End of training
Browse files- README.md +3 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +1664 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: train_copa_1757340255
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# train_copa_1757340255
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
- Num Input Tokens Seen: 281984
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lntuning
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: train_copa_1757340255
|
|
|
|
| 16 |
|
| 17 |
# train_copa_1757340255
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the copa dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0593
|
| 22 |
- Num Input Tokens Seen: 281984
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.05929639935493469,
|
| 4 |
+
"eval_runtime": 0.6226,
|
| 5 |
+
"eval_samples_per_second": 64.25,
|
| 6 |
+
"eval_steps_per_second": 16.062,
|
| 7 |
+
"num_input_tokens_seen": 281984,
|
| 8 |
+
"total_flos": 1.2698062524776448e+16,
|
| 9 |
+
"train_loss": 0.06748742392027958,
|
| 10 |
+
"train_runtime": 155.7099,
|
| 11 |
+
"train_samples_per_second": 23.12,
|
| 12 |
+
"train_steps_per_second": 5.78
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.05929639935493469,
|
| 4 |
+
"eval_runtime": 0.6226,
|
| 5 |
+
"eval_samples_per_second": 64.25,
|
| 6 |
+
"eval_steps_per_second": 16.062,
|
| 7 |
+
"num_input_tokens_seen": 281984
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"num_input_tokens_seen": 281984,
|
| 4 |
+
"total_flos": 1.2698062524776448e+16,
|
| 5 |
+
"train_loss": 0.06748742392027958,
|
| 6 |
+
"train_runtime": 155.7099,
|
| 7 |
+
"train_samples_per_second": 23.12,
|
| 8 |
+
"train_steps_per_second": 5.78
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1664 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 225,
|
| 3 |
+
"best_metric": 0.05929639935493469,
|
| 4 |
+
"best_model_checkpoint": "saves_stability/lntuning/llama-3-8b-instruct/train_copa_1757340255/checkpoint-225",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 45,
|
| 7 |
+
"global_step": 900,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.05555555555555555,
|
| 14 |
+
"grad_norm": 12.413766860961914,
|
| 15 |
+
"learning_rate": 2.2222222222222225e-06,
|
| 16 |
+
"loss": 0.5401,
|
| 17 |
+
"num_input_tokens_seen": 1632,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.1111111111111111,
|
| 22 |
+
"grad_norm": 10.926953315734863,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 0.589,
|
| 25 |
+
"num_input_tokens_seen": 3232,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.16666666666666666,
|
| 30 |
+
"grad_norm": 11.459014892578125,
|
| 31 |
+
"learning_rate": 7.777777777777777e-06,
|
| 32 |
+
"loss": 0.5822,
|
| 33 |
+
"num_input_tokens_seen": 4832,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.2222222222222222,
|
| 38 |
+
"grad_norm": 9.86060619354248,
|
| 39 |
+
"learning_rate": 1.0555555555555555e-05,
|
| 40 |
+
"loss": 0.6143,
|
| 41 |
+
"num_input_tokens_seen": 6432,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.2777777777777778,
|
| 46 |
+
"grad_norm": 8.448033332824707,
|
| 47 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 48 |
+
"loss": 0.4866,
|
| 49 |
+
"num_input_tokens_seen": 7968,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.3333333333333333,
|
| 54 |
+
"grad_norm": 8.12926197052002,
|
| 55 |
+
"learning_rate": 1.6111111111111115e-05,
|
| 56 |
+
"loss": 0.2775,
|
| 57 |
+
"num_input_tokens_seen": 9504,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3888888888888889,
|
| 62 |
+
"grad_norm": 6.437356472015381,
|
| 63 |
+
"learning_rate": 1.888888888888889e-05,
|
| 64 |
+
"loss": 0.3527,
|
| 65 |
+
"num_input_tokens_seen": 11104,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.4444444444444444,
|
| 70 |
+
"grad_norm": 6.665041923522949,
|
| 71 |
+
"learning_rate": 2.1666666666666667e-05,
|
| 72 |
+
"loss": 0.4136,
|
| 73 |
+
"num_input_tokens_seen": 12704,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.5,
|
| 78 |
+
"grad_norm": 2.534313201904297,
|
| 79 |
+
"learning_rate": 2.4444444444444445e-05,
|
| 80 |
+
"loss": 0.2081,
|
| 81 |
+
"num_input_tokens_seen": 14240,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.5,
|
| 86 |
+
"eval_loss": 0.11573882400989532,
|
| 87 |
+
"eval_runtime": 0.5875,
|
| 88 |
+
"eval_samples_per_second": 68.083,
|
| 89 |
+
"eval_steps_per_second": 17.021,
|
| 90 |
+
"num_input_tokens_seen": 14240,
|
| 91 |
+
"step": 45
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.5555555555555556,
|
| 95 |
+
"grad_norm": 5.011988639831543,
|
| 96 |
+
"learning_rate": 2.7222222222222223e-05,
|
| 97 |
+
"loss": 0.1648,
|
| 98 |
+
"num_input_tokens_seen": 15808,
|
| 99 |
+
"step": 50
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6111111111111112,
|
| 103 |
+
"grad_norm": 0.8455270528793335,
|
| 104 |
+
"learning_rate": 3e-05,
|
| 105 |
+
"loss": 0.1657,
|
| 106 |
+
"num_input_tokens_seen": 17344,
|
| 107 |
+
"step": 55
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6666666666666666,
|
| 111 |
+
"grad_norm": 6.417602062225342,
|
| 112 |
+
"learning_rate": 3.277777777777778e-05,
|
| 113 |
+
"loss": 0.0886,
|
| 114 |
+
"num_input_tokens_seen": 18912,
|
| 115 |
+
"step": 60
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.7222222222222222,
|
| 119 |
+
"grad_norm": 0.35575202107429504,
|
| 120 |
+
"learning_rate": 3.555555555555556e-05,
|
| 121 |
+
"loss": 0.1172,
|
| 122 |
+
"num_input_tokens_seen": 20448,
|
| 123 |
+
"step": 65
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.7777777777777778,
|
| 127 |
+
"grad_norm": 0.35196706652641296,
|
| 128 |
+
"learning_rate": 3.8333333333333334e-05,
|
| 129 |
+
"loss": 0.0435,
|
| 130 |
+
"num_input_tokens_seen": 21984,
|
| 131 |
+
"step": 70
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.8333333333333334,
|
| 135 |
+
"grad_norm": 0.32802021503448486,
|
| 136 |
+
"learning_rate": 4.111111111111111e-05,
|
| 137 |
+
"loss": 0.1677,
|
| 138 |
+
"num_input_tokens_seen": 23552,
|
| 139 |
+
"step": 75
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.8888888888888888,
