 1176 |
+
"grad_norm": 1.6296205380869357,
|
| 1177 |
+
"learning_rate": 5e-06,
|
| 1178 |
+
"loss": 0.4623,
|
| 1179 |
+
"step": 1660
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.6911392405063292,
|
| 1183 |
+
"grad_norm": 1.4330065379895607,
|
| 1184 |
+
"learning_rate": 5e-06,
|
| 1185 |
+
"loss": 0.4625,
|
| 1186 |
+
"step": 1670
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.70126582278481,
|
| 1190 |
+
"grad_norm": 1.3278761181861125,
|
| 1191 |
+
"learning_rate": 5e-06,
|
| 1192 |
+
"loss": 0.4627,
|
| 1193 |
+
"step": 1680
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.7113924050632912,
|
| 1197 |
+
"grad_norm": 1.4304366765861536,
|
| 1198 |
+
"learning_rate": 5e-06,
|
| 1199 |
+
"loss": 0.4646,
|
| 1200 |
+
"step": 1690
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.721518987341772,
|
| 1204 |
+
"grad_norm": 1.3460103891901807,
|
| 1205 |
+
"learning_rate": 5e-06,
|
| 1206 |
+
"loss": 0.4648,
|
| 1207 |
+
"step": 1700
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.731645569620253,
|
| 1211 |
+
"grad_norm": 1.4336438757742112,
|
| 1212 |
+
"learning_rate": 5e-06,
|
| 1213 |
+
"loss": 0.4708,
|
| 1214 |
+
"step": 1710
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.7417721518987341,
|
| 1218 |
+
"grad_norm": 1.333163056745564,
|
| 1219 |
+
"learning_rate": 5e-06,
|
| 1220 |
+
"loss": 0.4591,
|
| 1221 |
+
"step": 1720
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.7518987341772152,
|
| 1225 |
+
"grad_norm": 1.3819280069426765,
|
| 1226 |
+
"learning_rate": 5e-06,
|
| 1227 |
+
"loss": 0.4651,
|
| 1228 |
+
"step": 1730
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.7620253164556963,
|
| 1232 |
+
"grad_norm": 1.413028622360716,
|
| 1233 |
+
"learning_rate": 5e-06,
|
| 1234 |
+
"loss": 0.4714,
|
| 1235 |
+
"step": 1740
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.7721518987341773,
|
| 1239 |
+
"grad_norm": 1.3989242299354576,
|
| 1240 |
+
"learning_rate": 5e-06,
|
| 1241 |
+
"loss": 0.466,
|
| 1242 |
+
"step": 1750
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.7822784810126582,
|
| 1246 |
+
"grad_norm": 1.3712554248183217,
|
| 1247 |
+
"learning_rate": 5e-06,
|
| 1248 |
+
"loss": 0.4727,
|
| 1249 |
+
"step": 1760
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.7924050632911392,
|
| 1253 |
+
"grad_norm": 1.4780769754046013,
|
| 1254 |
+
"learning_rate": 5e-06,
|
| 1255 |
+
"loss": 0.4679,
|
| 1256 |
+
"step": 1770
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.80253164556962,
|
| 1260 |
+
"grad_norm": 1.4002831190634755,
|
| 1261 |
+
"learning_rate": 5e-06,
|
| 1262 |
+
"loss": 0.4647,
|
| 1263 |
+
"step": 1780
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.8126582278481012,
|
| 1267 |
+
"grad_norm": 1.3944675393169206,
|
| 1268 |
+
"learning_rate": 5e-06,
|
| 1269 |
+
"loss": 0.4644,
|
| 1270 |
+
"step": 1790
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.8227848101265822,
|
| 1274 |
+
"grad_norm": 1.4867966880006709,
|
| 1275 |
+
"learning_rate": 5e-06,
|
| 1276 |
+
"loss": 0.4675,
|
| 1277 |
+
"step": 1800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.8329113924050633,
|
| 1281 |
+
"grad_norm": 1.404496416486814,
|
| 1282 |
+
"learning_rate": 5e-06,
|
| 1283 |
+
"loss": 0.4624,
|
| 1284 |
+
"step": 1810
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.8430379746835444,
|
| 1288 |
+
"grad_norm": 1.5305845929183253,
|
| 1289 |
+
"learning_rate": 5e-06,
|
| 1290 |
+
"loss": 0.4739,
|
| 1291 |
+
"step": 1820
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.8531645569620254,
|
| 1295 |
+
"grad_norm": 1.3084464311737185,
|
| 1296 |
+
"learning_rate": 5e-06,
|
