End of training
Browse files- README.md +3 -2
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +1530 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Llama-3.1-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: openmathinstruct2-llama-3.1-8B-Instruct-lr5-ep2
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# openmathinstruct2-llama-3.1-8B-Instruct-lr5-ep2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 4 |
base_model: meta-llama/Llama-3.1-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: openmathinstruct2-llama-3.1-8B-Instruct-lr5-ep2
|
|
|
|
| 16 |
|
| 17 |
# openmathinstruct2-llama-3.1-8B-Instruct-lr5-ep2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct) on the openmathinstruct2_cot_20k_train dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7634
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.7634065747261047,
|
| 4 |
+
"eval_runtime": 31.2005,
|
| 5 |
+
"eval_samples_per_second": 59.262,
|
| 6 |
+
"eval_steps_per_second": 7.436,
|
| 7 |
+
"total_flos": 34318199685120.0,
|
| 8 |
+
"train_loss": 0.6641153321816371,
|
| 9 |
+
"train_runtime": 2465.8736,
|
| 10 |
+
"train_samples_per_second": 13.491,
|
| 11 |
+
"train_steps_per_second": 0.844
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.7634065747261047,
|
| 4 |
+
"eval_runtime": 31.2005,
|
| 5 |
+
"eval_samples_per_second": 59.262,
|
| 6 |
+
"eval_steps_per_second": 7.436
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"total_flos": 34318199685120.0,
|
| 4 |
+
"train_loss": 0.6641153321816371,
|
| 5 |
+
"train_runtime": 2465.8736,
|
| 6 |
+
"train_samples_per_second": 13.491,
|
| 7 |
+
"train_steps_per_second": 0.844
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1530 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2080,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.009615384615384616,
|
| 13 |
+
"grad_norm": 9.013119895723426,
|
| 14 |
+
"learning_rate": 4.807692307692308e-07,
|
| 15 |
+
"loss": 1.3011,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.019230769230769232,
|
| 20 |
+
"grad_norm": 8.170199017296364,
|
| 21 |
+
"learning_rate": 9.615384615384617e-07,
|
| 22 |
+
"loss": 1.1715,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.028846153846153848,
|
| 27 |
+
"grad_norm": 5.232892874296514,
|
| 28 |
+
"learning_rate": 1.4423076923076922e-06,
|
| 29 |
+
"loss": 0.9015,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.038461538461538464,
|
| 34 |
+
"grad_norm": 2.862322298571592,
|
| 35 |
+
"learning_rate": 1.9230769230769234e-06,
|
| 36 |
+
"loss": 0.7978,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04807692307692308,
|
| 41 |
+
"grad_norm": 2.7249762448136132,
|
| 42 |
+
"learning_rate": 2.403846153846154e-06,
|
| 43 |
+
"loss": 0.7609,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.057692307692307696,
|
| 48 |
+
"grad_norm": 3.169444209352466,
|
| 49 |
+
"learning_rate": 2.8846153846153845e-06,
|
| 50 |
+
"loss": 0.8146,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0673076923076923,
|
| 55 |
+
"grad_norm": 2.6189394700073807,
|
| 56 |
+
"learning_rate": 3.365384615384616e-06,
|
| 57 |
+
"loss": 0.777,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07692307692307693,
|
| 62 |
+
"grad_norm": 2.4397555428298077,
|
| 63 |
+
"learning_rate": 3.846153846153847e-06,
|
| 64 |
+
"loss": 0.78,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08653846153846154,
|
| 69 |
+
"grad_norm": 2.812934479310922,
|
| 70 |
+
"learning_rate": 4.326923076923077e-06,
|
| 71 |
+
"loss": 0.7721,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.09615384615384616,
|
| 76 |
+
"grad_norm": 2.5334288416130746,
|
| 77 |
+
"learning_rate": 4.807692307692308e-06,
|
| 78 |
+
"loss": 0.7669,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.10576923076923077,
|
| 83 |
+
"grad_norm": 2.342690607930004,
|
| 84 |
+
"learning_rate": 5.288461538461539e-06,
|
| 85 |
+
"loss": 0.7459,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.11538461538461539,
|
| 90 |
+
"grad_norm": 2.3269599095205264,
|
| 91 |
+
"learning_rate": 5.769230769230769e-06,
|
| 92 |
+
"loss": 0.7851,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.125,
|
| 97 |
+
"grad_norm": 2.42709508438667,
|
| 98 |
+
"learning_rate": 6.25e-06,
|
| 99 |
+
"loss": 0.7703,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1346153846153846,
|
| 104 |
+
"grad_norm": 2.6377643494039273,
|
| 105 |
+
"learning_rate": 6.730769230769232e-06,
|
| 106 |
+
"loss": 0.7496,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.14423076923076922,
|
| 111 |
+
"grad_norm": 2.875686037935072,
|
| 112 |
+
"learning_rate": 7.211538461538462e-06,
|
| 113 |
+
"loss": 0.784,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.15384615384615385,
|
| 118 |
+
"grad_norm": 2.1768155795174917,
|
| 119 |
+
"learning_rate": 7.692307692307694e-06,
|
| 120 |
+
"loss": 0.8022,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16346153846153846,
|
| 125 |
+
"grad_norm": 2.46654983618583,
|
| 126 |
+
"learning_rate": 8.173076923076923e-06,
|
| 127 |
+
"loss": 0.7741,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.17307692307692307,
|
| 132 |
+
"grad_norm": 2.4450582546281554,
|
| 133 |
+
"learning_rate": 8.653846153846155e-06,
|
| 134 |
+
"loss": 0.7827,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.18269230769230768,
|
| 139 |
+
"grad_norm": 2.4657718631331567,
|
| 140 |
+
"learning_rate": 9.134615384615384e-06,
|
| 141 |
+
"loss": 0.7894,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.19230769230769232,
|
| 146 |
+
"grad_norm": 2.54311809351163,
|
| 147 |
+
"learning_rate": 9.615384615384616e-06,
|
| 148 |
+
"loss": 0.8164,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.20192307692307693,
|
| 153 |
+
"grad_norm": 2.097681181559205,
|
| 154 |
+
"learning_rate": 9.999971836433636e-06,
|
| 155 |
+
"loss": 0.8139,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.21153846153846154,
|
| 160 |
+
"grad_norm": 2.2579921721680516,
|
| 161 |
+
"learning_rate": 9.998986144924253e-06,