|
| 143 |
+
"grad_norm": 0.16649259626865387,
|
| 144 |
+
"learning_rate": 4.388888888888889e-05,
|
| 145 |
+
"loss": 0.1363,
|
| 146 |
+
"num_input_tokens_seen": 25120,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.9444444444444444,
|
| 151 |
+
"grad_norm": 0.053206074982881546,
|
| 152 |
+
"learning_rate": 4.666666666666667e-05,
|
| 153 |
+
"loss": 0.0173,
|
| 154 |
+
"num_input_tokens_seen": 26656,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.0,
|
| 159 |
+
"grad_norm": 11.185739517211914,
|
| 160 |
+
"learning_rate": 4.9444444444444446e-05,
|
| 161 |
+
"loss": 0.2591,
|
| 162 |
+
"num_input_tokens_seen": 28192,
|
| 163 |
+
"step": 90
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.0,
|
| 167 |
+
"eval_loss": 0.07012443244457245,
|
| 168 |
+
"eval_runtime": 0.5977,
|
| 169 |
+
"eval_samples_per_second": 66.923,
|
| 170 |
+
"eval_steps_per_second": 16.731,
|
| 171 |
+
"num_input_tokens_seen": 28192,
|
| 172 |
+
"step": 90
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.0555555555555556,
|
| 176 |
+
"grad_norm": 4.218883991241455,
|
| 177 |
+
"learning_rate": 4.9996991493233693e-05,
|
| 178 |
+
"loss": 0.0344,
|
| 179 |
+
"num_input_tokens_seen": 29792,
|
| 180 |
+
"step": 95
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.1111111111111112,
|
| 184 |
+
"grad_norm": 0.030968718230724335,
|
| 185 |
+
"learning_rate": 4.99847706754774e-05,
|
| 186 |
+
"loss": 0.0927,
|
| 187 |
+
"num_input_tokens_seen": 31328,
|
| 188 |
+
"step": 100
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.1666666666666667,
|
| 192 |
+
"grad_norm": 4.61262321472168,
|
| 193 |
+
"learning_rate": 4.9963154107272295e-05,
|
| 194 |
+
"loss": 0.0883,
|
| 195 |
+
"num_input_tokens_seen": 32832,
|
| 196 |
+
"step": 105
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.2222222222222223,
|
| 200 |
+
"grad_norm": 5.351931571960449,
|
| 201 |
+
"learning_rate": 4.993214991772563e-05,
|
| 202 |
+
"loss": 0.0412,
|
| 203 |
+
"num_input_tokens_seen": 34304,
|
| 204 |
+
"step": 110
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.2777777777777777,
|
| 208 |
+
"grad_norm": 0.05530461296439171,
|
| 209 |
+
"learning_rate": 4.989176976624511e-05,
|
| 210 |
+
"loss": 0.1526,
|
| 211 |
+
"num_input_tokens_seen": 35840,
|
| 212 |
+
"step": 115
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.3333333333333333,
|
| 216 |
+
"grad_norm": 5.916304588317871,
|
| 217 |
+
"learning_rate": 4.9842028838154285e-05,
|
| 218 |
+
"loss": 0.2116,
|
| 219 |
+
"num_input_tokens_seen": 37376,
|
| 220 |
+
"step": 120
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.3888888888888888,
|
| 224 |
+
"grad_norm": 0.3439420759677887,
|
| 225 |
+
"learning_rate": 4.978294583898196e-05,
|
| 226 |
+
"loss": 0.0965,
|
| 227 |
+
"num_input_tokens_seen": 38944,
|
| 228 |
+
"step": 125
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.4444444444444444,
|
| 232 |
+
"grad_norm": 0.2820066511631012,
|
| 233 |
+
"learning_rate": 4.971454298742779e-05,
|
| 234 |
+
"loss": 0.0462,
|
| 235 |
+
"num_input_tokens_seen": 40512,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.5,
|
| 240 |
+
"grad_norm": 0.4881448745727539,
|
| 241 |
+
"learning_rate": 4.963684600700679e-05,
|
| 242 |
+
"loss": 0.0367,
|
| 243 |
+
"num_input_tokens_seen": 42080,
|
| 244 |
+
"step": 135
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.5,
|
| 248 |
+
"eval_loss": 0.06827564537525177,
|
| 249 |
+
"eval_runtime": 0.5886,
|
| 250 |
+
"eval_samples_per_second": 67.952,
|
| 251 |
+
"eval_steps_per_second": 16.988,
|
| 252 |
+
"num_input_tokens_seen": 42080,
|
| 253 |
+
"step": 135
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.5555555555555556,
|
| 257 |
+
"grad_norm": 4.441239833831787,
|
| 258 |
+
"learning_rate": 4.9549884116375714e-05,
|
| 259 |
+
"loss": 0.143,
|
| 260 |
+
"num_input_tokens_seen": 43680,
|
| 261 |
+
"step": 140
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.6111111111111112,
|
| 265 |
+
"grad_norm": 0.04426378384232521,
|
| 266 |
+
"learning_rate": 4.9453690018345144e-05,
|
| 267 |
+
"loss": 0.0385,
|
| 268 |
+
"num_input_tokens_seen": 45216,
|
| 269 |
+
"step": 145
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.6666666666666665,
|
| 273 |
+
"grad_norm": 2.143786907196045,
|
| 274 |
+
"learning_rate": 4.934829988758131e-05,
|
| 275 |
+
"loss": 0.192,
|
| 276 |
+
"num_input_tokens_seen": 46720,
|
| 277 |
+
"step": 150
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.7222222222222223,
|
| 281 |
+
"grad_norm": 0.15666843950748444,
|
| 282 |
+
"learning_rate": 4.923375335700223e-05,
|
| 283 |
+
"loss": 0.0042,
|
| 284 |
+
"num_input_tokens_seen": 48288,
|
| 285 |
+
"step": 155
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.7777777777777777,
|
| 289 |
+
"grad_norm": 4.361499786376953,
|
| 290 |
+
"learning_rate": 4.9110093502873476e-05,
|
| 291 |
+
"loss": 0.0937,
|
| 292 |
+
"num_input_tokens_seen": 49888,
|
| 293 |
+
"step": 160
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.8333333333333335,
|
| 297 |
+
"grad_norm": 0.02156808227300644,
|
| 298 |
+
"learning_rate": 4.897736682860885e-05,
|
| 299 |
+
"loss": 0.0234,
|
| 300 |
+
"num_input_tokens_seen": 51424,
|
| 301 |
+
"step": 165
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.8888888888888888,
|
| 305 |
+
"grad_norm": 6.929230690002441,
|
| 306 |
+
"learning_rate": 4.883562324728241e-05,
|
| 307 |
+
"loss": 0.1932,
|
| 308 |
+
"num_input_tokens_seen": 52992,
|
| 309 |
+
"step": 170
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.9444444444444444,
|
| 313 |
+
"grad_norm": 0.6252282857894897,
|
| 314 |
+
"learning_rate": 4.868491606285823e-05,
|
| 315 |
+
"loss": 0.0798,
|
| 316 |
+
"num_input_tokens_seen": 54592,
|
| 317 |
+
"step": 175
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.0,
|
| 321 |
+
"grad_norm": 0.18464066088199615,
|
| 322 |
+
"learning_rate": 4.8525301950144894e-05,
|
| 323 |
+
"loss": 0.1331,
|
| 324 |
+
"num_input_tokens_seen": 56192,
|
| 325 |
+
"step": 180
