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 +1109 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: mistral_7b_0-3_oh-dcft-v3.1-gemini-1.5-flash
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# mistral_7b_0-3_oh-dcft-v3.1-gemini-1.5-flash
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.5188
|
| 21 |
|
|
|
|
| 4 |
base_model: mistralai/Mistral-7B-v0.3
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: mistral_7b_0-3_oh-dcft-v3.1-gemini-1.5-flash
|
|
|
|
| 16 |
|
| 17 |
# mistral_7b_0-3_oh-dcft-v3.1-gemini-1.5-flash
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the mlfoundations-dev/oh-dcft-v3.1-gemini-1.5-flash dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.5188
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5188336968421936,
|
| 4 |
+
"eval_runtime": 51.9694,
|
| 5 |
+
"eval_samples_per_second": 258.421,
|
| 6 |
+
"eval_steps_per_second": 1.02,
|
| 7 |
+
"total_flos": 2507530756423680.0,
|
| 8 |
+
"train_loss": 0.4253670667598625,
|
| 9 |
+
"train_runtime": 9374.4269,
|
| 10 |
+
"train_samples_per_second": 81.656,
|
| 11 |
+
"train_steps_per_second": 0.16
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.5188336968421936,
|
| 4 |
+
"eval_runtime": 51.9694,
|
| 5 |
+
"eval_samples_per_second": 258.421,
|
| 6 |
+
"eval_steps_per_second": 1.02
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2507530756423680.0,
|
| 4 |
+
"train_loss": 0.4253670667598625,
|
| 5 |
+
"train_runtime": 9374.4269,
|
| 6 |
+
"train_samples_per_second": 81.656,
|
| 7 |
+
"train_steps_per_second": 0.16
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1497,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02004008016032064,
|
| 13 |
+
"grad_norm": 11.386945789449237,
|
| 14 |
+
"learning_rate": 6.666666666666667e-07,
|
| 15 |
+
"loss": 0.914,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04008016032064128,
|
| 20 |
+
"grad_norm": 3.686249022148282,
|
| 21 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 22 |
+
"loss": 0.7761,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06012024048096192,
|
| 27 |
+
"grad_norm": 1.797608399405574,
|
| 28 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 29 |
+
"loss": 0.6765,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.08016032064128256,
|
| 34 |
+
"grad_norm": 2.53598544592724,
|
| 35 |
+
"learning_rate": 2.666666666666667e-06,
|
| 36 |
+
"loss": 0.6319,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10020040080160321,
|
| 41 |
+
"grad_norm": 1.8597121806159485,
|
| 42 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 43 |
+
"loss": 0.6028,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12024048096192384,
|
| 48 |
+
"grad_norm": 1.8378706472462616,
|
| 49 |
+
"learning_rate": 4.000000000000001e-06,
|
| 50 |
+
"loss": 0.5846,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1402805611222445,
|
| 55 |
+
"grad_norm": 4.6666512736387356,
|
| 56 |
+
"learning_rate": 4.666666666666667e-06,
|
| 57 |
+
"loss": 0.5761,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.16032064128256512,
|
| 62 |
+
"grad_norm": 2.67064642378821,
|
| 63 |
+
"learning_rate": 4.999862725880242e-06,
|
| 64 |
+
"loss": 0.5758,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18036072144288579,
|
| 69 |
+
"grad_norm": 2.6004134353948736,
|
| 70 |
+
"learning_rate": 4.998764633422446e-06,
|
| 71 |
+
"loss": 0.5667,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.20040080160320642,
|
| 76 |
+
"grad_norm": 2.0205968573263413,
|
| 77 |
+
"learning_rate": 4.996568984453794e-06,
|
| 78 |
+
"loss": 0.5651,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.22044088176352705,
|
| 83 |
+
"grad_norm": 1.7372715757375135,
|
| 84 |
+
"learning_rate": 4.993276850606589e-06,
|
| 85 |
+
"loss": 0.5593,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.24048096192384769,
|
| 90 |
+
"grad_norm": 2.3534996172184846,
|
| 91 |
+
"learning_rate": 4.988889838675462e-06,
|
| 92 |
+
"loss": 0.5566,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2605210420841683,
|
| 97 |
+
"grad_norm": 2.1712051144800846,
|
| 98 |
+