|
| 162 |
+
"loss": 0.7919,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.22115384615384615,
|
| 167 |
+
"grad_norm": 2.1097886183574466,
|
| 168 |
+
"learning_rate": 9.996592592355083e-06,
|
| 169 |
+
"loss": 0.8056,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.23076923076923078,
|
| 174 |
+
"grad_norm": 2.22507095955906,
|
| 175 |
+
"learning_rate": 9.992791852820709e-06,
|
| 176 |
+
"loss": 0.8108,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2403846153846154,
|
| 181 |
+
"grad_norm": 2.1931221071365425,
|
| 182 |
+
"learning_rate": 9.987584996720813e-06,
|
| 183 |
+
"loss": 0.8041,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.25,
|
| 188 |
+
"grad_norm": 2.1220375928827044,
|
| 189 |
+
"learning_rate": 9.980973490458728e-06,
|
| 190 |
+
"loss": 0.8149,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.25961538461538464,
|
| 195 |
+
"grad_norm": 2.1068957326041904,
|
| 196 |
+
"learning_rate": 9.972959196028456e-06,
|
| 197 |
+
"loss": 0.7914,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2692307692307692,
|
| 202 |
+
"grad_norm": 1.9372196057401814,
|
| 203 |
+
"learning_rate": 9.96354437049027e-06,
|
| 204 |
+
"loss": 0.8033,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.27884615384615385,
|
| 209 |
+
"grad_norm": 2.2606013253960846,
|
| 210 |
+
"learning_rate": 9.952731665335071e-06,
|
| 211 |
+
"loss": 0.8039,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.28846153846153844,
|
| 216 |
+
"grad_norm": 2.1614476516877406,
|
| 217 |
+
"learning_rate": 9.940524125737641e-06,
|
| 218 |
+
"loss": 0.7647,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2980769230769231,
|
| 223 |
+
"grad_norm": 2.220222508508704,
|
| 224 |
+
"learning_rate": 9.92692518969903e-06,
|
| 225 |
+
"loss": 0.7901,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.3076923076923077,
|
| 230 |
+
"grad_norm": 2.201097779810873,
|
| 231 |
+
"learning_rate": 9.911938687078324e-06,
|
| 232 |
+
"loss": 0.7964,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3173076923076923,
|
| 237 |
+
"grad_norm": 2.029609328977134,
|
| 238 |
+
"learning_rate": 9.895568838514042e-06,
|
| 239 |
+
"loss": 0.7764,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3269230769230769,
|
| 244 |
+
"grad_norm": 2.2142291176229834,
|
| 245 |
+
"learning_rate": 9.87782025423547e-06,
|
| 246 |
+
"loss": 0.8005,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.33653846153846156,
|
| 251 |
+
"grad_norm": 2.084315605131572,
|
| 252 |
+
"learning_rate": 9.8586979327643e-06,
|
| 253 |
+
"loss": 0.7795,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.34615384615384615,
|
| 258 |
+
"grad_norm": 2.3024285135825457,
|
| 259 |
+
"learning_rate": 9.838207259506891e-06,
|
| 260 |
+
"loss": 0.7935,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3557692307692308,
|
| 265 |
+
"grad_norm": 2.5288625621043326,
|
| 266 |
+
"learning_rate": 9.816354005237583e-06,
|
| 267 |
+
"loss": 0.7749,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.36538461538461536,
|
| 272 |
+
"grad_norm": 2.0010687704613743,
|
| 273 |
+
"learning_rate": 9.793144324473473e-06,
|
| 274 |
+
"loss": 0.7988,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.375,
|
| 279 |
+
"grad_norm": 2.1536447186627496,
|
| 280 |
+
"learning_rate": 9.768584753741134e-06,
|
| 281 |
+
"loss": 0.7955,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.38461538461538464,
|
| 286 |
+
"grad_norm": 2.025116331685143,
|
| 287 |
+
"learning_rate": 9.742682209735727e-06,
|
| 288 |
+
"loss": 0.8014,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3942307692307692,
|
| 293 |
+
"grad_norm": 2.027340612157234,
|
| 294 |
+
"learning_rate": 9.715443987373062e-06,
|
| 295 |
+
"loss": 0.8341,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.40384615384615385,
|
| 300 |
+
"grad_norm": 1.968065863088579,
|
| 301 |
+
"learning_rate": 9.686877757735126e-06,
|
| 302 |
+
"loss": 0.7922,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.41346153846153844,
|
| 307 |
+
"grad_norm": 2.016993760012052,
|
| 308 |
+
"learning_rate": 9.656991565909703e-06,
|
| 309 |
+
"loss": 0.8328,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4230769230769231,
|
| 314 |
+
"grad_norm": 2.1902508822456213,
|
| 315 |
+
"learning_rate": 9.62579382872462e-06,
|
| 316 |
+
"loss": 0.8029,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4326923076923077,
|
| 321 |
+
"grad_norm": 2.0187820173893534,
|
| 322 |
+
"learning_rate": 9.593293332377325e-06,
|
| 323 |
+
"loss": 0.7634,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.4423076923076923,
|
| 328 |
+
"grad_norm": 2.1364408496498437,
|
| 329 |
+
"learning_rate": 9.55949922996045e-06,
|
| 330 |
+
"loss": 0.774,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4519230769230769,
|
| 335 |
+
"grad_norm": 2.0116381073985945,
|
| 336 |
+
"learning_rate": 9.52442103888402e-06,
|
| 337 |
+
"loss": 0.8331,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.46153846153846156,
|
| 342 |
+
"grad_norm": 2.100311438402205,
|
| 343 |
+
"learning_rate": 9.488068638195072e-06,
|
| 344 |
+
"loss": 0.7478,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.47115384615384615,
|
| 349 |
+
"grad_norm": 2.059472131869095,
|
| 350 |
+
"learning_rate": 9.450452265795423e-06,
|
| 351 |
+
"loss": 0.8202,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.4807692307692308,
|
| 356 |
+
"grad_norm": 2.074026027178722,
|
| 357 |
+
"learning_rate": 9.411582515558391e-06,
|
| 358 |
+
"loss": 0.8177,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4807692307692308,
|
| 363 |
+
"eval_loss": 0.7822802066802979,
|
| 364 |
+
"eval_runtime": 31.3783,
|
| 365 |
+
"eval_samples_per_second": 58.926,
|
| 366 |
+
"eval_steps_per_second": 7.394,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.49038461538461536,
|
| 371 |
+
"grad_norm": 1.9938002782487565,
|
| 372 |
+
"learning_rate": 9.371470334345232e-06,
|
| 373 |
+
"loss": 0.804,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.5,