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 2.0,
|
| 329 |
+
"eval_loss": 0.05967045947909355,
|
| 330 |
+
"eval_runtime": 0.5919,
|
| 331 |
+
"eval_samples_per_second": 67.58,
|
| 332 |
+
"eval_steps_per_second": 16.895,
|
| 333 |
+
"num_input_tokens_seen": 56192,
|
| 334 |
+
"step": 180
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 2.0555555555555554,
|
| 338 |
+
"grad_norm": 1.8507845401763916,
|
| 339 |
+
"learning_rate": 4.835684093348244e-05,
|
| 340 |
+
"loss": 0.0186,
|
| 341 |
+
"num_input_tokens_seen": 57760,
|
| 342 |
+
"step": 185
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.111111111111111,
|
| 346 |
+
"grad_norm": 0.018928587436676025,
|
| 347 |
+
"learning_rate": 4.817959636416969e-05,
|
| 348 |
+
"loss": 0.0333,
|
| 349 |
+
"num_input_tokens_seen": 59264,
|
| 350 |
+
"step": 190
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.1666666666666665,
|
| 354 |
+
"grad_norm": 0.03185804933309555,
|
| 355 |
+
"learning_rate": 4.7993634896640394e-05,
|
| 356 |
+
"loss": 0.0572,
|
| 357 |
+
"num_input_tokens_seen": 60768,
|
| 358 |
+
"step": 195
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.2222222222222223,
|
| 362 |
+
"grad_norm": 6.1525678634643555,
|
| 363 |
+
"learning_rate": 4.779902646339722e-05,
|
| 364 |
+
"loss": 0.1263,
|
| 365 |
+
"num_input_tokens_seen": 62272,
|
| 366 |
+
"step": 200
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 2.2777777777777777,
|
| 370 |
+
"grad_norm": 0.24679124355316162,
|
| 371 |
+
"learning_rate": 4.759584424871302e-05,
|
| 372 |
+
"loss": 0.0479,
|
| 373 |
+
"num_input_tokens_seen": 63808,
|
| 374 |
+
"step": 205
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.3333333333333335,
|
| 378 |
+
"grad_norm": 4.617246150970459,
|
| 379 |
+
"learning_rate": 4.7384164661109176e-05,
|
| 380 |
+
"loss": 0.1704,
|
| 381 |
+
"num_input_tokens_seen": 65344,
|
| 382 |
+
"step": 210
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.388888888888889,
|
| 386 |
+
"grad_norm": 1.2960439920425415,
|
| 387 |
+
"learning_rate": 4.7164067304621536e-05,
|
| 388 |
+
"loss": 0.0061,
|
| 389 |
+
"num_input_tokens_seen": 66912,
|
| 390 |
+
"step": 215
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 2.4444444444444446,
|
| 394 |
+
"grad_norm": 6.159663677215576,
|
| 395 |
+
"learning_rate": 4.693563494886455e-05,
|
| 396 |
+
"loss": 0.1972,
|
| 397 |
+
"num_input_tokens_seen": 68480,
|
| 398 |
+
"step": 220
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.5,
|
| 402 |
+
"grad_norm": 0.7147485017776489,
|
| 403 |
+
"learning_rate": 4.669895349790502e-05,
|
| 404 |
+
"loss": 0.0047,
|
| 405 |
+
"num_input_tokens_seen": 70048,
|
| 406 |
+
"step": 225
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.5,
|
| 410 |
+
"eval_loss": 0.05929639935493469,
|
| 411 |
+
"eval_runtime": 0.5943,
|
| 412 |
+
"eval_samples_per_second": 67.308,
|
| 413 |
+
"eval_steps_per_second": 16.827,
|
| 414 |
+
"num_input_tokens_seen": 70048,
|
| 415 |
+
"step": 225
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.5555555555555554,
|
| 419 |
+
"grad_norm": 0.47322940826416016,
|
| 420 |
+
"learning_rate": 4.645411195795709e-05,
|
| 421 |
+
"loss": 0.0096,
|
| 422 |
+
"num_input_tokens_seen": 71680,
|
| 423 |
+
"step": 230
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.611111111111111,
|
| 427 |
+
"grad_norm": 5.344168663024902,
|
| 428 |
+
"learning_rate": 4.620120240391065e-05,
|
| 429 |
+
"loss": 0.1928,
|
| 430 |
+
"num_input_tokens_seen": 73280,
|
| 431 |
+
"step": 235
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.6666666666666665,
|
| 435 |
+
"grad_norm": 0.012929716147482395,
|
| 436 |
+
"learning_rate": 4.5940319944705736e-05,
|
| 437 |
+
"loss": 0.0027,
|
| 438 |
+
"num_input_tokens_seen": 74752,
|
| 439 |
+
"step": 240
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.7222222222222223,
|
| 443 |
+
"grad_norm": 0.05127738416194916,
|
| 444 |
+
"learning_rate": 4.567156268756594e-05,
|
| 445 |
+
"loss": 0.106,
|
| 446 |
+
"num_input_tokens_seen": 76288,
|
| 447 |
+
"step": 245
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.7777777777777777,
|
| 451 |
+
"grad_norm": 0.03636276721954346,
|
| 452 |
+
"learning_rate": 4.539503170110431e-05,
|
| 453 |
+
"loss": 0.0649,
|
| 454 |
+
"num_input_tokens_seen": 77888,
|
| 455 |
+
"step": 250
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 2.8333333333333335,
|
| 459 |
+
"grad_norm": 0.31299880146980286,
|
| 460 |
+
"learning_rate": 4.5110830977315556e-05,
|
| 461 |
+
"loss": 0.0032,
|
| 462 |
+
"num_input_tokens_seen": 79456,
|
| 463 |
+
"step": 255
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 2.888888888888889,
|
| 467 |
+
"grad_norm": 0.5931472778320312,
|
| 468 |
+
"learning_rate": 4.4819067392468944e-05,
|
| 469 |
+
"loss": 0.0646,
|
| 470 |
+
"num_input_tokens_seen": 81088,
|
| 471 |
+
"step": 260
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.9444444444444446,
|
| 475 |
+
"grad_norm": 0.7903435230255127,
|
| 476 |
+
"learning_rate": 4.4519850666916484e-05,
|
| 477 |
+
"loss": 0.0188,
|
| 478 |
+
"num_input_tokens_seen": 82592,
|
| 479 |
+
"step": 265
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.0,
|
| 483 |
+
"grad_norm": 4.708324909210205,
|
| 484 |
+
"learning_rate": 4.4213293323831585e-05,
|
| 485 |
+
"loss": 0.0918,
|
| 486 |
+
"num_input_tokens_seen": 84192,
|
| 487 |
+
"step": 270
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 3.0,
|
| 491 |
+
"eval_loss": 0.059632014483213425,
|
| 492 |
+
"eval_runtime": 0.5811,
|
| 493 |
+
"eval_samples_per_second": 68.838,
|
| 494 |
+
"eval_steps_per_second": 17.21,
|
| 495 |
+
"num_input_tokens_seen": 84192,
|
| 496 |
+
"step": 270
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 3.0555555555555554,
|
| 500 |
+
"grad_norm": 0.02156168594956398,
|
| 501 |
+
"learning_rate": 4.38995106468937e-05,
|
| 502 |
+
"loss": 0.0337,
|
| 503 |
+
"num_input_tokens_seen": 85696,
|
| 504 |
+
"step": 275
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 3.111111111111111,
|
| 508 |
+
"grad_norm": 1.075160264968872,
|
| 509 |
+