"learning_rate": 4.9834100898331425e-06,
|
| 99 |
+
"loss": 0.554,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.280561122244489,
|
| 104 |
+
"grad_norm": 2.988740945183358,
|
| 105 |
+
"learning_rate": 4.976840278585413e-06,
|
| 106 |
+
"loss": 0.5412,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.30060120240480964,
|
| 111 |
+
"grad_norm": 1.7513590918597028,
|
| 112 |
+
"learning_rate": 4.969183611465766e-06,
|
| 113 |
+
"loss": 0.5494,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.32064128256513025,
|
| 118 |
+
"grad_norm": 1.911642443149266,
|
| 119 |
+
"learning_rate": 4.9604438254703845e-06,
|
| 120 |
+
"loss": 0.5342,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3406813627254509,
|
| 125 |
+
"grad_norm": 1.7028624872037597,
|
| 126 |
+
"learning_rate": 4.95062518623422e-06,
|
| 127 |
+
"loss": 0.5427,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.36072144288577157,
|
| 132 |
+
"grad_norm": 1.89229572623421,
|
| 133 |
+
"learning_rate": 4.939732485949066e-06,
|
| 134 |
+
"loss": 0.5379,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3807615230460922,
|
| 139 |
+
"grad_norm": 1.458083413953667,
|
| 140 |
+
"learning_rate": 4.927771041024627e-06,
|
| 141 |
+
"loss": 0.5355,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.40080160320641284,
|
| 146 |
+
"grad_norm": 2.1624686975408336,
|
| 147 |
+
"learning_rate": 4.914746689493731e-06,
|
| 148 |
+
"loss": 0.54,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.42084168336673344,
|
| 153 |
+
"grad_norm": 1.841839502355017,
|
| 154 |
+
"learning_rate": 4.900665788162962e-06,
|
| 155 |
+
"loss": 0.5306,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4408817635270541,
|
| 160 |
+
"grad_norm": 1.5623403972638494,
|
| 161 |
+
"learning_rate": 4.885535209510081e-06,
|
| 162 |
+
"loss": 0.531,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.46092184368737477,
|
| 167 |
+
"grad_norm": 2.0959167807238672,
|
| 168 |
+
"learning_rate": 4.869362338329773e-06,
|
| 169 |
+
"loss": 0.528,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.48096192384769537,
|
| 174 |
+
"grad_norm": 2.877452797335848,
|
| 175 |
+
"learning_rate": 4.852155068129343e-06,
|
| 176 |
+
"loss": 0.53,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.501002004008016,
|
| 181 |
+
"grad_norm": 2.406716714855783,
|
| 182 |
+
"learning_rate": 4.8339217972761274e-06,
|
| 183 |
+
"loss": 0.5242,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5210420841683366,
|
| 188 |
+
"grad_norm": 2.6180161170191956,
|
| 189 |
+
"learning_rate": 4.814671424898493e-06,
|
| 190 |
+
"loss": 0.5218,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5410821643286573,
|
| 195 |
+
"grad_norm": 1.9325754069366325,
|
| 196 |
+
"learning_rate": 4.794413346542432e-06,
|
| 197 |
+
"loss": 0.516,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.561122244488978,
|
| 202 |
+
"grad_norm": 1.6202018849080733,
|
| 203 |
+
"learning_rate": 4.773157449585871e-06,
|
| 204 |
+
"loss": 0.519,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5811623246492986,
|
| 209 |
+
"grad_norm": 2.5386712058137846,
|
| 210 |
+
"learning_rate": 4.750914108412927e-06,
|
| 211 |
+
"loss": 0.5168,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6012024048096193,
|
| 216 |
+
"grad_norm": 1.6204343506402092,
|
| 217 |
+
"learning_rate": 4.727694179350476e-06,
|
| 218 |
+
"loss": 0.5125,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6212424849699398,
|
| 223 |
+
"grad_norm": 1.6404070884775066,
|
| 224 |
+
"learning_rate": 4.703508995369497e-06,
|
| 225 |
+
"loss": 0.513,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6412825651302605,
|
| 230 |
+
"grad_norm": 1.6188579894187152,
|
| 231 |
+
"learning_rate": 4.6783703605537715e-06,
|
| 232 |
+
"loss": 0.5147,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6613226452905812,
|
| 237 |
+
"grad_norm": 1.4143375173363102,
|
| 238 |
+
"learning_rate": 4.652290544338663e-06,
|
| 239 |
+
"loss": 0.5173,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6813627254509018,
|
| 244 |
+
"grad_norm": 2.468479009235932,
|
| 245 |