|
| 378 |
+
"grad_norm": 1.9596865504803214,
|
| 379 |
+
"learning_rate": 9.330127018922195e-06,
|
| 380 |
+
"loss": 0.7844,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.5096153846153846,
|
| 385 |
+
"grad_norm": 1.8726235477694055,
|
| 386 |
+
"learning_rate": 9.287564212779012e-06,
|
| 387 |
+
"loss": 0.756,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.5192307692307693,
|
| 392 |
+
"grad_norm": 2.091468778530183,
|
| 393 |
+
"learning_rate": 9.243793902849764e-06,
|
| 394 |
+
"loss": 0.8057,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.5288461538461539,
|
| 399 |
+
"grad_norm": 2.032448528673805,
|
| 400 |
+
"learning_rate": 9.198828416136991e-06,
|
| 401 |
+
"loss": 0.8065,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5384615384615384,
|
| 406 |
+
"grad_norm": 2.0648947345620137,
|
| 407 |
+
"learning_rate": 9.152680416240059e-06,
|
| 408 |
+
"loss": 0.8101,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5480769230769231,
|
| 413 |
+
"grad_norm": 1.864661141720549,
|
| 414 |
+
"learning_rate": 9.10536289978872e-06,
|
| 415 |
+
"loss": 0.7597,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5576923076923077,
|
| 420 |
+
"grad_norm": 1.8389259729678664,
|
| 421 |
+
"learning_rate": 9.056889192782865e-06,
|
| 422 |
+
"loss": 0.806,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5673076923076923,
|
| 427 |
+
"grad_norm": 1.997501225318945,
|
| 428 |
+
"learning_rate": 9.007272946839559e-06,
|
| 429 |
+
"loss": 0.7696,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5769230769230769,
|
| 434 |
+
"grad_norm": 1.9510249081212931,
|
| 435 |
+
"learning_rate": 8.95652813534831e-06,
|
| 436 |
+
"loss": 0.7779,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5865384615384616,
|
| 441 |
+
"grad_norm": 2.043423090422744,
|
| 442 |
+
"learning_rate": 8.90466904953579e-06,
|
| 443 |
+
"loss": 0.8142,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5961538461538461,
|
| 448 |
+
"grad_norm": 2.1520676063407,
|
| 449 |
+
"learning_rate": 8.851710294440974e-06,
|
| 450 |
+
"loss": 0.804,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.6057692307692307,
|
| 455 |
+
"grad_norm": 1.8353482991425896,
|
| 456 |
+
"learning_rate": 8.797666784801954e-06,
|
| 457 |
+
"loss": 0.7701,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.6153846153846154,
|
| 462 |
+
"grad_norm": 1.8647174071179153,
|
| 463 |
+
"learning_rate": 8.742553740855507e-06,
|
| 464 |
+
"loss": 0.7641,
|
| 465 |
+
"step": 640
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.625,
|
| 469 |
+
"grad_norm": 1.859731623435647,
|
| 470 |
+
"learning_rate": 8.68638668405062e-06,
|
| 471 |
+
"loss": 0.8293,
|
| 472 |
+
"step": 650
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.6346153846153846,
|
| 476 |
+
"grad_norm": 2.035763511676626,
|
| 477 |
+
"learning_rate": 8.629181432677213e-06,
|
| 478 |
+
"loss": 0.8386,
|
| 479 |
+
"step": 660
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6442307692307693,
|
| 483 |
+
"grad_norm": 1.9498309927555528,
|
| 484 |
+
"learning_rate": 8.570954097411224e-06,
|
| 485 |
+
"loss": 0.826,
|
| 486 |
+
"step": 670
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6538461538461539,
|
| 490 |
+
"grad_norm": 2.157631582471241,
|
| 491 |
+
"learning_rate": 8.511721076777388e-06,
|
| 492 |
+
"loss": 0.7933,
|
| 493 |
+
"step": 680
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6634615384615384,
|
| 497 |
+
"grad_norm": 1.817540210529388,
|
| 498 |
+
"learning_rate": 8.451499052530923e-06,
|
| 499 |
+
"loss": 0.794,
|
| 500 |
+
"step": 690
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6730769230769231,
|
| 504 |
+
"grad_norm": 2.003699550055361,
|
| 505 |
+
"learning_rate": 8.390304984959455e-06,
|
| 506 |
+
"loss": 0.8152,
|
| 507 |
+
"step": 700
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6826923076923077,
|
| 511 |
+
"grad_norm": 1.7853207622424534,
|
| 512 |
+
"learning_rate": 8.328156108106518e-06,
|
| 513 |
+
"loss": 0.7869,
|
| 514 |
+
"step": 710
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6923076923076923,
|
| 518 |
+
"grad_norm": 1.951815011854146,
|
| 519 |
+
"learning_rate": 8.265069924917925e-06,
|
| 520 |
+
"loss": 0.8086,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.7019230769230769,
|
| 525 |
+
"grad_norm": 1.904892870037223,
|
| 526 |
+
"learning_rate": 8.20106420231244e-06,
|
| 527 |
+
"loss": 0.7959,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.7115384615384616,
|
| 532 |
+
"grad_norm": 1.9417662866059235,
|
| 533 |
+
"learning_rate": 8.136156966178082e-06,
|
| 534 |
+
"loss": 0.8072,
|
| 535 |
+
"step": 740
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.7211538461538461,
|
| 539 |
+
"grad_norm": 1.8480817149859605,
|
| 540 |
+
"learning_rate": 8.070366496295505e-06,
|
| 541 |
+
"loss": 0.7814,
|
| 542 |
+
"step": 750
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7307692307692307,
|
| 546 |
+
"grad_norm": 1.8327729681591678,
|
| 547 |
+
"learning_rate": 8.003711321189895e-06,
|
| 548 |
+
"loss": 0.7786,
|
| 549 |
+
"step": 760
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.7403846153846154,
|
| 553 |
+
"grad_norm": 2.1198206040027494,
|
| 554 |
+
"learning_rate": 7.93621021291277e-06,
|
| 555 |
+
"loss": 0.7485,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.75,
|
| 560 |
+
"grad_norm": 1.9784614334378419,
|
| 561 |
+
"learning_rate": 7.86788218175523e-06,
|
| 562 |
+
"loss": 0.7736,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7596153846153846,
|
| 567 |
+
"grad_norm": 1.733613545143569,
|
| 568 |
+
"learning_rate": 7.798746470894113e-06,
|
| 569 |
+
"loss": 0.7613,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.7692307692307693,
|
| 574 |
+
"grad_norm": 1.7905885212318795,
|
| 575 |
+
"learning_rate": 7.728822550972523e-06,
|
| 576 |
+
"loss": 0.7902,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7788461538461539,
|
| 581 |
+
"grad_norm": 2.02868421572116,
|
| 582 |
+
"learning_rate": 7.658130114616364e-06,