"learning_rate": 4.357862063693486e-05,
|
| 510 |
+
"loss": 0.0222,
|
| 511 |
+
"num_input_tokens_seen": 87296,
|
| 512 |
+
"step": 280
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 3.1666666666666665,
|
| 516 |
+
"grad_norm": 0.09805523604154587,
|
| 517 |
+
"learning_rate": 4.325074396756437e-05,
|
| 518 |
+
"loss": 0.01,
|
| 519 |
+
"num_input_tokens_seen": 88832,
|
| 520 |
+
"step": 285
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 3.2222222222222223,
|
| 524 |
+
"grad_norm": 0.028949294239282608,
|
| 525 |
+
"learning_rate": 4.2916003939788403e-05,
|
| 526 |
+
"loss": 0.0024,
|
| 527 |
+
"num_input_tokens_seen": 90368,
|
| 528 |
+
"step": 290
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 3.2777777777777777,
|
| 532 |
+
"grad_norm": 0.4382733702659607,
|
| 533 |
+
"learning_rate": 4.257452643564155e-05,
|
| 534 |
+
"loss": 0.0078,
|
| 535 |
+
"num_input_tokens_seen": 91968,
|
| 536 |
+
"step": 295
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 3.3333333333333335,
|
| 540 |
+
"grad_norm": 2.2116539478302,
|
| 541 |
+
"learning_rate": 4.22264398708477e-05,
|
| 542 |
+
"loss": 0.0124,
|
| 543 |
+
"num_input_tokens_seen": 93568,
|
| 544 |
+
"step": 300
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 3.388888888888889,
|
| 548 |
+
"grad_norm": 0.011147049255669117,
|
| 549 |
+
"learning_rate": 4.1871875146528195e-05,
|
| 550 |
+
"loss": 0.01,
|
| 551 |
+
"num_input_tokens_seen": 95168,
|
| 552 |
+
"step": 305
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 3.4444444444444446,
|
| 556 |
+
"grad_norm": 0.057261109352111816,
|
| 557 |
+
"learning_rate": 4.1510965599975196e-05,
|
| 558 |
+
"loss": 0.0729,
|
| 559 |
+
"num_input_tokens_seen": 96704,
|
| 560 |
+
"step": 310
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 3.5,
|
| 564 |
+
"grad_norm": 0.028863638639450073,
|
| 565 |
+
"learning_rate": 4.114384695450906e-05,
|
| 566 |
+
"loss": 0.1091,
|
| 567 |
+
"num_input_tokens_seen": 98304,
|
| 568 |
+
"step": 315
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 3.5,
|
| 572 |
+
"eval_loss": 0.06172233074903488,
|
| 573 |
+
"eval_runtime": 0.5816,
|
| 574 |
+
"eval_samples_per_second": 68.77,
|
| 575 |
+
"eval_steps_per_second": 17.193,
|
| 576 |
+
"num_input_tokens_seen": 98304,
|
| 577 |
+
"step": 315
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 3.5555555555555554,
|
| 581 |
+
"grad_norm": 0.021796802058815956,
|
| 582 |
+
"learning_rate": 4.077065726843828e-05,
|
| 583 |
+
"loss": 0.0015,
|
| 584 |
+
"num_input_tokens_seen": 99840,
|
| 585 |
+
"step": 320
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 3.611111111111111,
|
| 589 |
+
"grad_norm": 0.8977848291397095,
|
| 590 |
+
"learning_rate": 4.039153688314145e-05,
|
| 591 |
+
"loss": 0.032,
|
| 592 |
+
"num_input_tokens_seen": 101376,
|
| 593 |
+
"step": 325
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.6666666666666665,
|
| 597 |
+
"grad_norm": 0.11903944611549377,
|
| 598 |
+
"learning_rate": 4.000662837029062e-05,
|
| 599 |
+
"loss": 0.0704,
|
| 600 |
+
"num_input_tokens_seen": 102976,
|
| 601 |
+
"step": 330
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 3.7222222222222223,
|
| 605 |
+
"grad_norm": 3.8452537059783936,
|
| 606 |
+
"learning_rate": 3.961607647823583e-05,
|
| 607 |
+
"loss": 0.0592,
|
| 608 |
+
"num_input_tokens_seen": 104576,
|
| 609 |
+
"step": 335
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.7777777777777777,
|
| 613 |
+
"grad_norm": 1.4291119575500488,
|
| 614 |
+
"learning_rate": 3.9220028077571295e-05,
|
| 615 |
+
"loss": 0.1871,
|
| 616 |
+
"num_input_tokens_seen": 106112,
|
| 617 |
+
"step": 340
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 3.8333333333333335,
|
| 621 |
+
"grad_norm": 0.04512714594602585,
|
| 622 |
+
"learning_rate": 3.881863210590332e-05,
|
| 623 |
+
"loss": 0.1058,
|
| 624 |
+
"num_input_tokens_seen": 107712,
|
| 625 |
+
"step": 345
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 3.888888888888889,
|
| 629 |
+
"grad_norm": 0.29252851009368896,
|
| 630 |
+
"learning_rate": 3.841203951184095e-05,
|
| 631 |
+
"loss": 0.1107,
|
| 632 |
+
"num_input_tokens_seen": 109344,
|
| 633 |
+
"step": 350
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.9444444444444446,
|
| 637 |
+
"grad_norm": 6.142524719238281,
|
| 638 |
+
"learning_rate": 3.8000403198230387e-05,
|
| 639 |
+
"loss": 0.1373,
|
| 640 |
+
"num_input_tokens_seen": 110944,
|
| 641 |
+
"step": 355
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 4.0,
|
| 645 |
+
"grad_norm": 0.19722937047481537,
|
| 646 |
+
"learning_rate": 3.75838779646545e-05,
|
| 647 |
+
"loss": 0.0053,
|
| 648 |
+
"num_input_tokens_seen": 112544,
|
| 649 |
+
"step": 360
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 4.0,
|
| 653 |
+
"eval_loss": 0.062212228775024414,
|
| 654 |
+
"eval_runtime": 0.5812,
|
| 655 |
+
"eval_samples_per_second": 68.82,
|
| 656 |
+
"eval_steps_per_second": 17.205,
|
| 657 |
+
"num_input_tokens_seen": 112544,
|
| 658 |
+
"step": 360
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 4.055555555555555,
|
| 662 |
+
"grad_norm": 0.09612889587879181,
|
| 663 |
+
"learning_rate": 3.7162620449219e-05,
|
| 664 |
+
"loss": 0.0397,
|
| 665 |
+
"num_input_tokens_seen": 114144,
|
| 666 |
+
"step": 365
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 4.111111111111111,
|
| 670 |
+
"grad_norm": 2.852125406265259,
|
| 671 |
+
"learning_rate": 3.673678906964727e-05,
|
| 672 |
+
"loss": 0.0304,
|
| 673 |
+
"num_input_tokens_seen": 115712,
|
| 674 |
+
"step": 370
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 4.166666666666667,
|
| 678 |
+
"grad_norm": 0.6219462156295776,
|
| 679 |
+
"learning_rate": 3.630654396370594e-05,
|
| 680 |
+
"loss": 0.1283,
|
| 681 |
+
"num_input_tokens_seen": 117312,
|
| 682 |
+
"step": 375
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 4.222222222222222,
|
| 686 |
+
"grad_norm": 0.216511532664299,
|
| 687 |
+
"learning_rate": 3.5872046928983626e-05,
|
| 688 |
+
"loss": 0.0795,
|
| 689 |
+
"num_input_tokens_seen": 118848,
|
| 690 |
+