| 1297 |
+
"loss": 0.4655,
|
| 1298 |
+
"step": 1830
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.8632911392405065,
|
| 1302 |
+
"grad_norm": 1.3059438675535409,
|
| 1303 |
+
"learning_rate": 5e-06,
|
| 1304 |
+
"loss": 0.4678,
|
| 1305 |
+
"step": 1840
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.8734177215189873,
|
| 1309 |
+
"grad_norm": 1.3867484618996055,
|
| 1310 |
+
"learning_rate": 5e-06,
|
| 1311 |
+
"loss": 0.4609,
|
| 1312 |
+
"step": 1850
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.8835443037974684,
|
| 1316 |
+
"grad_norm": 1.3420942943215763,
|
| 1317 |
+
"learning_rate": 5e-06,
|
| 1318 |
+
"loss": 0.4663,
|
| 1319 |
+
"step": 1860
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.8936708860759492,
|
| 1323 |
+
"grad_norm": 1.421266278043212,
|
| 1324 |
+
"learning_rate": 5e-06,
|
| 1325 |
+
"loss": 0.4665,
|
| 1326 |
+
"step": 1870
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.9037974683544303,
|
| 1330 |
+
"grad_norm": 1.36634296470369,
|
| 1331 |
+
"learning_rate": 5e-06,
|
| 1332 |
+
"loss": 0.4676,
|
| 1333 |
+
"step": 1880
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.9139240506329114,
|
| 1337 |
+
"grad_norm": 1.2659671091408768,
|
| 1338 |
+
"learning_rate": 5e-06,
|
| 1339 |
+
"loss": 0.4669,
|
| 1340 |
+
"step": 1890
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.9240506329113924,
|
| 1344 |
+
"grad_norm": 1.373264332933901,
|
| 1345 |
+
"learning_rate": 5e-06,
|
| 1346 |
+
"loss": 0.4701,
|
| 1347 |
+
"step": 1900
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.9341772151898735,
|
| 1351 |
+
"grad_norm": 1.3568414058185374,
|
| 1352 |
+
"learning_rate": 5e-06,
|
| 1353 |
+
"loss": 0.4693,
|
| 1354 |
+
"step": 1910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.9443037974683546,
|
| 1358 |
+
"grad_norm": 1.3940261082697252,
|
| 1359 |
+
"learning_rate": 5e-06,
|
| 1360 |
+
"loss": 0.4763,
|
| 1361 |
+
"step": 1920
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.9544303797468354,
|
| 1365 |
+
"grad_norm": 1.3879188735035681,
|
| 1366 |
+
"learning_rate": 5e-06,
|
| 1367 |
+
"loss": 0.4756,
|
| 1368 |
+
"step": 1930
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.9645569620253165,
|
| 1372 |
+
"grad_norm": 1.363003506285662,
|
| 1373 |
+
"learning_rate": 5e-06,
|
| 1374 |
+
"loss": 0.4667,
|
| 1375 |
+
"step": 1940
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.9746835443037973,
|
| 1379 |
+
"grad_norm": 1.4533412320271968,
|
| 1380 |
+
"learning_rate": 5e-06,
|
| 1381 |
+
"loss": 0.469,
|
| 1382 |
+
"step": 1950
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.9848101265822784,
|
| 1386 |
+
"grad_norm": 1.4479310447515883,
|
| 1387 |
+
"learning_rate": 5e-06,
|
| 1388 |
+
"loss": 0.4748,
|
| 1389 |
+
"step": 1960
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.9949367088607595,
|
| 1393 |
+
"grad_norm": 1.2966648783230335,
|
| 1394 |
+
"learning_rate": 5e-06,
|
| 1395 |
+
"loss": 0.468,
|
| 1396 |
+
"step": 1970
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.0,
|
| 1400 |
+
"eval_loss": 0.1398227959871292,
|
| 1401 |
+
"eval_runtime": 506.8788,
|
| 1402 |
+
"eval_samples_per_second": 26.247,
|
| 1403 |
+
"eval_steps_per_second": 0.41,
|
| 1404 |
+
"step": 1975
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.0050632911392405,
|
| 1408 |
+
"grad_norm": 3.06600016953371,
|
| 1409 |
+
"learning_rate": 5e-06,
|
| 1410 |
+
"loss": 0.4085,
|
| 1411 |
+
"step": 1980
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.0151898734177216,
|
| 1415 |
+
"grad_norm": 2.0901467389116335,
|
| 1416 |
+
"learning_rate": 5e-06,
|
| 1417 |
+
"loss": 0.3545,
|