|
| 583 |
+
"loss": 0.7962,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7884615384615384,
|
| 588 |
+
"grad_norm": 1.8163329758157731,
|
| 589 |
+
"learning_rate": 7.586689070888284e-06,
|
| 590 |
+
"loss": 0.7625,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7980769230769231,
|
| 595 |
+
"grad_norm": 2.0352374441953542,
|
| 596 |
+
"learning_rate": 7.5145195396807244e-06,
|
| 597 |
+
"loss": 0.7836,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8076923076923077,
|
| 602 |
+
"grad_norm": 2.0684855751019144,
|
| 603 |
+
"learning_rate": 7.441641846049557e-06,
|
| 604 |
+
"loss": 0.8016,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.8173076923076923,
|
| 609 |
+
"grad_norm": 2.042151868500934,
|
| 610 |
+
"learning_rate": 7.368076514489947e-06,
|
| 611 |
+
"loss": 0.8012,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.8269230769230769,
|
| 616 |
+
"grad_norm": 1.9253750246288859,
|
| 617 |
+
"learning_rate": 7.2938442631560714e-06,
|
| 618 |
+
"loss": 0.799,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.8365384615384616,
|
| 623 |
+
"grad_norm": 1.8896143274467088,
|
| 624 |
+
"learning_rate": 7.218965998026259e-06,
|
| 625 |
+
"loss": 0.8099,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.8461538461538461,
|
| 630 |
+
"grad_norm": 1.7590979198633765,
|
| 631 |
+
"learning_rate": 7.143462807015271e-06,
|
| 632 |
+
"loss": 0.7726,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8557692307692307,
|
| 637 |
+
"grad_norm": 1.9332789610600787,
|
| 638 |
+
"learning_rate": 7.067355954035316e-06,
|
| 639 |
+
"loss": 0.808,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8653846153846154,
|
| 644 |
+
"grad_norm": 1.8265190820741775,
|
| 645 |
+
"learning_rate": 6.990666873007506e-06,
|
| 646 |
+
"loss": 0.7604,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.875,
|
| 651 |
+
"grad_norm": 1.8486951872238018,
|
| 652 |
+
"learning_rate": 6.913417161825449e-06,
|
| 653 |
+
"loss": 0.7861,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8846153846153846,
|
| 658 |
+
"grad_norm": 2.0473377615419412,
|
| 659 |
+
"learning_rate": 6.8356285762726385e-06,
|
| 660 |
+
"loss": 0.7618,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8942307692307693,
|
| 665 |
+
"grad_norm": 1.7908009618022396,
|
| 666 |
+
"learning_rate": 6.757323023895388e-06,
|
| 667 |
+
"loss": 0.7547,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.9038461538461539,
|
| 672 |
+
"grad_norm": 1.7499349918180362,
|
| 673 |
+
"learning_rate": 6.678522557833025e-06,
|
| 674 |
+
"loss": 0.7595,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.9134615384615384,
|
| 679 |
+
"grad_norm": 1.9193288351945685,
|
| 680 |
+
"learning_rate": 6.599249370607083e-06,
|
| 681 |
+
"loss": 0.7465,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.9230769230769231,
|
| 686 |
+
"grad_norm": 1.7382141318072688,
|
| 687 |
+
"learning_rate": 6.519525787871235e-06,
|
| 688 |
+
"loss": 0.7829,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9326923076923077,
|
| 693 |
+
"grad_norm": 1.6639815518816288,
|
| 694 |
+
"learning_rate": 6.439374262123731e-06,
|
| 695 |
+
"loss": 0.7483,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9423076923076923,
|
| 700 |
+
"grad_norm": 2.040809216745246,
|
| 701 |
+
"learning_rate": 6.358817366384122e-06,
|
| 702 |
+
"loss": 0.7695,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9519230769230769,
|
| 707 |
+
"grad_norm": 1.9765992795556684,
|
| 708 |
+
"learning_rate": 6.277877787836034e-06,
|
| 709 |
+
"loss": 0.8039,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9615384615384616,
|
| 714 |
+
"grad_norm": 1.7379165898055615,
|
| 715 |
+
"learning_rate": 6.1965783214377895e-06,
|
| 716 |
+
"loss": 0.7708,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9615384615384616,
|
| 721 |
+
"eval_loss": 0.757188618183136,
|
| 722 |
+
"eval_runtime": 30.718,
|
| 723 |
+
"eval_samples_per_second": 60.193,
|
| 724 |
+
"eval_steps_per_second": 7.553,
|
| 725 |
+
"step": 1000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.9711538461538461,
|
| 729 |
+
"grad_norm": 1.91002204642886,
|
| 730 |
+
"learning_rate": 6.114941863502682e-06,
|
| 731 |
+
"loss": 0.791,
|
| 732 |
+
"step": 1010
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.9807692307692307,
|
| 736 |
+
"grad_norm": 1.7547449600143168,
|
| 737 |
+
"learning_rate": 6.032991405250702e-06,
|
| 738 |
+
"loss": 0.7584,
|
| 739 |
+
"step": 1020
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.9903846153846154,
|
| 743 |
+
"grad_norm": 1.7518146591240198,
|
| 744 |
+
"learning_rate": 5.950750026333534e-06,
|
| 745 |
+
"loss": 0.7911,
|
| 746 |
+
"step": 1030
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.0,
|
| 750 |
+
"grad_norm": 1.8856712522978225,
|
| 751 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 752 |
+
"loss": 0.7762,
|
| 753 |
+
"step": 1040
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.0096153846153846,
|
| 757 |
+
"grad_norm": 2.6930087290136417,
|
| 758 |
+
"learning_rate": 5.785487228246339e-06,
|
| 759 |
+
"loss": 0.6172,
|
| 760 |
+
"step": 1050
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.0192307692307692,
|
| 764 |
+
"grad_norm": 2.0128509132494217,
|
| 765 |
+
"learning_rate": 5.7025123519254644e-06,
|
| 766 |
+
"loss": 0.5335,
|
| 767 |
+
"step": 1060
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.0288461538461537,
|
| 771 |
+
"grad_norm": 2.081106502499814,
|
| 772 |
+
"learning_rate": 5.619339627529876e-06,
|
| 773 |
+
"loss": 0.5475,
|
| 774 |
+
"step": 1070
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.0384615384615385,
|
| 778 |
+
"grad_norm": 2.089872779664623,
|
| 779 |
+
"learning_rate": 5.53599247893724e-06,
|
| 780 |
+
"loss": 0.5491,
|
| 781 |
+
"step": 1080
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.0480769230769231,
|
| 785 |
+
"grad_norm": 2.487028007682742,
|
| 786 |
+
"learning_rate": 5.45249437914819e-06,
|
| 787 |
+
"loss": 0.5396,
|
| 788 |