"step": 380
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 4.277777777777778,
|
| 694 |
+
"grad_norm": 2.406106472015381,
|
| 695 |
+
"learning_rate": 3.543346136204545e-05,
|
| 696 |
+
"loss": 0.0155,
|
| 697 |
+
"num_input_tokens_seen": 120416,
|
| 698 |
+
"step": 385
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 4.333333333333333,
|
| 702 |
+
"grad_norm": 0.8766459822654724,
|
| 703 |
+
"learning_rate": 3.499095219698631e-05,
|
| 704 |
+
"loss": 0.0599,
|
| 705 |
+
"num_input_tokens_seen": 122016,
|
| 706 |
+
"step": 390
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 4.388888888888889,
|
| 710 |
+
"grad_norm": 6.439513683319092,
|
| 711 |
+
"learning_rate": 3.454468584340588e-05,
|
| 712 |
+
"loss": 0.0956,
|
| 713 |
+
"num_input_tokens_seen": 123584,
|
| 714 |
+
"step": 395
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 4.444444444444445,
|
| 718 |
+
"grad_norm": 0.011177383363246918,
|
| 719 |
+
"learning_rate": 3.409483012382879e-05,
|
| 720 |
+
"loss": 0.0031,
|
| 721 |
+
"num_input_tokens_seen": 125216,
|
| 722 |
+
"step": 400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 4.5,
|
| 726 |
+
"grad_norm": 0.6205223798751831,
|
| 727 |
+
"learning_rate": 3.364155421059342e-05,
|
| 728 |
+
"loss": 0.0101,
|
| 729 |
+
"num_input_tokens_seen": 126784,
|
| 730 |
+
"step": 405
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 4.5,
|
| 734 |
+
"eval_loss": 0.06303347647190094,
|
| 735 |
+
"eval_runtime": 0.5821,
|
| 736 |
+
"eval_samples_per_second": 68.711,
|
| 737 |
+
"eval_steps_per_second": 17.178,
|
| 738 |
+
"num_input_tokens_seen": 126784,
|
| 739 |
+
"step": 405
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 4.555555555555555,
|
| 743 |
+
"grad_norm": 0.12966503202915192,
|
| 744 |
+
"learning_rate": 3.318502856223311e-05,
|
| 745 |
+
"loss": 0.0062,
|
| 746 |
+
"num_input_tokens_seen": 128352,
|
| 747 |
+
"step": 410
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 4.611111111111111,
|
| 751 |
+
"grad_norm": 0.27338626980781555,
|
| 752 |
+
"learning_rate": 3.272542485937369e-05,
|
| 753 |
+
"loss": 0.0062,
|
| 754 |
+
"num_input_tokens_seen": 129920,
|
| 755 |
+
"step": 415
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 4.666666666666667,
|
| 759 |
+
"grad_norm": 0.04085496813058853,
|
| 760 |
+
"learning_rate": 3.2262915940171376e-05,
|
| 761 |
+
"loss": 0.0027,
|
| 762 |
+
"num_input_tokens_seen": 131488,
|
| 763 |
+
"step": 420
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 4.722222222222222,
|
| 767 |
+
"grad_norm": 0.0658736526966095,
|
| 768 |
+
"learning_rate": 3.1797675735315455e-05,
|
| 769 |
+
"loss": 0.0676,
|
| 770 |
+
"num_input_tokens_seen": 133056,
|
| 771 |
+
"step": 425
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 4.777777777777778,
|
| 775 |
+
"grad_norm": 0.40652352571487427,
|
| 776 |
+
"learning_rate": 3.132987920262005e-05,
|
| 777 |
+
"loss": 0.038,
|
| 778 |
+
"num_input_tokens_seen": 134624,
|
| 779 |
+
"step": 430
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 4.833333333333333,
|
| 783 |
+
"grad_norm": 0.22422264516353607,
|
| 784 |
+
"learning_rate": 3.085970226122962e-05,
|
| 785 |
+
"loss": 0.0227,
|
| 786 |
+
"num_input_tokens_seen": 136192,
|
| 787 |
+
"step": 435
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 4.888888888888889,
|
| 791 |
+
"grad_norm": 0.009336884133517742,
|
| 792 |
+
"learning_rate": 3.0387321725463e-05,
|
| 793 |
+
"loss": 0.0115,
|
| 794 |
+
"num_input_tokens_seen": 137824,
|
| 795 |
+
"step": 440
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 4.944444444444445,
|
| 799 |
+
"grad_norm": 0.28929874300956726,
|
| 800 |
+
"learning_rate": 2.9912915238320754e-05,
|
| 801 |
+
"loss": 0.0154,
|
| 802 |
+
"num_input_tokens_seen": 139392,
|
| 803 |
+
"step": 445
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 5.0,
|
| 807 |
+
"grad_norm": 0.1447066068649292,
|
| 808 |
+
"learning_rate": 2.9436661204680882e-05,
|
| 809 |
+
"loss": 0.0808,
|
| 810 |
+
"num_input_tokens_seen": 140960,
|
| 811 |
+
"step": 450
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 5.0,
|
| 815 |
+
"eval_loss": 0.06195898726582527,
|
| 816 |
+
"eval_runtime": 0.5844,
|
| 817 |
+
"eval_samples_per_second": 68.443,
|
| 818 |
+
"eval_steps_per_second": 17.111,
|
| 819 |
+
"num_input_tokens_seen": 140960,
|
| 820 |
+
"step": 450
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 5.055555555555555,
|
| 824 |
+
"grad_norm": 0.04051095247268677,
|
| 825 |
+
"learning_rate": 2.8958738724208072e-05,
|
| 826 |
+
"loss": 0.0635,
|
| 827 |
+
"num_input_tokens_seen": 142560,
|
| 828 |
+
"step": 455
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 5.111111111111111,
|
| 832 |
+
"grad_norm": 0.023053204640746117,
|
| 833 |
+
"learning_rate": 2.8479327524001636e-05,
|
| 834 |
+
"loss": 0.0316,
|
| 835 |
+
"num_input_tokens_seen": 144096,
|
| 836 |
+
"step": 460
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 5.166666666666667,
|
| 840 |
+
"grad_norm": 0.23852871358394623,
|
| 841 |
+
"learning_rate": 2.7998607891007495e-05,
|
| 842 |
+
"loss": 0.0049,
|
| 843 |
+
"num_input_tokens_seen": 145632,
|
| 844 |
+
"step": 465
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 5.222222222222222,
|
| 848 |
+
"grad_norm": 2.9483938217163086,
|
| 849 |
+
"learning_rate": 2.7516760604219617e-05,
|
| 850 |
+
"loss": 0.0206,
|
| 851 |
+
"num_input_tokens_seen": 147232,
|
| 852 |
+
"step": 470
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 5.277777777777778,
|
| 856 |
+
"grad_norm": 0.058457694947719574,
|
| 857 |
+
"learning_rate": 2.7033966866696457e-05,
|
| 858 |
+
"loss": 0.0586,
|
| 859 |
+
"num_input_tokens_seen": 148864,
|
| 860 |
+
"step": 475
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 5.333333333333333,
|
| 864 |
+
"grad_norm": 0.023417210206389427,
|
| 865 |
+
"learning_rate": 2.6550408237417885e-05,
|
| 866 |
+
"loss": 0.003,
|
| 867 |
+
"num_input_tokens_seen": 150432,
|
| 868 |
+
"step": 480
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.388888888888889,
|
| 872 |
+