+
"learning_rate": 4.6252822755227605e-06,
|
| 246 |
+
"loss": 0.5108,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7014028056112225,
|
| 251 |
+
"grad_norm": 2.1271900619659854,
|
| 252 |
+
"learning_rate": 4.5973587360553185e-06,
|
| 253 |
+
"loss": 0.5131,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7214428857715431,
|
| 258 |
+
"grad_norm": 1.5534652608564319,
|
| 259 |
+
"learning_rate": 4.568533554602535e-06,
|
| 260 |
+
"loss": 0.5083,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7414829659318637,
|
| 265 |
+
"grad_norm": 1.6027013883759695,
|
| 266 |
+
"learning_rate": 4.5388207998957986e-06,
|
| 267 |
+
"loss": 0.5133,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7615230460921844,
|
| 272 |
+
"grad_norm": 1.530068712838613,
|
| 273 |
+
"learning_rate": 4.508234973865145e-06,
|
| 274 |
+
"loss": 0.5157,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.781563126252505,
|
| 279 |
+
"grad_norm": 1.8195363489136889,
|
| 280 |
+
"learning_rate": 4.476791004561293e-06,
|
| 281 |
+
"loss": 0.512,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8016032064128257,
|
| 286 |
+
"grad_norm": 1.637599921361452,
|
| 287 |
+
"learning_rate": 4.4445042388696935e-06,
|
| 288 |
+
"loss": 0.5095,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8216432865731463,
|
| 293 |
+
"grad_norm": 1.5903329614208361,
|
| 294 |
+
"learning_rate": 4.411390435020164e-06,
|
| 295 |
+
"loss": 0.5131,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8416833667334669,
|
| 300 |
+
"grad_norm": 1.5488153255590384,
|
| 301 |
+
"learning_rate": 4.377465754895757e-06,
|
| 302 |
+
"loss": 0.5052,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8617234468937875,
|
| 307 |
+
"grad_norm": 1.8543008230812608,
|
| 308 |
+
"learning_rate": 4.342746756144609e-06,
|
| 309 |
+
"loss": 0.5124,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8817635270541082,
|
| 314 |
+
"grad_norm": 1.5858389816681295,
|
| 315 |
+
"learning_rate": 4.307250384098645e-06,
|
| 316 |
+
"loss": 0.5094,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9018036072144289,
|
| 321 |
+
"grad_norm": 1.6693267301390629,
|
| 322 |
+
"learning_rate": 4.270993963503048e-06,
|
| 323 |
+
"loss": 0.5055,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9218436873747495,
|
| 328 |
+
"grad_norm": 1.4123291299839993,
|
| 329 |
+
"learning_rate": 4.233995190060559e-06,
|
| 330 |
+
"loss": 0.511,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9418837675350702,
|
| 335 |
+
"grad_norm": 1.4294865023874812,
|
| 336 |
+
"learning_rate": 4.196272121794714e-06,
|
| 337 |
+
"loss": 0.5122,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.9619238476953907,
|
| 342 |
+
"grad_norm": 1.4747480031112472,
|
| 343 |
+
"learning_rate": 4.15784317023626e-06,
|
| 344 |
+
"loss": 0.5068,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.9819639278557114,
|
| 349 |
+
"grad_norm": 1.6382019572026507,
|
| 350 |
+
"learning_rate": 4.118727091437013e-06,
|
| 351 |
+
"loss": 0.5002,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.0,
|
| 356 |
+
"eval_loss": 0.5069113373756409,
|
| 357 |
+
"eval_runtime": 45.884,
|
| 358 |
+
"eval_samples_per_second": 292.695,
|
| 359 |
+
"eval_steps_per_second": 1.155,
|
| 360 |
+
"step": 499
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.002004008016032,
|
| 364 |
+
"grad_norm": 3.2862336764098803,
|
| 365 |
+
"learning_rate": 4.078942976815581e-06,
|
| 366 |
+
"loss": 0.4942,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.0220440881763526,
|
| 371 |
+
"grad_norm": 1.7759252107171999,
|
| 372 |
+
"learning_rate": 4.038510243839396e-06,
|
| 373 |
+
"loss": 0.418,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.0420841683366733,
|
| 378 |
+
"grad_norm": 1.7448507912514557,
|
| 379 |
+
"learning_rate": 3.997448626547613e-06,
|
| 380 |
+
"loss": 0.4136,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.062124248496994,
|
| 385 |
+
"grad_norm": 1.4081466868123829,
|
| 386 |
+
"learning_rate": 3.955778165919502e-06,
|
| 387 |
+
"loss": 0.4131,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.0821643286573146,