+
"step": 1090
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.0576923076923077,
|
| 792 |
+
"grad_norm": 2.1183702883792015,
|
| 793 |
+
"learning_rate": 5.368868843675642e-06,
|
| 794 |
+
"loss": 0.5602,
|
| 795 |
+
"step": 1100
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 1.0673076923076923,
|
| 799 |
+
"grad_norm": 2.231623903215653,
|
| 800 |
+
"learning_rate": 5.285139423922138e-06,
|
| 801 |
+
"loss": 0.5341,
|
| 802 |
+
"step": 1110
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 1.0769230769230769,
|
| 806 |
+
"grad_norm": 1.8590648999967128,
|
| 807 |
+
"learning_rate": 5.201329700547077e-06,
|
| 808 |
+
"loss": 0.5499,
|
| 809 |
+
"step": 1120
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 1.0865384615384615,
|
| 813 |
+
"grad_norm": 2.031456951588171,
|
| 814 |
+
"learning_rate": 5.117463276825711e-06,
|
| 815 |
+
"loss": 0.5541,
|
| 816 |
+
"step": 1130
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 1.0961538461538463,
|
| 820 |
+
"grad_norm": 1.996025281675884,
|
| 821 |
+
"learning_rate": 5.033563772001782e-06,
|
| 822 |
+
"loss": 0.5229,
|
| 823 |
+
"step": 1140
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.1057692307692308,
|
| 827 |
+
"grad_norm": 2.01803070793536,
|
| 828 |
+
"learning_rate": 4.949654814635623e-06,
|
| 829 |
+
"loss": 0.5534,
|
| 830 |
+
"step": 1150
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.1153846153846154,
|
| 834 |
+
"grad_norm": 2.0131103606600163,
|
| 835 |
+
"learning_rate": 4.865760035949695e-06,
|
| 836 |
+
"loss": 0.5503,
|
| 837 |
+
"step": 1160
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.125,
|
| 841 |
+
"grad_norm": 1.9109776540155083,
|
| 842 |
+
"learning_rate": 4.781903063173321e-06,
|
| 843 |
+
"loss": 0.5587,
|
| 844 |
+
"step": 1170
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 1.1346153846153846,
|
| 848 |
+
"grad_norm": 2.240195731738998,
|
| 849 |
+
"learning_rate": 4.69810751288857e-06,
|
| 850 |
+
"loss": 0.5345,
|
| 851 |
+
"step": 1180
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 1.1442307692307692,
|
| 855 |
+
"grad_norm": 2.21815291482556,
|
| 856 |
+
"learning_rate": 4.61439698437914e-06,
|
| 857 |
+
"loss": 0.5105,
|
| 858 |
+
"step": 1190
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 1.1538461538461537,
|
| 862 |
+
"grad_norm": 1.959424085301074,
|
| 863 |
+
"learning_rate": 4.530795052984104e-06,
|
| 864 |
+
"loss": 0.5472,
|
| 865 |
+
"step": 1200
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 1.1634615384615385,
|
| 869 |
+
"grad_norm": 1.9724626057174217,
|
| 870 |
+
"learning_rate": 4.447325263458401e-06,
|
| 871 |
+
"loss": 0.5425,
|
| 872 |
+
"step": 1210
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.1730769230769231,
|
| 876 |
+
"grad_norm": 1.9404277504052594,
|
| 877 |
+
"learning_rate": 4.364011123341947e-06,
|
| 878 |
+
"loss": 0.5383,
|
| 879 |
+
"step": 1220
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.1826923076923077,
|
| 883 |
+
"grad_norm": 1.975957154057773,
|
| 884 |
+
"learning_rate": 4.280876096339222e-06,
|
| 885 |
+
"loss": 0.5635,
|
| 886 |
+
"step": 1230
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.1923076923076923,
|
| 890 |
+
"grad_norm": 3.7893846099890203,
|
| 891 |
+
"learning_rate": 4.1979435957111984e-06,
|
| 892 |
+
"loss": 0.5508,
|
| 893 |
+
"step": 1240
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 1.2019230769230769,
|
| 897 |
+
"grad_norm": 2.088379071314605,
|
| 898 |
+
"learning_rate": 4.11523697768149e-06,
|
| 899 |
+
"loss": 0.5917,
|
| 900 |
+
"step": 1250
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 1.2115384615384615,
|
| 904 |
+
"grad_norm": 2.0699986543823914,
|
| 905 |
+
"learning_rate": 4.032779534858544e-06,
|
| 906 |
+
"loss": 0.5549,
|
| 907 |
+
"step": 1260
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 1.2211538461538463,
|
| 911 |
+
"grad_norm": 2.4531750427167744,
|
| 912 |
+
"learning_rate": 3.9505944896757635e-06,
|
| 913 |
+
"loss": 0.5314,
|
| 914 |
+
"step": 1270
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 1.2307692307692308,
|
| 918 |
+
"grad_norm": 2.130560765169762,
|
| 919 |
+
"learning_rate": 3.86870498785139e-06,
|
| 920 |
+
"loss": 0.5327,
|
| 921 |
+
"step": 1280
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 1.2403846153846154,
|
| 925 |
+
"grad_norm": 2.2163755882504947,
|
| 926 |
+
"learning_rate": 3.7871340918699945e-06,
|
| 927 |
+
"loss": 0.509,
|
| 928 |
+
"step": 1290
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.25,
|
| 932 |
+
"grad_norm": 2.1107171131060136,
|
| 933 |
+
"learning_rate": 3.705904774487396e-06,
|
| 934 |
+
"loss": 0.5153,
|
| 935 |
+
"step": 1300
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 1.2596153846153846,
|
| 939 |
+
"grad_norm": 2.1786622272309413,
|
| 940 |
+
"learning_rate": 3.6250399122608713e-06,
|
| 941 |
+
"loss": 0.5589,
|
| 942 |
+
"step": 1310
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 1.2692307692307692,
|
| 946 |
+
"grad_norm": 1.9962608207111503,
|
| 947 |
+
"learning_rate": 3.544562279106436e-06,
|
| 948 |
+
"loss": 0.5425,
|
| 949 |
+
"step": 1320
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 1.2788461538461537,
|
| 953 |
+
"grad_norm": 1.8464394968329445,
|
| 954 |
+
"learning_rate": 3.464494539885047e-06,
|
| 955 |
+
"loss": 0.5349,
|
| 956 |
+
"step": 1330
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 1.2884615384615383,
|
| 960 |
+
"grad_norm": 2.138267726845245,
|
| 961 |
+
"learning_rate": 3.3848592440195118e-06,
|
| 962 |
+
"loss": 0.5461,
|
| 963 |
+
"step": 1340
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 1.2980769230769231,
|
| 967 |
+
"grad_norm": 2.0087619862264767,
|
| 968 |
+
"learning_rate": 3.3056788191439116e-06,
|
| 969 |
+
"loss": 0.559,
|
| 970 |
+
"step": 1350
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 1.3076923076923077,
|
| 974 |
+
"grad_norm": 2.1318708987886854,
|
| 975 |
+
"learning_rate": 3.226975564787322e-06,
|
| 976 |
+