"grad_norm": 0.29810523986816406,
|
| 873 |
+
"learning_rate": 2.6066266563008267e-05,
|
| 874 |
+
"loss": 0.0771,
|
| 875 |
+
"num_input_tokens_seen": 152032,
|
| 876 |
+
"step": 485
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 5.444444444444445,
|
| 880 |
+
"grad_norm": 0.020725004374980927,
|
| 881 |
+
"learning_rate": 2.5581723909351406e-05,
|
| 882 |
+
"loss": 0.025,
|
| 883 |
+
"num_input_tokens_seen": 153600,
|
| 884 |
+
"step": 490
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 5.5,
|
| 888 |
+
"grad_norm": 0.4676690399646759,
|
| 889 |
+
"learning_rate": 2.5096962493123012e-05,
|
| 890 |
+
"loss": 0.0104,
|
| 891 |
+
"num_input_tokens_seen": 155200,
|
| 892 |
+
"step": 495
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 5.5,
|
| 896 |
+
"eval_loss": 0.0627414733171463,
|
| 897 |
+
"eval_runtime": 0.5812,
|
| 898 |
+
"eval_samples_per_second": 68.818,
|
| 899 |
+
"eval_steps_per_second": 17.204,
|
| 900 |
+
"num_input_tokens_seen": 155200,
|
| 901 |
+
"step": 495
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 5.555555555555555,
|
| 905 |
+
"grad_norm": 0.01173495128750801,
|
| 906 |
+
"learning_rate": 2.461216461326642e-05,
|
| 907 |
+
"loss": 0.0801,
|
| 908 |
+
"num_input_tokens_seen": 156800,
|
| 909 |
+
"step": 500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 5.611111111111111,
|
| 913 |
+
"grad_norm": 0.007062096148729324,
|
| 914 |
+
"learning_rate": 2.4127512582437485e-05,
|
| 915 |
+
"loss": 0.0009,
|
| 916 |
+
"num_input_tokens_seen": 158336,
|
| 917 |
+
"step": 505
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 5.666666666666667,
|
| 921 |
+
"grad_norm": 0.04896526783704758,
|
| 922 |
+
"learning_rate": 2.364318865844416e-05,
|
| 923 |
+
"loss": 0.0027,
|
| 924 |
+
"num_input_tokens_seen": 159904,
|
| 925 |
+
"step": 510
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 5.722222222222222,
|
| 929 |
+
"grad_norm": 6.96190071105957,
|
| 930 |
+
"learning_rate": 2.3159374975706884e-05,
|
| 931 |
+
"loss": 0.0464,
|
| 932 |
+
"num_input_tokens_seen": 161504,
|
| 933 |
+
"step": 515
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 5.777777777777778,
|
| 937 |
+
"grad_norm": 0.008594535291194916,
|
| 938 |
+
"learning_rate": 2.2676253476765196e-05,
|
| 939 |
+
"loss": 0.0016,
|
| 940 |
+
"num_input_tokens_seen": 163008,
|
| 941 |
+
"step": 520
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 5.833333333333333,
|
| 945 |
+
"grad_norm": 0.40661683678627014,
|
| 946 |
+
"learning_rate": 2.2194005843856636e-05,
|
| 947 |
+
"loss": 0.0174,
|
| 948 |
+
"num_input_tokens_seen": 164512,
|
| 949 |
+
"step": 525
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 5.888888888888889,
|
| 953 |
+
"grad_norm": 0.24199238419532776,
|
| 954 |
+
"learning_rate": 2.1712813430593436e-05,
|
| 955 |
+
"loss": 0.1174,
|
| 956 |
+
"num_input_tokens_seen": 166080,
|
| 957 |
+
"step": 530
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 5.944444444444445,
|
| 961 |
+
"grad_norm": 0.025658387690782547,
|
| 962 |
+
"learning_rate": 2.1232857193762924e-05,
|
| 963 |
+
"loss": 0.0317,
|
| 964 |
+
"num_input_tokens_seen": 167680,
|
| 965 |
+
"step": 535
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 6.0,
|
| 969 |
+
"grad_norm": 0.006939691957086325,
|
| 970 |
+
"learning_rate": 2.0754317625276983e-05,
|
| 971 |
+
"loss": 0.0012,
|
| 972 |
+
"num_input_tokens_seen": 169216,
|
| 973 |
+
"step": 540
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 6.0,
|
| 977 |
+
"eval_loss": 0.06366995722055435,
|
| 978 |
+
"eval_runtime": 0.5877,
|
| 979 |
+
"eval_samples_per_second": 68.063,
|
| 980 |
+
"eval_steps_per_second": 17.016,
|
| 981 |
+
"num_input_tokens_seen": 169216,
|
| 982 |
+
"step": 540
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 6.055555555555555,
|
| 986 |
+
"grad_norm": 0.017029551789164543,
|
| 987 |
+
"learning_rate": 2.02773746842965e-05,
|
| 988 |
+
"loss": 0.0027,
|
| 989 |
+
"num_input_tokens_seen": 170784,
|
| 990 |
+
"step": 545
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 6.111111111111111,
|
| 994 |
+
"grad_norm": 4.089417457580566,
|
| 995 |
+
"learning_rate": 1.980220772955602e-05,
|
| 996 |
+
"loss": 0.0248,
|
| 997 |
+
"num_input_tokens_seen": 172352,
|
| 998 |
+
"step": 550
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 6.166666666666667,
|
| 1002 |
+
"grad_norm": 0.15079845488071442,
|
| 1003 |
+
"learning_rate": 1.932899545191433e-05,
|
| 1004 |
+
"loss": 0.002,
|
| 1005 |
+
"num_input_tokens_seen": 173920,
|
| 1006 |
+
"step": 555
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 6.222222222222222,
|
| 1010 |
+
"grad_norm": 0.006432878784835339,
|
| 1011 |
+
"learning_rate": 1.8857915807156092e-05,
|
| 1012 |
+
"loss": 0.0032,
|
| 1013 |
+
"num_input_tokens_seen": 175488,
|
| 1014 |
+
"step": 560
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 6.277777777777778,
|
| 1018 |
+
"grad_norm": 2.5554280281066895,
|
| 1019 |
+
"learning_rate": 1.838914594906995e-05,
|
| 1020 |
+
"loss": 0.0127,
|
| 1021 |
+
"num_input_tokens_seen": 177024,
|
| 1022 |
+
"step": 565
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 6.333333333333333,
|
| 1026 |
+
"grad_norm": 0.41308820247650146,
|
| 1027 |
+
"learning_rate": 1.792286216282824e-05,
|
| 1028 |
+
"loss": 0.025,
|
| 1029 |
+
"num_input_tokens_seen": 178624,
|
| 1030 |
+
"step": 570
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 6.388888888888889,
|
| 1034 |
+
"grad_norm": 0.43808722496032715,
|
| 1035 |
+
"learning_rate": 1.7459239798693364e-05,
|
| 1036 |
+
"loss": 0.1628,
|
| 1037 |
+
"num_input_tokens_seen": 180096,
|
| 1038 |
+
"step": 575
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 6.444444444444445,
|
| 1042 |
+
"grad_norm": 2.488274335861206,
|
| 1043 |
+
"learning_rate": 1.699845320607571e-05,
|
| 1044 |
+
"loss": 0.0157,
|
| 1045 |
+
"num_input_tokens_seen": 181600,
|
| 1046 |
+
"step": 580
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 6.5,
|
| 1050 |
+
"grad_norm": 1.1927860975265503,
|
| 1051 |
+
"learning_rate": 1.6540675667967974e-05,
|