| 1418 |
+
"step": 1990
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.0253164556962027,
|
| 1422 |
+
"grad_norm": 1.7198026240231687,
|
| 1423 |
+
"learning_rate": 5e-06,
|
| 1424 |
+
"loss": 0.3486,
|
| 1425 |
+
"step": 2000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 2.0354430379746837,
|
| 1429 |
+
"grad_norm": 1.6596559930998485,
|
| 1430 |
+
"learning_rate": 5e-06,
|
| 1431 |
+
"loss": 0.3485,
|
| 1432 |
+
"step": 2010
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 2.0455696202531644,
|
| 1436 |
+
"grad_norm": 1.6010412767976676,
|
| 1437 |
+
"learning_rate": 5e-06,
|
| 1438 |
+
"loss": 0.3469,
|
| 1439 |
+
"step": 2020
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.0556962025316454,
|
| 1443 |
+
"grad_norm": 1.619456570416184,
|
| 1444 |
+
"learning_rate": 5e-06,
|
| 1445 |
+
"loss": 0.3412,
|
| 1446 |
+
"step": 2030
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 2.0658227848101265,
|
| 1450 |
+
"grad_norm": 1.5477510460994646,
|
| 1451 |
+
"learning_rate": 5e-06,
|
| 1452 |
+
"loss": 0.3499,
|
| 1453 |
+
"step": 2040
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 2.0759493670886076,
|
| 1457 |
+
"grad_norm": 1.6995317518377548,
|
| 1458 |
+
"learning_rate": 5e-06,
|
| 1459 |
+
"loss": 0.3431,
|
| 1460 |
+
"step": 2050
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.0860759493670886,
|
| 1464 |
+
"grad_norm": 1.6672377260042517,
|
| 1465 |
+
"learning_rate": 5e-06,
|
| 1466 |
+
"loss": 0.3509,
|
| 1467 |
+
"step": 2060
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 2.0962025316455697,
|
| 1471 |
+
"grad_norm": 1.5137523020681678,
|
| 1472 |
+
"learning_rate": 5e-06,
|
| 1473 |
+
"loss": 0.3485,
|
| 1474 |
+
"step": 2070
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 2.1063291139240508,
|
| 1478 |
+
"grad_norm": 1.5728338684227074,
|
| 1479 |
+
"learning_rate": 5e-06,
|
| 1480 |
+
"loss": 0.3414,
|
| 1481 |
+
"step": 2080
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 2.116455696202532,
|
| 1485 |
+
"grad_norm": 1.7215854826795536,
|
| 1486 |
+
"learning_rate": 5e-06,
|
| 1487 |
+
"loss": 0.356,
|
| 1488 |
+
"step": 2090
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 2.1265822784810124,
|
| 1492 |
+
"grad_norm": 1.547269577634038,
|
| 1493 |
+
"learning_rate": 5e-06,
|
| 1494 |
+
"loss": 0.351,
|
| 1495 |
+
"step": 2100
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 2.1367088607594935,
|
| 1499 |
+
"grad_norm": 1.7846477440752144,
|
| 1500 |
+
"learning_rate": 5e-06,
|
| 1501 |
+
"loss": 0.3475,
|
| 1502 |
+
"step": 2110
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.1468354430379746,
|
| 1506 |
+
"grad_norm": 1.6766174751768645,
|
| 1507 |
+
"learning_rate": 5e-06,
|
| 1508 |
+
"loss": 0.3507,
|
| 1509 |
+
"step": 2120
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 2.1569620253164556,
|
| 1513 |
+
"grad_norm": 1.6129490070505184,
|
| 1514 |
+
"learning_rate": 5e-06,
|
| 1515 |
+
"loss": 0.3544,
|
| 1516 |
+
"step": 2130
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.1670886075949367,
|
| 1520 |
+
"grad_norm": 1.6006266055236678,
|
| 1521 |
+
"learning_rate": 5e-06,
|
| 1522 |
+
"loss": 0.3514,
|
| 1523 |
+
"step": 2140
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 2.1772151898734178,
|
| 1527 |
+
"grad_norm": 1.5978325737155568,
|
| 1528 |
+
"learning_rate": 5e-06,
|
| 1529 |
+
"loss": 0.3525,
|
| 1530 |
+
"step": 2150
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 2.187341772151899,
|
| 1534 |
+
"grad_norm": 1.5951449058616392,
|
| 1535 |
+
"learning_rate": 5e-06,
|
| 1536 |
+
"loss": 0.3507,
|
| 1537 |
+
"step": 2160