"loss": 0.5326,
|
| 977 |
+
"step": 1360
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 1.3173076923076923,
|
| 981 |
+
"grad_norm": 2.0595039744355725,
|
| 982 |
+
"learning_rate": 3.148771646093608e-06,
|
| 983 |
+
"loss": 0.5215,
|
| 984 |
+
"step": 1370
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.3269230769230769,
|
| 988 |
+
"grad_norm": 2.106731128516135,
|
| 989 |
+
"learning_rate": 3.0710890875790745e-06,
|
| 990 |
+
"loss": 0.5205,
|
| 991 |
+
"step": 1380
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 1.3365384615384617,
|
| 995 |
+
"grad_norm": 2.015598237138461,
|
| 996 |
+
"learning_rate": 2.993949766929711e-06,
|
| 997 |
+
"loss": 0.5155,
|
| 998 |
+
"step": 1390
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.3461538461538463,
|
| 1002 |
+
"grad_norm": 2.089932096970866,
|
| 1003 |
+
"learning_rate": 2.917375408839803e-06,
|
| 1004 |
+
"loss": 0.5321,
|
| 1005 |
+
"step": 1400
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.3557692307692308,
|
| 1009 |
+
"grad_norm": 2.1011121825872414,
|
| 1010 |
+
"learning_rate": 2.8413875788936067e-06,
|
| 1011 |
+
"loss": 0.5654,
|
| 1012 |
+
"step": 1410
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.3653846153846154,
|
| 1016 |
+
"grad_norm": 2.5049128653170607,
|
| 1017 |
+
"learning_rate": 2.766007677491871e-06,
|
| 1018 |
+
"loss": 0.5332,
|
| 1019 |
+
"step": 1420
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.375,
|
| 1023 |
+
"grad_norm": 2.199796386555421,
|
| 1024 |
+
"learning_rate": 2.6912569338248317e-06,
|
| 1025 |
+
"loss": 0.5129,
|
| 1026 |
+
"step": 1430
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.3846153846153846,
|
| 1030 |
+
"grad_norm": 2.122643085103428,
|
| 1031 |
+
"learning_rate": 2.6171563998934605e-06,
|
| 1032 |
+
"loss": 0.5315,
|
| 1033 |
+
"step": 1440
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.3942307692307692,
|
| 1037 |
+
"grad_norm": 1.9680294145904305,
|
| 1038 |
+
"learning_rate": 2.5437269445806146e-06,
|
| 1039 |
+
"loss": 0.5388,
|
| 1040 |
+
"step": 1450
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 1.4038461538461537,
|
| 1044 |
+
"grad_norm": 2.1649286481102505,
|
| 1045 |
+
"learning_rate": 2.4709892477737263e-06,
|
| 1046 |
+
"loss": 0.5449,
|
| 1047 |
+
"step": 1460
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 1.4134615384615383,
|
| 1051 |
+
"grad_norm": 1.9108821348630196,
|
| 1052 |
+
"learning_rate": 2.3989637945407547e-06,
|
| 1053 |
+
"loss": 0.5124,
|
| 1054 |
+
"step": 1470
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 1.4230769230769231,
|
| 1058 |
+
"grad_norm": 1.8705842302466216,
|
| 1059 |
+
"learning_rate": 2.3276708693609947e-06,
|
| 1060 |
+
"loss": 0.5271,
|
| 1061 |
+
"step": 1480
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 1.4326923076923077,
|
| 1065 |
+
"grad_norm": 2.218892686030775,
|
| 1066 |
+
"learning_rate": 2.2571305504123547e-06,
|
| 1067 |
+
"loss": 0.536,
|
| 1068 |
+
"step": 1490
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 1.4423076923076923,
|
| 1072 |
+
"grad_norm": 2.151762003437118,
|
| 1073 |
+
"learning_rate": 2.187362703916766e-06,
|
| 1074 |
+
"loss": 0.5513,
|
| 1075 |
+
"step": 1500
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 1.4423076923076923,
|
| 1079 |
+
"eval_loss": 0.7693456411361694,
|
| 1080 |
+
"eval_runtime": 31.5006,
|
| 1081 |
+
"eval_samples_per_second": 58.697,
|
| 1082 |
+
"eval_steps_per_second": 7.365,
|
| 1083 |
+
"step": 1500
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.4519230769230769,
|
| 1087 |
+
"grad_norm": 2.2335086500452768,
|
| 1088 |
+
"learning_rate": 2.1183869785452744e-06,
|
| 1089 |
+
"loss": 0.5343,
|
| 1090 |
+
"step": 1510
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.4615384615384617,
|
| 1094 |
+
"grad_norm": 2.201272848216364,
|
| 1095 |
+
"learning_rate": 2.050222799884387e-06,
|
| 1096 |
+
"loss": 0.5203,
|
| 1097 |
+
"step": 1520
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.4711538461538463,
|
| 1101 |
+
"grad_norm": 2.245522812744825,
|
| 1102 |
+
"learning_rate": 1.9828893649652653e-06,
|
| 1103 |
+
"loss": 0.5492,
|
| 1104 |
+
"step": 1530
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.4807692307692308,
|
| 1108 |
+
"grad_norm": 2.4219731931643134,
|
| 1109 |
+
"learning_rate": 1.9164056368572847e-06,
|
| 1110 |
+
"loss": 0.531,
|
| 1111 |
+
"step": 1540
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.4903846153846154,
|
| 1115 |
+
"grad_norm": 1.9852458155211052,
|
| 1116 |
+
"learning_rate": 1.8507903393274622e-06,
|
| 1117 |
+
"loss": 0.5413,
|
| 1118 |
+
"step": 1550
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.5,
|
| 1122 |
+
"grad_norm": 2.057532782936977,
|
| 1123 |
+
"learning_rate": 1.7860619515673034e-06,
|
| 1124 |
+
"loss": 0.5271,
|
| 1125 |
+
"step": 1560
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 1.5096153846153846,
|
| 1129 |
+
"grad_norm": 2.3471986544512187,
|
| 1130 |
+
"learning_rate": 1.7222387029885268e-06,
|
| 1131 |
+
"loss": 0.5579,
|
| 1132 |
+
"step": 1570
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 1.5192307692307692,
|
| 1136 |
+
"grad_norm": 2.12927805118844,
|
| 1137 |
+
"learning_rate": 1.6593385680891139e-06,
|
| 1138 |
+
"loss": 0.524,
|
| 1139 |
+
"step": 1580
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 1.5288461538461537,
|
| 1143 |
+
"grad_norm": 2.3452167989714785,
|
| 1144 |
+
"learning_rate": 1.5973792613911698e-06,
|
| 1145 |
+
"loss": 0.5263,
|
| 1146 |
+
"step": 1590
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 1.5384615384615383,
|
| 1150 |
+
"grad_norm": 2.3303569386029115,
|
| 1151 |
+
"learning_rate": 1.5363782324520033e-06,
|
| 1152 |
+
"loss": 0.5075,
|
| 1153 |
+
"step": 1600
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.5480769230769231,
|
| 1157 |
+
"grad_norm": 2.3477866769254514,
|
| 1158 |
+
"learning_rate": 1.476352660949802e-06,
|
| 1159 |
+
"loss": 0.5611,
|
| 1160 |
+
"step": 1610