| 1052 |
+
"loss": 0.0056,
|
| 1053 |
+
"num_input_tokens_seen": 183232,
|
| 1054 |
+
"step": 585
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 6.5,
|
| 1058 |
+
"eval_loss": 0.06768044084310532,
|
| 1059 |
+
"eval_runtime": 0.5879,
|
| 1060 |
+
"eval_samples_per_second": 68.036,
|
| 1061 |
+
"eval_steps_per_second": 17.009,
|
| 1062 |
+
"num_input_tokens_seen": 183232,
|
| 1063 |
+
"step": 585
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 6.555555555555555,
|
| 1067 |
+
"grad_norm": 0.0071192653849720955,
|
| 1068 |
+
"learning_rate": 1.60860793357805e-05,
|
| 1069 |
+
"loss": 0.0707,
|
| 1070 |
+
"num_input_tokens_seen": 184768,
|
| 1071 |
+
"step": 590
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 6.611111111111111,
|
| 1075 |
+
"grad_norm": 0.004635022487491369,
|
| 1076 |
+
"learning_rate": 1.56348351646022e-05,
|
| 1077 |
+
"loss": 0.0261,
|
| 1078 |
+
"num_input_tokens_seen": 186208,
|
| 1079 |
+
"step": 595
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 6.666666666666667,
|
| 1083 |
+
"grad_norm": 0.007054431829601526,
|
| 1084 |
+
"learning_rate": 1.5187112848911323e-05,
|
| 1085 |
+
"loss": 0.0091,
|
| 1086 |
+
"num_input_tokens_seen": 187776,
|
| 1087 |
+
"step": 600
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 6.722222222222222,
|
| 1091 |
+
"grad_norm": 0.7310401797294617,
|
| 1092 |
+
"learning_rate": 1.47430807587603e-05,
|
| 1093 |
+
"loss": 0.0049,
|
| 1094 |
+
"num_input_tokens_seen": 189344,
|
| 1095 |
+
"step": 605
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 6.777777777777778,
|
| 1099 |
+
"grad_norm": 0.06609062105417252,
|
| 1100 |
+
"learning_rate": 1.430290587645865e-05,
|
| 1101 |
+
"loss": 0.0027,
|
| 1102 |
+
"num_input_tokens_seen": 190912,
|
| 1103 |
+
"step": 610
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 6.833333333333333,
|
| 1107 |
+
"grad_norm": 0.014985096640884876,
|
| 1108 |
+
"learning_rate": 1.3866753733777765e-05,
|
| 1109 |
+
"loss": 0.0607,
|
| 1110 |
+
"num_input_tokens_seen": 192512,
|
| 1111 |
+
"step": 615
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 6.888888888888889,
|
| 1115 |
+
"grad_norm": 0.010730482637882233,
|
| 1116 |
+
"learning_rate": 1.343478834970121e-05,
|
| 1117 |
+
"loss": 0.0018,
|
| 1118 |
+
"num_input_tokens_seen": 194112,
|
| 1119 |
+
"step": 620
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 6.944444444444445,
|
| 1123 |
+
"grad_norm": 0.30494803190231323,
|
| 1124 |
+
"learning_rate": 1.3007172168743854e-05,
|
| 1125 |
+
"loss": 0.0662,
|
| 1126 |
+
"num_input_tokens_seen": 195712,
|
| 1127 |
+
"step": 625
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 7.0,
|
| 1131 |
+
"grad_norm": 0.06056293845176697,
|
| 1132 |
+
"learning_rate": 1.2584065999863102e-05,
|
| 1133 |
+
"loss": 0.0014,
|
| 1134 |
+
"num_input_tokens_seen": 197248,
|
| 1135 |
+
"step": 630
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 7.0,
|
| 1139 |
+
"eval_loss": 0.07016871124505997,
|
| 1140 |
+
"eval_runtime": 0.5849,
|
| 1141 |
+
"eval_samples_per_second": 68.386,
|
| 1142 |
+
"eval_steps_per_second": 17.096,
|
| 1143 |
+
"num_input_tokens_seen": 197248,
|
| 1144 |
+
"step": 630
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 7.055555555555555,
|
| 1148 |
+
"grad_norm": 0.03782691806554794,
|
| 1149 |
+
"learning_rate": 1.2165628955985314e-05,
|
| 1150 |
+
"loss": 0.0026,
|
| 1151 |
+
"num_input_tokens_seen": 198784,
|
| 1152 |
+
"step": 635
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 7.111111111111111,
|
| 1156 |
+
"grad_norm": 4.731870651245117,
|
| 1157 |
+
"learning_rate": 1.175201839416988e-05,
|
| 1158 |
+
"loss": 0.0685,
|
| 1159 |
+
"num_input_tokens_seen": 200352,
|
| 1160 |
+
"step": 640
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 7.166666666666667,
|
| 1164 |
+
"grad_norm": 0.1799234002828598,
|
| 1165 |
+
"learning_rate": 1.1343389856433658e-05,
|
| 1166 |
+
"loss": 0.0026,
|
| 1167 |
+
"num_input_tokens_seen": 201856,
|
| 1168 |
+
"step": 645
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 7.222222222222222,
|
| 1172 |
+
"grad_norm": 0.010190362110733986,
|
| 1173 |
+
"learning_rate": 1.0939897011258001e-05,
|
| 1174 |
+
"loss": 0.0256,
|
| 1175 |
+
"num_input_tokens_seen": 203488,
|
| 1176 |
+
"step": 650
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 7.277777777777778,
|
| 1180 |
+
"grad_norm": 0.007492389529943466,
|
| 1181 |
+
"learning_rate": 1.0541691595800337e-05,
|
| 1182 |
+
"loss": 0.0021,
|
| 1183 |
+
"num_input_tokens_seen": 205056,
|
| 1184 |
+
"step": 655
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 7.333333333333333,
|
| 1188 |
+
"grad_norm": 0.06244223192334175,
|
| 1189 |
+
"learning_rate": 1.0148923358832022e-05,
|
| 1190 |
+
"loss": 0.0582,
|
| 1191 |
+
"num_input_tokens_seen": 206656,
|
| 1192 |
+
"step": 660
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 7.388888888888889,
|
| 1196 |
+
"grad_norm": 0.04081568121910095,
|
| 1197 |
+
"learning_rate": 9.761740004423927e-06,
|
| 1198 |
+
"loss": 0.0057,
|
| 1199 |
+
"num_input_tokens_seen": 208288,
|
| 1200 |
+
"step": 665
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 7.444444444444445,
|
| 1204 |
+
"grad_norm": 7.480345726013184,
|
| 1205 |
+
"learning_rate": 9.380287136401e-06,
|
| 1206 |
+
"loss": 0.096,
|
| 1207 |
+
"num_input_tokens_seen": 209856,
|
| 1208 |
+
"step": 670
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 7.5,
|
| 1212 |
+
"grad_norm": 0.7831881046295166,
|
| 1213 |
+
"learning_rate": 9.00470820358663e-06,
|
| 1214 |
+
"loss": 0.0042,
|
| 1215 |
+
"num_input_tokens_seen": 211424,
|
| 1216 |
+
"step": 675
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 7.5,
|
| 1220 |
+
"eval_loss": 0.06858132034540176,
|
| 1221 |
+
"eval_runtime": 0.5834,
|
| 1222 |
+
"eval_samples_per_second": 68.565,
|
| 1223 |
+
"eval_steps_per_second": 17.141,
|
| 1224 |
+
"num_input_tokens_seen": 211424,
|
| 1225 |
+
"step": 675
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 7.555555555555555,
|
| 1229 |
+
"grad_norm": 0.01823975145816803,
|
| 1230 |