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 2.19746835443038,
|
| 1541 |
+
"grad_norm": 1.5513789073675182,
|
| 1542 |
+
"learning_rate": 5e-06,
|
| 1543 |
+
"loss": 0.3496,
|
| 1544 |
+
"step": 2170
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.207594936708861,
|
| 1548 |
+
"grad_norm": 1.5229143655144641,
|
| 1549 |
+
"learning_rate": 5e-06,
|
| 1550 |
+
"loss": 0.3494,
|
| 1551 |
+
"step": 2180
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 2.2177215189873416,
|
| 1555 |
+
"grad_norm": 1.660242717310412,
|
| 1556 |
+
"learning_rate": 5e-06,
|
| 1557 |
+
"loss": 0.3553,
|
| 1558 |
+
"step": 2190
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 2.2278481012658227,
|
| 1562 |
+
"grad_norm": 1.6739776038113652,
|
| 1563 |
+
"learning_rate": 5e-06,
|
| 1564 |
+
"loss": 0.3607,
|
| 1565 |
+
"step": 2200
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 2.2379746835443037,
|
| 1569 |
+
"grad_norm": 1.7005249152138244,
|
| 1570 |
+
"learning_rate": 5e-06,
|
| 1571 |
+
"loss": 0.3486,
|
| 1572 |
+
"step": 2210
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 2.248101265822785,
|
| 1576 |
+
"grad_norm": 1.6489218772469403,
|
| 1577 |
+
"learning_rate": 5e-06,
|
| 1578 |
+
"loss": 0.3592,
|
| 1579 |
+
"step": 2220
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 2.258227848101266,
|
| 1583 |
+
"grad_norm": 1.6209760997530658,
|
| 1584 |
+
"learning_rate": 5e-06,
|
| 1585 |
+
"loss": 0.3584,
|
| 1586 |
+
"step": 2230
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.268354430379747,
|
| 1590 |
+
"grad_norm": 1.6153685359109242,
|
| 1591 |
+
"learning_rate": 5e-06,
|
| 1592 |
+
"loss": 0.3554,
|
| 1593 |
+
"step": 2240
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 2.278481012658228,
|
| 1597 |
+
"grad_norm": 1.6268217139529644,
|
| 1598 |
+
"learning_rate": 5e-06,
|
| 1599 |
+
"loss": 0.3542,
|
| 1600 |
+
"step": 2250
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 2.2886075949367086,
|
| 1604 |
+
"grad_norm": 1.7752820541706038,
|
| 1605 |
+
"learning_rate": 5e-06,
|
| 1606 |
+
"loss": 0.3621,
|
| 1607 |
+
"step": 2260
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.2987341772151897,
|
| 1611 |
+
"grad_norm": 1.5709969247849427,
|
| 1612 |
+
"learning_rate": 5e-06,
|
| 1613 |
+
"loss": 0.3524,
|
| 1614 |
+
"step": 2270
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.3088607594936708,
|
| 1618 |
+
"grad_norm": 1.5534897904208944,
|
| 1619 |
+
"learning_rate": 5e-06,
|
| 1620 |
+
"loss": 0.3563,
|
| 1621 |
+
"step": 2280
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.318987341772152,
|
| 1625 |
+
"grad_norm": 1.6608945730503153,
|
| 1626 |
+
"learning_rate": 5e-06,
|
| 1627 |
+
"loss": 0.3565,
|
| 1628 |
+
"step": 2290
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.329113924050633,
|
| 1632 |
+
"grad_norm": 1.5801622846665264,
|
| 1633 |
+
"learning_rate": 5e-06,
|
| 1634 |
+
"loss": 0.3622,
|
| 1635 |
+
"step": 2300
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 2.339240506329114,
|
| 1639 |
+
"grad_norm": 1.5320500459364543,
|
| 1640 |
+
"learning_rate": 5e-06,
|
| 1641 |
+
"loss": 0.3589,
|
| 1642 |
+
"step": 2310
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 2.349367088607595,
|
| 1646 |
+
"grad_norm": 1.581287461651533,
|
| 1647 |
+
"learning_rate": 5e-06,
|
| 1648 |
+
"loss": 0.3613,
|
| 1649 |
+
"step": 2320
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 2.359493670886076,
|
| 1653 |
+
"grad_norm": 1.6038211850984352,
|
| 1654 |
+
"learning_rate": 5e-06,
|
| 1655 |
+
"loss": 0.3617,
|
| 1656 |
+
"step": 2330
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 2.369620253164557,
|
| 1660 |
+