|
| 392 |
+
"grad_norm": 1.4737485149847567,
|
| 393 |
+
"learning_rate": 3.913519200093035e-06,
|
| 394 |
+
"loss": 0.412,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.1022044088176353,
|
| 399 |
+
"grad_norm": 1.2635878378102765,
|
| 400 |
+
"learning_rate": 3.870692354438423e-06,
|
| 401 |
+
"loss": 0.4142,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.122244488977956,
|
| 406 |
+
"grad_norm": 1.553384214976449,
|
| 407 |
+
"learning_rate": 3.827318531491478e-06,
|
| 408 |
+
"loss": 0.4115,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.1422845691382766,
|
| 413 |
+
"grad_norm": 1.5870294004406946,
|
| 414 |
+
"learning_rate": 3.7834189007517002e-06,
|
| 415 |
+
"loss": 0.4108,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.1623246492985972,
|
| 420 |
+
"grad_norm": 1.6029534467749516,
|
| 421 |
+
"learning_rate": 3.739014888350053e-06,
|
| 422 |
+
"loss": 0.4116,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.182364729458918,
|
| 427 |
+
"grad_norm": 1.6194120236052227,
|
| 428 |
+
"learning_rate": 3.694128166591494e-06,
|
| 429 |
+
"loss": 0.4136,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.2024048096192386,
|
| 434 |
+
"grad_norm": 1.436406692065181,
|
| 435 |
+
"learning_rate": 3.6487806433773615e-06,
|
| 436 |
+
"loss": 0.4134,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.2224448897795592,
|
| 441 |
+
"grad_norm": 1.4163845777049167,
|
| 442 |
+
"learning_rate": 3.6029944515127585e-06,
|
| 443 |
+
"loss": 0.4091,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.2424849699398797,
|
| 448 |
+
"grad_norm": 1.5714505381282495,
|
| 449 |
+
"learning_rate": 3.5567919379041783e-06,
|
| 450 |
+
"loss": 0.4142,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.2625250501002003,
|
| 455 |
+
"grad_norm": 1.4088322992815865,
|
| 456 |
+
"learning_rate": 3.510195652652629e-06,
|
| 457 |
+
"loss": 0.4161,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.282565130260521,
|
| 462 |
+
"grad_norm": 1.6997351047753615,
|
| 463 |
+
"learning_rate": 3.463228338047589e-06,
|
| 464 |
+
"loss": 0.4059,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.3026052104208417,
|
| 469 |
+
"grad_norm": 1.5683685074334606,
|
| 470 |
+
"learning_rate": 3.4159129174671534e-06,
|
| 471 |
+
"loss": 0.4065,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.3226452905811623,
|
| 476 |
+
"grad_norm": 1.6253512472033467,
|
| 477 |
+
"learning_rate": 3.3682724841898067e-06,
|
| 478 |
+
"loss": 0.4108,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 1.342685370741483,
|
| 483 |
+
"grad_norm": 1.759831168192353,
|
| 484 |
+
"learning_rate": 3.320330290123261e-06,
|
| 485 |
+
"loss": 0.4103,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 1.3627254509018036,
|
| 490 |
+
"grad_norm": 1.7378970495344055,
|
| 491 |
+
"learning_rate": 3.2721097344558794e-06,
|
| 492 |
+
"loss": 0.4133,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.3827655310621243,
|
| 497 |
+
"grad_norm": 1.5427360104428394,
|
| 498 |
+
"learning_rate": 3.223634352236213e-06,
|
| 499 |
+
"loss": 0.4103,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.402805611222445,
|
| 504 |
+
"grad_norm": 1.4316640042632263,
|
| 505 |
+
"learning_rate": 3.1749278028862325e-06,
|
| 506 |
+
"loss": 0.4078,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.4228456913827654,
|
| 511 |
+
"grad_norm": 1.951831574889627,
|
| 512 |
+
"learning_rate": 3.1260138586538483e-06,
|
| 513 |
+
"loss": 0.4108,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.4428857715430863,
|
| 518 |
+
"grad_norm": 1.5703450558182246,
|
| 519 |
+
"learning_rate": 3.076916393010373e-06,
|
| 520 |
+
"loss": 0.4091,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.4629258517034067,
|
| 525 |
+
"grad_norm": 1.4374885381526987,
|
| 526 |
+
"learning_rate": 3.0276593689985737e-06,
|
| 527 |
+
"loss": 0.4056,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.4829659318637274,
|
| 532 |
+
"grad_norm": 1.2643689737278019,
|
| 533 |
+
"learning_rate": 2.9782668275370095e-06,
|
| 534 |
+