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.5576923076923077,
|
| 1164 |
+
"grad_norm": 1.9831558221073002,
|
| 1165 |
+
"learning_rate": 1.4173194518453415e-06,
|
| 1166 |
+
"loss": 0.5124,
|
| 1167 |
+
"step": 1620
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 1.5673076923076923,
|
| 1171 |
+
"grad_norm": 2.5946859927202732,
|
| 1172 |
+
"learning_rate": 1.3592952306210589e-06,
|
| 1173 |
+
"loss": 0.5193,
|
| 1174 |
+
"step": 1630
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 1.5769230769230769,
|
| 1178 |
+
"grad_norm": 2.3095838494615433,
|
| 1179 |
+
"learning_rate": 1.3022963385988153e-06,
|
| 1180 |
+
"loss": 0.5374,
|
| 1181 |
+
"step": 1640
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 1.5865384615384617,
|
| 1185 |
+
"grad_norm": 2.0028485071747806,
|
| 1186 |
+
"learning_rate": 1.246338828337707e-06,
|
| 1187 |
+
"loss": 0.512,
|
| 1188 |
+
"step": 1650
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 1.5961538461538463,
|
| 1192 |
+
"grad_norm": 2.029535183303286,
|
| 1193 |
+
"learning_rate": 1.1914384591132045e-06,
|
| 1194 |
+
"loss": 0.5128,
|
| 1195 |
+
"step": 1660
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 1.6057692307692308,
|
| 1199 |
+
"grad_norm": 2.147413310284848,
|
| 1200 |
+
"learning_rate": 1.1376106924788594e-06,
|
| 1201 |
+
"loss": 0.5316,
|
| 1202 |
+
"step": 1670
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 1.6153846153846154,
|
| 1206 |
+
"grad_norm": 2.3899145312729693,
|
| 1207 |
+
"learning_rate": 1.0848706879118893e-06,
|
| 1208 |
+
"loss": 0.5658,
|
| 1209 |
+
"step": 1680
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 1.625,
|
| 1213 |
+
"grad_norm": 2.0348123827167335,
|
| 1214 |
+
"learning_rate": 1.0332332985438248e-06,
|
| 1215 |
+
"loss": 0.5269,
|
| 1216 |
+
"step": 1690
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 1.6346153846153846,
|
| 1220 |
+
"grad_norm": 1.8531683575644768,
|
| 1221 |
+
"learning_rate": 9.82713066977427e-07,
|
| 1222 |
+
"loss": 0.5405,
|
| 1223 |
+
"step": 1700
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 1.6442307692307692,
|
| 1227 |
+
"grad_norm": 1.9592530320960306,
|
| 1228 |
+
"learning_rate": 9.333242211910687e-07,
|
| 1229 |
+
"loss": 0.5184,
|
| 1230 |
+
"step": 1710
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 1.6538461538461537,
|
| 1234 |
+
"grad_norm": 1.9954409383029053,
|
| 1235 |
+
"learning_rate": 8.850806705317183e-07,
|
| 1236 |
+
"loss": 0.4852,
|
| 1237 |
+
"step": 1720
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 1.6634615384615383,
|
| 1241 |
+
"grad_norm": 2.098998703354889,
|
| 1242 |
+
"learning_rate": 8.379960017976546e-07,
|
| 1243 |
+
"loss": 0.5139,
|
| 1244 |
+
"step": 1730
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 1.6730769230769231,
|
| 1248 |
+
"grad_norm": 2.5491851954980658,
|
| 1249 |
+
"learning_rate": 7.920834754120305e-07,
|
| 1250 |
+
"loss": 0.5392,
|
| 1251 |
+
"step": 1740
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 1.6826923076923077,
|
| 1255 |
+
"grad_norm": 2.3935228227975,
|
| 1256 |
+
"learning_rate": 7.473560216883524e-07,
|
| 1257 |
+
"loss": 0.533,
|
| 1258 |
+
"step": 1750
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.6923076923076923,
|
| 1262 |
+
"grad_norm": 2.170304242236588,
|
| 1263 |
+
"learning_rate": 7.03826237188916e-07,
|
| 1264 |
+
"loss": 0.5523,
|
| 1265 |
+
"step": 1760
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.7019230769230769,
|
| 1269 |
+
"grad_norm": 2.238634395178553,
|
| 1270 |
+
"learning_rate": 6.615063811772532e-07,
|
| 1271 |
+
"loss": 0.5122,
|
| 1272 |
+
"step": 1770
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 1.7115384615384617,
|
| 1276 |
+
"grad_norm": 2.0570639625612985,
|
| 1277 |
+
"learning_rate": 6.204083721655607e-07,
|
| 1278 |
+
"loss": 0.4928,
|
| 1279 |
+
"step": 1780
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 1.7211538461538463,
|
| 1283 |
+
"grad_norm": 2.103504186819526,
|
| 1284 |
+
"learning_rate": 5.805437845580958e-07,
|
| 1285 |
+
"loss": 0.5203,
|
| 1286 |
+
"step": 1790
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 1.7307692307692308,
|
| 1290 |
+
"grad_norm": 2.197105961623965,
|
| 1291 |
+
"learning_rate": 5.41923845391486e-07,
|
| 1292 |
+
"loss": 0.5124,
|
| 1293 |
+
"step": 1800
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 1.7403846153846154,
|
| 1297 |
+
"grad_norm": 2.1152456056469227,
|
| 1298 |
+
"learning_rate": 5.045594311728708e-07,
|
| 1299 |
+
"loss": 0.4916,
|
| 1300 |
+
"step": 1810
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.75,
|
| 1304 |
+
"grad_norm": 2.0939964989579973,
|
| 1305 |
+
"learning_rate": 4.6846106481675035e-07,
|
| 1306 |
+
"loss": 0.5111,
|
| 1307 |
+
"step": 1820
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.7596153846153846,
|
| 1311 |
+
"grad_norm": 2.4677432405873256,
|
| 1312 |
+
"learning_rate": 4.336389126814311e-07,
|
| 1313 |
+
"loss": 0.5346,
|
| 1314 |
+
"step": 1830
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 1.7692307692307692,
|
| 1318 |
+
"grad_norm": 1.9790895263145134,
|
| 1319 |
+
"learning_rate": 4.001027817058789e-07,
|
| 1320 |
+
"loss": 0.5051,
|
| 1321 |
+
"step": 1840
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 1.7788461538461537,
|
| 1325 |
+
"grad_norm": 2.2327516887468417,
|
| 1326 |
+
"learning_rate": 3.6786211664779583e-07,
|
| 1327 |
+
"loss": 0.5376,
|
| 1328 |
+
"step": 1850
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 1.7884615384615383,
|
| 1332 |
+
"grad_norm": 2.0971604625086777,
|
| 1333 |
+
"learning_rate": 3.369259974236988e-07,
|
| 1334 |
+
"loss": 0.4795,
|
| 1335 |
+
"step": 1860
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.7980769230769231,
|
| 1339 |
+
"grad_norm": 2.124716857194864,
|
| 1340 |
+
"learning_rate": 3.0730313655175647e-07,
|
| 1341 |
+
"loss": 0.5388,
|
| 1342 |
+
"step": 1870
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.8076923076923077,
|
| 1346 |
+
"grad_norm": 2.1095913682765977,