"grad_norm": 1.625966530104118,
|
| 1661 |
+
"learning_rate": 5e-06,
|
| 1662 |
+
"loss": 0.3598,
|
| 1663 |
+
"step": 2340
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 2.379746835443038,
|
| 1667 |
+
"grad_norm": 1.7645641137037487,
|
| 1668 |
+
"learning_rate": 5e-06,
|
| 1669 |
+
"loss": 0.362,
|
| 1670 |
+
"step": 2350
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.389873417721519,
|
| 1674 |
+
"grad_norm": 1.6940524616595496,
|
| 1675 |
+
"learning_rate": 5e-06,
|
| 1676 |
+
"loss": 0.361,
|
| 1677 |
+
"step": 2360
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 2.4,
|
| 1681 |
+
"grad_norm": 1.5384052283713547,
|
| 1682 |
+
"learning_rate": 5e-06,
|
| 1683 |
+
"loss": 0.3639,
|
| 1684 |
+
"step": 2370
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 2.410126582278481,
|
| 1688 |
+
"grad_norm": 1.6483060393926068,
|
| 1689 |
+
"learning_rate": 5e-06,
|
| 1690 |
+
"loss": 0.3586,
|
| 1691 |
+
"step": 2380
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 2.420253164556962,
|
| 1695 |
+
"grad_norm": 1.6074760148392138,
|
| 1696 |
+
"learning_rate": 5e-06,
|
| 1697 |
+
"loss": 0.3648,
|
| 1698 |
+
"step": 2390
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 2.430379746835443,
|
| 1702 |
+
"grad_norm": 1.7080201956192906,
|
| 1703 |
+
"learning_rate": 5e-06,
|
| 1704 |
+
"loss": 0.3554,
|
| 1705 |
+
"step": 2400
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 2.440506329113924,
|
| 1709 |
+
"grad_norm": 1.625757072296565,
|
| 1710 |
+
"learning_rate": 5e-06,
|
| 1711 |
+
"loss": 0.3531,
|
| 1712 |
+
"step": 2410
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 2.4506329113924052,
|
| 1716 |
+
"grad_norm": 1.6544192748138193,
|
| 1717 |
+
"learning_rate": 5e-06,
|
| 1718 |
+
"loss": 0.3659,
|
| 1719 |
+
"step": 2420
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 2.460759493670886,
|
| 1723 |
+
"grad_norm": 1.6016056620993897,
|
| 1724 |
+
"learning_rate": 5e-06,
|
| 1725 |
+
"loss": 0.3618,
|
| 1726 |
+
"step": 2430
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 2.470886075949367,
|
| 1730 |
+
"grad_norm": 1.6466444516635712,
|
| 1731 |
+
"learning_rate": 5e-06,
|
| 1732 |
+
"loss": 0.3651,
|
| 1733 |
+
"step": 2440
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 2.481012658227848,
|
| 1737 |
+
"grad_norm": 1.6213376274576505,
|
| 1738 |
+
"learning_rate": 5e-06,
|
| 1739 |
+
"loss": 0.3611,
|
| 1740 |
+
"step": 2450
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 2.491139240506329,
|
| 1744 |
+
"grad_norm": 1.61939669911579,
|
| 1745 |
+
"learning_rate": 5e-06,
|
| 1746 |
+
"loss": 0.3655,
|
| 1747 |
+
"step": 2460
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 2.50126582278481,
|
| 1751 |
+
"grad_norm": 1.7185252999562315,
|
| 1752 |
+
"learning_rate": 5e-06,
|
| 1753 |
+
"loss": 0.3717,
|
| 1754 |
+
"step": 2470
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 2.511392405063291,
|
| 1758 |
+
"grad_norm": 1.5701430968600476,
|
| 1759 |
+
"learning_rate": 5e-06,
|
| 1760 |
+
"loss": 0.3616,
|
| 1761 |
+
"step": 2480
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 2.5215189873417723,
|
| 1765 |
+
"grad_norm": 1.6525891145092144,
|
| 1766 |
+
"learning_rate": 5e-06,
|
| 1767 |
+
"loss": 0.3618,
|
| 1768 |
+
"step": 2490
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 2.5316455696202533,
|
| 1772 |
+
"grad_norm": 1.6331648861432553,
|
| 1773 |
+
"learning_rate": 5e-06,
|
| 1774 |
+
"loss": 0.3608,
|
| 1775 |
+
"step": 2500
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 2.5417721518987344,
|
| 1779 |
+
"grad_norm": 1.5954836812756767,
|
| 1780 |
+
"learning_rate": 5e-06,
|
| 1781 |
+