"loss": 0.4139,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.503006012024048,
|
| 539 |
+
"grad_norm": 1.570266686325652,
|
| 540 |
+
"learning_rate": 2.928762875686358e-06,
|
| 541 |
+
"loss": 0.4109,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.5230460921843687,
|
| 546 |
+
"grad_norm": 1.5743968516846465,
|
| 547 |
+
"learning_rate": 2.879171674883462e-06,
|
| 548 |
+
"loss": 0.4103,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.5430861723446894,
|
| 553 |
+
"grad_norm": 1.5844459981668437,
|
| 554 |
+
"learning_rate": 2.8295174291488383e-06,
|
| 555 |
+
"loss": 0.4109,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.56312625250501,
|
| 560 |
+
"grad_norm": 1.504354777149479,
|
| 561 |
+
"learning_rate": 2.779824373273397e-06,
|
| 562 |
+
"loss": 0.4118,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.5831663326653307,
|
| 567 |
+
"grad_norm": 1.3787147420821566,
|
| 568 |
+
"learning_rate": 2.7301167609901474e-06,
|
| 569 |
+
"loss": 0.4135,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.6032064128256514,
|
| 574 |
+
"grad_norm": 1.4777711687133863,
|
| 575 |
+
"learning_rate": 2.680418853136659e-06,
|
| 576 |
+
"loss": 0.4095,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.623246492985972,
|
| 581 |
+
"grad_norm": 1.468500779993122,
|
| 582 |
+
"learning_rate": 2.630754905814048e-06,
|
| 583 |
+
"loss": 0.4079,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.6432865731462925,
|
| 588 |
+
"grad_norm": 1.6134930461454124,
|
| 589 |
+
"learning_rate": 2.58114915854829e-06,
|
| 590 |
+
"loss": 0.4083,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.6633266533066133,
|
| 595 |
+
"grad_norm": 1.4588892173060275,
|
| 596 |
+
"learning_rate": 2.531625822459603e-06,
|
| 597 |
+
"loss": 0.4087,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.6833667334669338,
|
| 602 |
+
"grad_norm": 1.4497521826737287,
|
| 603 |
+
"learning_rate": 2.4822090684457147e-06,
|
| 604 |
+
"loss": 0.4113,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.7034068136272547,
|
| 609 |
+
"grad_norm": 1.4372338693007887,
|
| 610 |
+
"learning_rate": 2.432923015384743e-06,
|
| 611 |
+
"loss": 0.4047,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.723446893787575,
|
| 616 |
+
"grad_norm": 1.3561893796274846,
|
| 617 |
+
"learning_rate": 2.3837917183634815e-06,
|
| 618 |
+
"loss": 0.4085,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.7434869739478958,
|
| 623 |
+
"grad_norm": 1.3339197747345177,
|
| 624 |
+
"learning_rate": 2.3348391569368064e-06,
|
| 625 |
+
"loss": 0.411,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.7635270541082164,
|
| 630 |
+
"grad_norm": 1.2878261372500692,
|
| 631 |
+
"learning_rate": 2.2860892234239565e-06,
|
| 632 |
+
"loss": 0.4145,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.783567134268537,
|
| 637 |
+
"grad_norm": 1.3852796037018127,
|
| 638 |
+
"learning_rate": 2.237565711247391e-06,
|
| 639 |
+
"loss": 0.4109,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.8036072144288577,
|
| 644 |
+
"grad_norm": 1.343576505451422,
|
| 645 |
+
"learning_rate": 2.1892923033199075e-06,
|
| 646 |
+
"loss": 0.4078,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.8236472945891784,
|
| 651 |
+
"grad_norm": 1.4910844256604878,
|
| 652 |
+
"learning_rate": 2.141292560485708e-06,
|
| 653 |
+
"loss": 0.4064,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.843687374749499,
|
| 658 |
+
"grad_norm": 1.452632980241925,
|
| 659 |
+
"learning_rate": 2.0935899100210316e-06,
|
| 660 |
+
"loss": 0.4059,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.8637274549098195,
|
| 665 |
+
"grad_norm": 1.3416125112736146,
|
| 666 |
+
"learning_rate": 2.046207634199989e-06,
|
| 667 |
+
"loss": 0.405,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.8837675350701404,
|
| 672 |
+
"grad_norm": 1.3346098098736352,
|
| 673 |
+
"learning_rate": 1.9991688589311575e-06,
|
| 674 |
+
"loss": 0.4067,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.9038076152304608,
|
| 679 |
+
"grad_norm": 1.2974237604973982,
|
| 680 |
+
"learning_rate": 1.9524965424705026e-06,
|
| 681 |
+