|
| 1347 |
+
"learning_rate": 2.790018766980773e-07,
|
| 1348 |
+
"loss": 0.545,
|
| 1349 |
+
"step": 1880
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.8173076923076923,
|
| 1353 |
+
"grad_norm": 2.0130617931416275,
|
| 1354 |
+
"learning_rate": 2.520301883271797e-07,
|
| 1355 |
+
"loss": 0.5169,
|
| 1356 |
+
"step": 1890
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.8269230769230769,
|
| 1360 |
+
"grad_norm": 2.083745653036941,
|
| 1361 |
+
"learning_rate": 2.2639566745727203e-07,
|
| 1362 |
+
"loss": 0.4928,
|
| 1363 |
+
"step": 1900
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.8365384615384617,
|
| 1367 |
+
"grad_norm": 2.220101162444105,
|
| 1368 |
+
"learning_rate": 2.0210553352098815e-07,
|
| 1369 |
+
"loss": 0.5075,
|
| 1370 |
+
"step": 1910
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 1.8461538461538463,
|
| 1374 |
+
"grad_norm": 2.308065945824805,
|
| 1375 |
+
"learning_rate": 1.7916662733218848e-07,
|
| 1376 |
+
"loss": 0.5116,
|
| 1377 |
+
"step": 1920
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 1.8557692307692308,
|
| 1381 |
+
"grad_norm": 1.9607562596689283,
|
| 1382 |
+
"learning_rate": 1.575854091593837e-07,
|
| 1383 |
+
"loss": 0.5183,
|
| 1384 |
+
"step": 1930
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 1.8653846153846154,
|
| 1388 |
+
"grad_norm": 2.1443103265105634,
|
| 1389 |
+
"learning_rate": 1.3736795690633353e-07,
|
| 1390 |
+
"loss": 0.5123,
|
| 1391 |
+
"step": 1940
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 1.875,
|
| 1395 |
+
"grad_norm": 1.9320175413394869,
|
| 1396 |
+
"learning_rate": 1.185199644003332e-07,
|
| 1397 |
+
"loss": 0.4931,
|
| 1398 |
+
"step": 1950
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 1.8846153846153846,
|
| 1402 |
+
"grad_norm": 2.2160725485465567,
|
| 1403 |
+
"learning_rate": 1.0104673978866164e-07,
|
| 1404 |
+
"loss": 0.513,
|
| 1405 |
+
"step": 1960
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 1.8942307692307692,
|
| 1409 |
+
"grad_norm": 2.2787872837737697,
|
| 1410 |
+
"learning_rate": 8.495320404365348e-08,
|
| 1411 |
+
"loss": 0.5062,
|
| 1412 |
+
"step": 1970
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 1.9038461538461537,
|
| 1416 |
+
"grad_norm": 2.168799751436692,
|
| 1417 |
+
"learning_rate": 7.024388957680705e-08,
|
| 1418 |
+
"loss": 0.5168,
|
| 1419 |
+
"step": 1980
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 1.9134615384615383,
|
| 1423 |
+
"grad_norm": 2.064037817068646,
|
| 1424 |
+
"learning_rate": 5.6922938962329364e-08,
|
| 1425 |
+
"loss": 0.5154,
|
| 1426 |
+
"step": 1990
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 1.9230769230769231,
|
| 1430 |
+
"grad_norm": 2.085831001764275,
|
| 1431 |
+
"learning_rate": 4.499410377045765e-08,
|
| 1432 |
+
"loss": 0.5059,
|
| 1433 |
+
"step": 2000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 1.9230769230769231,
|
| 1437 |
+
"eval_loss": 0.7636520862579346,
|
| 1438 |
+
"eval_runtime": 31.002,
|
| 1439 |
+
"eval_samples_per_second": 59.641,
|
| 1440 |
+
"eval_steps_per_second": 7.483,
|
| 1441 |
+
"step": 2000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.9326923076923077,
|
| 1445 |
+
"grad_norm": 2.0866363192357245,
|
| 1446 |
+
"learning_rate": 3.446074351091566e-08,
|
| 1447 |
+
"loss": 0.535,
|
| 1448 |
+
"step": 2010
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 1.9423076923076923,
|
| 1452 |
+
"grad_norm": 1.9303541701779643,
|
| 1453 |
+
"learning_rate": 2.5325824686772138e-08,
|
| 1454 |
+
"loss": 0.5193,
|
| 1455 |
+
"step": 2020
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 1.9519230769230769,
|
| 1459 |
+
"grad_norm": 2.035058945568711,
|
| 1460 |
+
"learning_rate": 1.7591919958986348e-08,
|
| 1461 |
+
"loss": 0.5242,
|
| 1462 |
+
"step": 2030
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 1.9615384615384617,
|
| 1466 |
+
"grad_norm": 1.906676822404482,
|
| 1467 |
+
"learning_rate": 1.1261207421874309e-08,
|
| 1468 |
+
"loss": 0.4826,
|
| 1469 |
+
"step": 2040
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 1.9711538461538463,
|
| 1473 |
+
"grad_norm": 2.0734277093423965,
|
| 1474 |
+
"learning_rate": 6.335469989692255e-09,
|
| 1475 |
+
"loss": 0.5064,
|
| 1476 |
+
"step": 2050
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 1.9807692307692308,
|
| 1480 |
+
"grad_norm": 2.320470177512831,
|
| 1481 |
+
"learning_rate": 2.816094894513843e-09,
|
| 1482 |
+
"loss": 0.5264,
|
| 1483 |
+
"step": 2060
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 1.9903846153846154,
|
| 1487 |
+
"grad_norm": 2.2494248679357955,
|
| 1488 |
+
"learning_rate": 7.040732955487795e-10,
|
| 1489 |
+
"loss": 0.51,
|
| 1490 |
+
"step": 2070
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 2.0,
|
| 1494 |
+
"grad_norm": 2.2699709074042307,
|
| 1495 |
+
"learning_rate": 0.0,
|
| 1496 |
+
"loss": 0.5211,
|
| 1497 |
+
"step": 2080
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 2.0,
|
| 1501 |
+
"step": 2080,
|
| 1502 |
+
"total_flos": 34318199685120.0,
|
| 1503 |
+
"train_loss": 0.6641153321816371,
|
| 1504 |
+
"train_runtime": 2465.8736,
|
| 1505 |
+
"train_samples_per_second": 13.491,
|
| 1506 |
+
"train_steps_per_second": 0.844
|
| 1507 |
+
}
|
| 1508 |
+
],
|
| 1509 |
+
"logging_steps": 10,
|
| 1510 |
+
"max_steps": 2080,
|
| 1511 |
+
"num_input_tokens_seen": 0,
|
| 1512 |
+
"num_train_epochs": 2,
|
| 1513 |
+
"save_steps": 50000,
|
| 1514 |
+
"stateful_callbacks": {
|
| 1515 |
+
"TrainerControl": {
|
| 1516 |
+
"args": {
|
| 1517 |
+
"should_epoch_stop": false,
|
| 1518 |
+
"should_evaluate": false,
|
| 1519 |
+
"should_log": false,
|
| 1520 |
+
"should_save": true,
|
| 1521 |
+
"should_training_stop": true
|
| 1522 |
+
},
|
| 1523 |
+
"attributes": {}
|
| 1524 |
+
}
|
| 1525 |
+
},
|
| 1526 |
+
"total_flos": 34318199685120.0,
|
| 1527 |
+
"train_batch_size": 1,
|
| 1528 |
+
"trial_name": null,
|
| 1529 |
+
"trial_params": null
|
| 1530 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|