"loss": 0.3564,
|
| 1782 |
+
"step": 2510
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 2.5518987341772155,
|
| 1786 |
+
"grad_norm": 1.6221389574344411,
|
| 1787 |
+
"learning_rate": 5e-06,
|
| 1788 |
+
"loss": 0.3603,
|
| 1789 |
+
"step": 2520
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 2.562025316455696,
|
| 1793 |
+
"grad_norm": 1.645557544125129,
|
| 1794 |
+
"learning_rate": 5e-06,
|
| 1795 |
+
"loss": 0.3633,
|
| 1796 |
+
"step": 2530
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 2.572151898734177,
|
| 1800 |
+
"grad_norm": 1.7704063332654312,
|
| 1801 |
+
"learning_rate": 5e-06,
|
| 1802 |
+
"loss": 0.3658,
|
| 1803 |
+
"step": 2540
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 2.5822784810126582,
|
| 1807 |
+
"grad_norm": 1.6779345444764124,
|
| 1808 |
+
"learning_rate": 5e-06,
|
| 1809 |
+
"loss": 0.3706,
|
| 1810 |
+
"step": 2550
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 2.5924050632911393,
|
| 1814 |
+
"grad_norm": 1.6024752118154975,
|
| 1815 |
+
"learning_rate": 5e-06,
|
| 1816 |
+
"loss": 0.3618,
|
| 1817 |
+
"step": 2560
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 2.6025316455696204,
|
| 1821 |
+
"grad_norm": 1.5688766989229237,
|
| 1822 |
+
"learning_rate": 5e-06,
|
| 1823 |
+
"loss": 0.3719,
|
| 1824 |
+
"step": 2570
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 2.6126582278481014,
|
| 1828 |
+
"grad_norm": 1.5972321010318007,
|
| 1829 |
+
"learning_rate": 5e-06,
|
| 1830 |
+
"loss": 0.3695,
|
| 1831 |
+
"step": 2580
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 2.622784810126582,
|
| 1835 |
+
"grad_norm": 1.7926779657765413,
|
| 1836 |
+
"learning_rate": 5e-06,
|
| 1837 |
+
"loss": 0.3671,
|
| 1838 |
+
"step": 2590
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.632911392405063,
|
| 1842 |
+
"grad_norm": 1.6497318559260037,
|
| 1843 |
+
"learning_rate": 5e-06,
|
| 1844 |
+
"loss": 0.3648,
|
| 1845 |
+
"step": 2600
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 2.643037974683544,
|
| 1849 |
+
"grad_norm": 1.825449432151336,
|
| 1850 |
+
"learning_rate": 5e-06,
|
| 1851 |
+
"loss": 0.3675,
|
| 1852 |
+
"step": 2610
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 2.6531645569620252,
|
| 1856 |
+
"grad_norm": 1.6048797855449353,
|
| 1857 |
+
"learning_rate": 5e-06,
|
| 1858 |
+
"loss": 0.3638,
|
| 1859 |
+
"step": 2620
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 2.6632911392405063,
|
| 1863 |
+
"grad_norm": 1.6207591062208428,
|
| 1864 |
+
"learning_rate": 5e-06,
|
| 1865 |
+
"loss": 0.3683,
|
| 1866 |
+
"step": 2630
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 2.6734177215189874,
|
| 1870 |
+
"grad_norm": 1.7272301549499494,
|
| 1871 |
+
"learning_rate": 5e-06,
|
| 1872 |
+
"loss": 0.3685,
|
| 1873 |
+
"step": 2640
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 2.6835443037974684,
|
| 1877 |
+
"grad_norm": 1.6364841258418936,
|
| 1878 |
+
"learning_rate": 5e-06,
|
| 1879 |
+
"loss": 0.3684,
|
| 1880 |
+
"step": 2650
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.6936708860759495,
|
| 1884 |
+
"grad_norm": 1.6937643730038003,
|
| 1885 |
+
"learning_rate": 5e-06,
|
| 1886 |
+
"loss": 0.3723,
|
| 1887 |
+
"step": 2660
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 2.7037974683544306,
|
| 1891 |
+
"grad_norm": 1.5774730979594618,
|
| 1892 |
+
"learning_rate": 5e-06,
|
| 1893 |
+
"loss": 0.3671,
|
| 1894 |
+
"step": 2670
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 2.7139240506329116,
|
| 1898 |
+
"grad_norm": 1.6768910733845062,
|
| 1899 |
+
"learning_rate": 5e-06,
|
| 1900 |
+
"loss": 0.3715,
|
| 1901 |
+
"step": 2680