"loss": 0.4024,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.9238476953907817,
|
| 686 |
+
"grad_norm": 1.471473412639078,
|
| 687 |
+
"learning_rate": 1.9062134642161198e-06,
|
| 688 |
+
"loss": 0.4091,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.9438877755511021,
|
| 693 |
+
"grad_norm": 1.357694886526054,
|
| 694 |
+
"learning_rate": 1.8603422135902737e-06,
|
| 695 |
+
"loss": 0.4057,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.9639278557114228,
|
| 700 |
+
"grad_norm": 1.2865807531711169,
|
| 701 |
+
"learning_rate": 1.8149051790141628e-06,
|
| 702 |
+
"loss": 0.4066,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.9839679358717435,
|
| 707 |
+
"grad_norm": 1.297389613332899,
|
| 708 |
+
"learning_rate": 1.7699245369807778e-06,
|
| 709 |
+
"loss": 0.4045,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.0,
|
| 714 |
+
"eval_loss": 0.4974905848503113,
|
| 715 |
+
"eval_runtime": 51.1615,
|
| 716 |
+
"eval_samples_per_second": 262.502,
|
| 717 |
+
"eval_steps_per_second": 1.036,
|
| 718 |
+
"step": 998
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.004008016032064,
|
| 722 |
+
"grad_norm": 2.171277446089175,
|
| 723 |
+
"learning_rate": 1.725422241231205e-06,
|
| 724 |
+
"loss": 0.3859,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.024048096192385,
|
| 729 |
+
"grad_norm": 1.7817202202370321,
|
| 730 |
+
"learning_rate": 1.6814200120396438e-06,
|
| 731 |
+
"loss": 0.3244,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.0440881763527052,
|
| 736 |
+
"grad_norm": 1.619953212777582,
|
| 737 |
+
"learning_rate": 1.6379393256123737e-06,
|
| 738 |
+
"loss": 0.3211,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 2.064128256513026,
|
| 743 |
+
"grad_norm": 1.40219382175259,
|
| 744 |
+
"learning_rate": 1.595001403605844e-06,
|
| 745 |
+
"loss": 0.3158,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 2.0841683366733466,
|
| 750 |
+
"grad_norm": 1.4533085140998225,
|
| 751 |
+
"learning_rate": 1.552627202769e-06,
|
| 752 |
+
"loss": 0.318,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.1042084168336674,
|
| 757 |
+
"grad_norm": 1.4269652142649647,
|
| 758 |
+
"learning_rate": 1.5108374047149061e-06,
|
| 759 |
+
"loss": 0.3197,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 2.124248496993988,
|
| 764 |
+
"grad_norm": 1.537954200218747,
|
| 765 |
+
"learning_rate": 1.4696524058266516e-06,
|
| 766 |
+
"loss": 0.3196,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 2.1442885771543088,
|
| 771 |
+
"grad_norm": 1.6449193889553193,
|
| 772 |
+
"learning_rate": 1.4290923073024712e-06,
|
| 773 |
+
"loss": 0.3214,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.164328657314629,
|
| 778 |
+
"grad_norm": 1.3598251114998061,
|
| 779 |
+
"learning_rate": 1.3891769053449355e-06,
|
| 780 |
+
"loss": 0.321,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.18436873747495,
|
| 785 |
+
"grad_norm": 1.551617792162964,
|
| 786 |
+
"learning_rate": 1.3499256814990003e-06,
|
| 787 |
+
"loss": 0.3172,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 2.2044088176352705,
|
| 792 |
+
"grad_norm": 1.487413687096442,
|
| 793 |
+
"learning_rate": 1.3113577931436332e-06,
|
| 794 |
+
"loss": 0.319,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.2244488977955914,
|
| 799 |
+
"grad_norm": 1.379093893132986,
|
| 800 |
+
"learning_rate": 1.2734920641416556e-06,
|
| 801 |
+
"loss": 0.3201,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 2.244488977955912,
|
| 806 |
+
"grad_norm": 1.3525463916397538,
|
| 807 |
+
"learning_rate": 1.236346975652358e-06,
|
| 808 |
+
"loss": 0.32,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 2.2645290581162323,
|
| 813 |
+
"grad_norm": 1.4572131883779704,
|
| 814 |
+
"learning_rate": 1.1999406571113962e-06,
|
| 815 |
+
"loss": 0.3171,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.284569138276553,
|
| 820 |
+
"grad_norm": 1.4042689852309367,
|
| 821 |
+
"learning_rate": 1.1642908773823274e-06,
|
| 822 |
+
"loss": 0.3166,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 2.3046092184368736,
|
| 827 |
+
"grad_norm": 1.3000414229131876,
|
| 828 |
+
"learning_rate": 1.1294150360841577e-06,
|
| 829 |
+
"loss": 0.3211,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.3246492985971945,
|
| 834 |
+
"grad_norm": 1.3811842558284932,
|
| 835 |
+
"learning_rate": 1.0953301550990882e-06,
|
| 836 |
+
"loss": 0.3213,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.344689378757515,
|
| 841 |
+
"grad_norm": 1.3259540554288098,
|
| 842 |
+
"learning_rate": 1.0620528702646312e-06,
|
| 843 |
+
"loss": 0.3193,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 2.364729458917836,
|
| 848 |
+
"grad_norm": 1.4720294318649223,
|
| 849 |
+
"learning_rate": 1.0295994232541454e-06,
|
| 850 |
+
"loss": 0.32,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 2.3847695390781563,
|
| 855 |
+
"grad_norm": 1.3734699044137106,
|
| 856 |
+
"learning_rate": 9.979856536497435e-07,
|
| 857 |
+
"loss": 0.3178,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 2.404809619238477,
|
| 862 |
+
"grad_norm": 1.3553774328605976,
|
| 863 |
+
"learning_rate": 9.672269912114582e-07,
|
| 864 |
+
"loss": 0.3134,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 2.4248496993987976,
|
| 869 |
+
"grad_norm": 1.350392284981299,
|
| 870 |
+
"learning_rate": 9.373384483464223e-07,
|
| 871 |
+
"loss": 0.3218,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 2.4448897795591185,
|
| 876 |
+
"grad_norm": 1.4402104791397123,
|
| 877 |
+
"learning_rate": 9.08334612781753e-07,
|
| 878 |
+
"loss": 0.3205,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.464929859719439,
|
| 883 |
+
"grad_norm": 1.3646686545789422,
|
| 884 |
+
"learning_rate": 8.80229640444705e-07,
|
| 885 |
+
"loss": 0.3216,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 2.4849699398797593,
|
| 890 |
+
"grad_norm": 1.3499126545275342,
|
| 891 |
+
"learning_rate": 8.530372485535751e-07,
|
| 892 |
+
"loss": 0.3197,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.50501002004008,
|
| 897 |
+
"grad_norm": 1.4013520657650143,
|
| 898 |
+
"learning_rate": 8.267707089227288e-07,
|
| 899 |
+
"loss": 0.3174,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.5250501002004007,
|
| 904 |
+
"grad_norm": 1.3543143703928378,
|
| 905 |
+
"learning_rate": 8.01442841485013e-07,
|
| 906 |
+
"loss": 0.3237,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.5450901803607215,
|
| 911 |
+
"grad_norm": 1.3528190597320051,
|
| 912 |
+
"learning_rate": 7.770660080347213e-07,
|
| 913 |
+
"loss": 0.3226,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.565130260521042,
|
| 918 |
+
"grad_norm": 1.316933143733213,
|
| 919 |
+
"learning_rate": 7.536521061941613e-07,
|
| 920 |
+
"loss": 0.3178,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.585170340681363,
|
| 925 |
+
"grad_norm": 1.2849910299224803,
|
| 926 |
+
"learning_rate": 7.312125636067723e-07,
|
| 927 |
+
"loss": 0.3225,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.6052104208416833,
|
| 932 |
+
"grad_norm": 1.290228394422378,
|
| 933 |
+
"learning_rate": 7.097583323596257e-07,
|
| 934 |
+
"loss": 0.3222,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.6252505010020037,
|
| 939 |
+
"grad_norm": 1.426441845965089,
|
| 940 |
+
"learning_rate": 6.892998836380285e-07,
|
| 941 |
+
"loss": 0.3173,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.6452905811623246,
|
| 946 |
+
"grad_norm": 1.2801979621928115,
|
| 947 |
+
"learning_rate": 6.698472026148459e-07,
|
| 948 |
+
"loss": 0.3234,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.6653306613226455,
|
| 953 |
+
"grad_norm": 1.3220972865520855,
|
| 954 |
+
"learning_rate": 6.514097835770269e-07,
|
| 955 |
+
"loss": 0.3175,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 2.685370741482966,
|
| 960 |
+
"grad_norm": 1.3667409352018853,
|
| 961 |
+
"learning_rate": 6.3399662529172e-07,
|
| 962 |
+
"loss": 0.3184,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.7054108216432864,
|
| 967 |
+
"grad_norm": 1.3528891456726284,
|
| 968 |
+
"learning_rate": 6.176162266142376e-07,
|
| 969 |
+
"loss": 0.323,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 2.7254509018036073,