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.7240506329113923,
|
| 1905 |
+
"grad_norm": 1.6389744367173145,
|
| 1906 |
+
"learning_rate": 5e-06,
|
| 1907 |
+
"loss": 0.3674,
|
| 1908 |
+
"step": 2690
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 2.7341772151898733,
|
| 1912 |
+
"grad_norm": 1.7387883655123013,
|
| 1913 |
+
"learning_rate": 5e-06,
|
| 1914 |
+
"loss": 0.3701,
|
| 1915 |
+
"step": 2700
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 2.7443037974683544,
|
| 1919 |
+
"grad_norm": 1.622855925296899,
|
| 1920 |
+
"learning_rate": 5e-06,
|
| 1921 |
+
"loss": 0.3692,
|
| 1922 |
+
"step": 2710
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 2.7544303797468355,
|
| 1926 |
+
"grad_norm": 1.6049744247340423,
|
| 1927 |
+
"learning_rate": 5e-06,
|
| 1928 |
+
"loss": 0.3605,
|
| 1929 |
+
"step": 2720
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.7645569620253165,
|
| 1933 |
+
"grad_norm": 1.674299981616671,
|
| 1934 |
+
"learning_rate": 5e-06,
|
| 1935 |
+
"loss": 0.3681,
|
| 1936 |
+
"step": 2730
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.7746835443037976,
|
| 1940 |
+
"grad_norm": 1.6563350827590924,
|
| 1941 |
+
"learning_rate": 5e-06,
|
| 1942 |
+
"loss": 0.3732,
|
| 1943 |
+
"step": 2740
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 2.7848101265822782,
|
| 1947 |
+
"grad_norm": 1.6562436415268564,
|
| 1948 |
+
"learning_rate": 5e-06,
|
| 1949 |
+
"loss": 0.3666,
|
| 1950 |
+
"step": 2750
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 2.7949367088607593,
|
| 1954 |
+
"grad_norm": 1.5565903819361853,
|
| 1955 |
+
"learning_rate": 5e-06,
|
| 1956 |
+
"loss": 0.3706,
|
| 1957 |
+
"step": 2760
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 2.8050632911392404,
|
| 1961 |
+
"grad_norm": 1.7142515863322454,
|
| 1962 |
+
"learning_rate": 5e-06,
|
| 1963 |
+
"loss": 0.3679,
|
| 1964 |
+
"step": 2770
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 2.8151898734177214,
|
| 1968 |
+
"grad_norm": 1.5712425643932983,
|
| 1969 |
+
"learning_rate": 5e-06,
|
| 1970 |
+
"loss": 0.3761,
|
| 1971 |
+
"step": 2780
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 2.8253164556962025,
|
| 1975 |
+
"grad_norm": 1.597329501161643,
|
| 1976 |
+
"learning_rate": 5e-06,
|
| 1977 |
+
"loss": 0.3697,
|
| 1978 |
+
"step": 2790
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.8354430379746836,
|
| 1982 |
+
"grad_norm": 1.6642023307988005,
|
| 1983 |
+
"learning_rate": 5e-06,
|
| 1984 |
+
"loss": 0.3613,
|
| 1985 |
+
"step": 2800
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.8455696202531646,
|
| 1989 |
+
"grad_norm": 1.6302362931142675,
|
| 1990 |
+
"learning_rate": 5e-06,
|
| 1991 |
+
"loss": 0.3773,
|
| 1992 |
+
"step": 2810
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.8556962025316457,
|
| 1996 |
+
"grad_norm": 1.594346958058013,
|
| 1997 |
+
"learning_rate": 5e-06,
|
| 1998 |
+
"loss": 0.3699,
|
| 1999 |
+
"step": 2820
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 2.8658227848101268,
|
| 2003 |
+
"grad_norm": 1.5803480355180004,
|
| 2004 |
+
"learning_rate": 5e-06,
|
| 2005 |
+
"loss": 0.3711,
|
| 2006 |
+
"step": 2830
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 2.875949367088608,
|
| 2010 |
+
"grad_norm": 1.4800494871531567,
|
| 2011 |
+
"learning_rate": 5e-06,
|
| 2012 |
+
"loss": 0.3691,
|
| 2013 |
+
"step": 2840
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 2.8860759493670884,
|
| 2017 |
+
"grad_norm": 1.6631915526393122,
|
| 2018 |
+
"learning_rate": 5e-06,
|
| 2019 |
+
"loss": 0.3747,
|
| 2020 |
+
"step": 2850
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 2.8962025316455695,