|
| 974 |
+
"grad_norm": 1.327321152396037,
|
| 975 |
+
"learning_rate": 6.022765823400116e-07,
|
| 976 |
+
"loss": 0.3215,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.7454909819639277,
|
| 981 |
+
"grad_norm": 1.277409240062218,
|
| 982 |
+
"learning_rate": 5.879851793025669e-07,
|
| 983 |
+
"loss": 0.3182,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.7655310621242486,
|
| 988 |
+
"grad_norm": 1.336386040490389,
|
| 989 |
+
"learning_rate": 5.747489927194165e-07,
|
| 990 |
+
"loss": 0.3202,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.785571142284569,
|
| 995 |
+
"grad_norm": 1.2533448631397004,
|
| 996 |
+
"learning_rate": 5.625744827876601e-07,
|
| 997 |
+
"loss": 0.319,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 2.80561122244489,
|
| 1002 |
+
"grad_norm": 1.290234810317301,
|
| 1003 |
+
"learning_rate": 5.514675915309507e-07,
|
| 1004 |
+
"loss": 0.3185,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 2.8256513026052104,
|
| 1009 |
+
"grad_norm": 1.311881104807796,
|
| 1010 |
+
"learning_rate": 5.414337398993661e-07,
|
| 1011 |
+
"loss": 0.3156,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 2.845691382765531,
|
| 1016 |
+
"grad_norm": 1.3245343587549412,
|
| 1017 |
+
"learning_rate": 5.324778251236008e-07,
|
| 1018 |
+
"loss": 0.3191,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 2.8657314629258517,
|
| 1023 |
+
"grad_norm": 1.318165777680666,
|
| 1024 |
+
"learning_rate": 5.246042183247698e-07,
|
| 1025 |
+
"loss": 0.3141,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 2.8857715430861726,
|
| 1030 |
+
"grad_norm": 1.3833599862766017,
|
| 1031 |
+
"learning_rate": 5.178167623809932e-07,
|
| 1032 |
+
"loss": 0.3199,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 2.905811623246493,
|
| 1037 |
+
"grad_norm": 1.3572877385241544,
|
| 1038 |
+
"learning_rate": 5.121187700517977e-07,
|
| 1039 |
+
"loss": 0.3213,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 2.9258517034068134,
|
| 1044 |
+
"grad_norm": 1.289697605126265,
|
| 1045 |
+
"learning_rate": 5.075130223612563e-07,
|
| 1046 |
+
"loss": 0.3209,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 2.9458917835671343,
|
| 1051 |
+
"grad_norm": 1.3713215738345175,
|
| 1052 |
+
"learning_rate": 5.040017672406508e-07,
|
| 1053 |
+
"loss": 0.3152,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 2.9659318637274548,
|
| 1058 |
+
"grad_norm": 1.3065474715872867,
|
| 1059 |
+
"learning_rate": 5.015867184313224e-07,
|
| 1060 |
+
"loss": 0.3152,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 2.9859719438877756,
|
| 1065 |
+
"grad_norm": 1.29789945819745,
|
| 1066 |
+
"learning_rate": 5.002690546482441e-07,
|
| 1067 |
+
"loss": 0.3184,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 3.0,
|
| 1072 |
+
"eval_loss": 0.5188336968421936,
|
| 1073 |
+
"eval_runtime": 52.0293,
|
| 1074 |
+
"eval_samples_per_second": 258.124,
|
| 1075 |
+
"eval_steps_per_second": 1.019,
|
| 1076 |
+
"step": 1497
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.0,
|
| 1080 |
+
"step": 1497,
|
| 1081 |
+
"total_flos": 2507530756423680.0,
|
| 1082 |
+
"train_loss": 0.4253670667598625,
|
| 1083 |
+
"train_runtime": 9374.4269,
|
| 1084 |
+
"train_samples_per_second": 81.656,
|
| 1085 |
+
"train_steps_per_second": 0.16
|
| 1086 |
+
}
|
| 1087 |
+
],
|
| 1088 |
+
"logging_steps": 10,
|
| 1089 |
+
"max_steps": 1497,
|
| 1090 |
+
"num_input_tokens_seen": 0,
|
| 1091 |
+
"num_train_epochs": 3,
|
| 1092 |
+
"save_steps": 500,
|
| 1093 |
+
"stateful_callbacks": {
|
| 1094 |
+
"TrainerControl": {
|
| 1095 |
+
"args": {
|
| 1096 |
+
"should_epoch_stop": false,
|
| 1097 |
+
"should_evaluate": false,
|
| 1098 |
+
"should_log": false,
|
| 1099 |
+
"should_save": true,
|
| 1100 |
+
"should_training_stop": true
|
| 1101 |
+
},
|
| 1102 |
+
"attributes": {}
|
| 1103 |
+
}
|
| 1104 |
+
},
|
| 1105 |
+
"total_flos": 2507530756423680.0,
|
| 1106 |
+
"train_batch_size": 16,
|
| 1107 |
+
"trial_name": null,
|
| 1108 |
+
"trial_params": null
|
| 1109 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|