|
| 2024 |
+
"grad_norm": 1.630862108030936,
|
| 2025 |
+
"learning_rate": 5e-06,
|
| 2026 |
+
"loss": 0.3674,
|
| 2027 |
+
"step": 2860
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.9063291139240506,
|
| 2031 |
+
"grad_norm": 1.6893754841333202,
|
| 2032 |
+
"learning_rate": 5e-06,
|
| 2033 |
+
"loss": 0.373,
|
| 2034 |
+
"step": 2870
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 2.9164556962025316,
|
| 2038 |
+
"grad_norm": 1.705060151220597,
|
| 2039 |
+
"learning_rate": 5e-06,
|
| 2040 |
+
"loss": 0.3723,
|
| 2041 |
+
"step": 2880
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 2.9265822784810127,
|
| 2045 |
+
"grad_norm": 1.6188859279801948,
|
| 2046 |
+
"learning_rate": 5e-06,
|
| 2047 |
+
"loss": 0.3796,
|
| 2048 |
+
"step": 2890
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 2.9367088607594938,
|
| 2052 |
+
"grad_norm": 1.629353248790061,
|
| 2053 |
+
"learning_rate": 5e-06,
|
| 2054 |
+
"loss": 0.3828,
|
| 2055 |
+
"step": 2900
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.946835443037975,
|
| 2059 |
+
"grad_norm": 1.7509445495897462,
|
| 2060 |
+
"learning_rate": 5e-06,
|
| 2061 |
+
"loss": 0.3713,
|
| 2062 |
+
"step": 2910
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 2.9569620253164555,
|
| 2066 |
+
"grad_norm": 1.6650857655447995,
|
| 2067 |
+
"learning_rate": 5e-06,
|
| 2068 |
+
"loss": 0.369,
|
| 2069 |
+
"step": 2920
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.9670886075949365,
|
| 2073 |
+
"grad_norm": 1.8285248313648312,
|
| 2074 |
+
"learning_rate": 5e-06,
|
| 2075 |
+
"loss": 0.3721,
|
| 2076 |
+
"step": 2930
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 2.9772151898734176,
|
| 2080 |
+
"grad_norm": 1.6243458834916522,
|
| 2081 |
+
"learning_rate": 5e-06,
|
| 2082 |
+
"loss": 0.3759,
|
| 2083 |
+
"step": 2940
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 2.9873417721518987,
|
| 2087 |
+
"grad_norm": 1.6911778320262214,
|
| 2088 |
+
"learning_rate": 5e-06,
|
| 2089 |
+
"loss": 0.3746,
|
| 2090 |
+
"step": 2950
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.9974683544303797,
|
| 2094 |
+
"grad_norm": 1.7509767039817499,
|
| 2095 |
+
"learning_rate": 5e-06,
|
| 2096 |
+
"loss": 0.3711,
|
| 2097 |
+
"step": 2960
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 2.9984810126582278,
|
| 2101 |
+
"eval_loss": 0.15120381116867065,
|
| 2102 |
+
"eval_runtime": 508.8119,
|
| 2103 |
+
"eval_samples_per_second": 26.147,
|
| 2104 |
+
"eval_steps_per_second": 0.409,
|
| 2105 |
+
"step": 2961
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 2.9984810126582278,
|
| 2109 |
+
"step": 2961,
|
| 2110 |
+
"total_flos": 2479683262218240.0,
|
| 2111 |
+
"train_loss": 0.46617310421968466,
|
| 2112 |
+
"train_runtime": 84499.4991,
|
| 2113 |
+
"train_samples_per_second": 8.974,
|
| 2114 |
+
"train_steps_per_second": 0.035
|
| 2115 |
+
}
|
| 2116 |
+
],
|
| 2117 |
+
"logging_steps": 10,
|
| 2118 |
+
"max_steps": 2961,
|
| 2119 |
+
"num_input_tokens_seen": 0,
|
| 2120 |
+
"num_train_epochs": 3,
|
| 2121 |
+
"save_steps": 500,
|
| 2122 |
+
"stateful_callbacks": {
|
| 2123 |
+
"TrainerControl": {
|
| 2124 |
+
"args": {
|
| 2125 |
+
"should_epoch_stop": false,
|
| 2126 |
+
"should_evaluate": false,
|
| 2127 |
+
"should_log": false,
|
| 2128 |
+
"should_save": true,
|
| 2129 |
+
"should_training_stop": true
|
| 2130 |
+
},
|
| 2131 |
+
"attributes": {}
|
| 2132 |
+
}
|
| 2133 |
+
},
|
| 2134 |
+
"total_flos": 2479683262218240.0,
|
| 2135 |
+
"train_batch_size": 8,
|
| 2136 |
+
"trial_name": null,
|
| 2137 |
+
"trial_params": null
|
| 2138 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|