+
"learning_rate": 8.635144445857406e-06,
|
| 1231 |
+
"loss": 0.0013,
|
| 1232 |
+
"num_input_tokens_seen": 212928,
|
| 1233 |
+
"step": 680
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 7.611111111111111,
|
| 1237 |
+
"grad_norm": 0.018310269340872765,
|
| 1238 |
+
"learning_rate": 8.271734841028553e-06,
|
| 1239 |
+
"loss": 0.0025,
|
| 1240 |
+
"num_input_tokens_seen": 214496,
|
| 1241 |
+
"step": 685
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 7.666666666666667,
|
| 1245 |
+
"grad_norm": 0.01911945454776287,
|
| 1246 |
+
"learning_rate": 7.914616052590071e-06,
|
| 1247 |
+
"loss": 0.0016,
|
| 1248 |
+
"num_input_tokens_seen": 216064,
|
| 1249 |
+
"step": 690
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 7.722222222222222,
|
| 1253 |
+
"grad_norm": 0.017379632219672203,
|
| 1254 |
+
"learning_rate": 7.563922378313218e-06,
|
| 1255 |
+
"loss": 0.0439,
|
| 1256 |
+
"num_input_tokens_seen": 217600,
|
| 1257 |
+
"step": 695
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 7.777777777777778,
|
| 1261 |
+
"grad_norm": 0.5448654890060425,
|
| 1262 |
+
"learning_rate": 7.219785699746573e-06,
|
| 1263 |
+
"loss": 0.0035,
|
| 1264 |
+
"num_input_tokens_seen": 219136,
|
| 1265 |
+
"step": 700
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 7.833333333333333,
|
| 1269 |
+
"grad_norm": 0.15058286488056183,
|
| 1270 |
+
"learning_rate": 6.882335432620779e-06,
|
| 1271 |
+
"loss": 0.0106,
|
| 1272 |
+
"num_input_tokens_seen": 220704,
|
| 1273 |
+
"step": 705
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 7.888888888888889,
|
| 1277 |
+
"grad_norm": 0.2723049223423004,
|
| 1278 |
+
"learning_rate": 6.55169847818059e-06,
|
| 1279 |
+
"loss": 0.0235,
|
| 1280 |
+
"num_input_tokens_seen": 222272,
|
| 1281 |
+
"step": 710
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 7.944444444444445,
|
| 1285 |
+
"grad_norm": 0.11821077018976212,
|
| 1286 |
+
"learning_rate": 6.22799917546252e-06,
|
| 1287 |
+
"loss": 0.0026,
|
| 1288 |
+
"num_input_tokens_seen": 223872,
|
| 1289 |
+
"step": 715
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 8.0,
|
| 1293 |
+
"grad_norm": 2.420344352722168,
|
| 1294 |
+
"learning_rate": 5.9113592545359945e-06,
|
| 1295 |
+
"loss": 0.0692,
|
| 1296 |
+
"num_input_tokens_seen": 225440,
|
| 1297 |
+
"step": 720
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 8.0,
|
| 1301 |
+
"eval_loss": 0.06697994470596313,
|
| 1302 |
+
"eval_runtime": 0.5855,
|
| 1303 |
+
"eval_samples_per_second": 68.323,
|
| 1304 |
+
"eval_steps_per_second": 17.081,
|
| 1305 |
+
"num_input_tokens_seen": 225440,
|
| 1306 |
+
"step": 720
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 8.055555555555555,
|
| 1310 |
+
"grad_norm": 1.6140986680984497,
|
| 1311 |
+
"learning_rate": 5.601897790725643e-06,
|
| 1312 |
+
"loss": 0.0069,
|
| 1313 |
+
"num_input_tokens_seen": 226912,
|
| 1314 |
+
"step": 725
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 8.11111111111111,
|
| 1318 |
+
"grad_norm": 0.10575804114341736,
|
| 1319 |
+
"learning_rate": 5.299731159831953e-06,
|
| 1320 |
+
"loss": 0.0014,
|
| 1321 |
+
"num_input_tokens_seen": 228512,
|
| 1322 |
+
"step": 730
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 8.166666666666666,
|
| 1326 |
+
"grad_norm": 0.4418221414089203,
|
| 1327 |
+
"learning_rate": 5.004972994367102e-06,
|
| 1328 |
+
"loss": 0.05,
|
| 1329 |
+
"num_input_tokens_seen": 230080,
|
| 1330 |
+
"step": 735
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 8.222222222222221,
|
| 1334 |
+
"grad_norm": 0.006636911537498236,
|
| 1335 |
+
"learning_rate": 4.7177341408224e-06,
|
| 1336 |
+
"loss": 0.0009,
|
| 1337 |
+
"num_input_tokens_seen": 231616,
|
| 1338 |
+
"step": 740
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 8.277777777777779,
|
| 1342 |
+
"grad_norm": 0.005838257260620594,
|
| 1343 |
+
"learning_rate": 4.438122617983443e-06,
|
| 1344 |
+
"loss": 0.0232,
|
| 1345 |
+
"num_input_tokens_seen": 233184,
|
| 1346 |
+
"step": 745
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 8.333333333333334,
|
| 1350 |
+
"grad_norm": 0.16078728437423706,
|
| 1351 |
+
"learning_rate": 4.166243576308712e-06,
|
| 1352 |
+
"loss": 0.0033,
|
| 1353 |
+
"num_input_tokens_seen": 234688,
|
| 1354 |
+
"step": 750
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 8.38888888888889,
|
| 1358 |
+
"grad_norm": 0.15908759832382202,
|
| 1359 |
+
"learning_rate": 3.9021992583867325e-06,
|
| 1360 |
+
"loss": 0.1145,
|
| 1361 |
+
"num_input_tokens_seen": 236256,
|
| 1362 |
+
"step": 755
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 8.444444444444445,
|
| 1366 |
+
"grad_norm": 0.19336074590682983,
|
| 1367 |
+
"learning_rate": 3.6460889604868626e-06,
|
| 1368 |
+
"loss": 0.0028,
|
| 1369 |
+
"num_input_tokens_seen": 237856,
|
| 1370 |
+
"step": 760
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 8.5,
|
| 1374 |
+
"grad_norm": 0.05375547334551811,
|
| 1375 |
+
"learning_rate": 3.398008995217988e-06,
|
| 1376 |
+
"loss": 0.0005,
|
| 1377 |
+
"num_input_tokens_seen": 239392,
|
| 1378 |
+
"step": 765
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 8.5,
|
| 1382 |
+
"eval_loss": 0.06791551411151886,
|
| 1383 |
+
"eval_runtime": 0.5839,
|
| 1384 |
+
"eval_samples_per_second": 68.503,
|
| 1385 |
+
"eval_steps_per_second": 17.126,
|
| 1386 |
+
"num_input_tokens_seen": 239392,
|
| 1387 |
+
"step": 765
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 8.555555555555555,
|
| 1391 |
+
"grad_norm": 0.12801700830459595,
|
| 1392 |
+
"learning_rate": 3.158052655309332e-06,
|
| 1393 |
+
"loss": 0.0756,
|
| 1394 |
+
"num_input_tokens_seen": 240960,
|
| 1395 |
+
"step": 770
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 8.61111111111111,
|
| 1399 |
+
"grad_norm": 0.013426029123365879,
|
| 1400 |
+
"learning_rate": 2.9263101785268254e-06,
|
| 1401 |
+
"loss": 0.0063,
|
| 1402 |
+
"num_input_tokens_seen": 242496,
|
| 1403 |
+
"step": 775
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 8.666666666666666,
|
| 1407 |
+
"grad_norm": 1.9325268268585205,
|
| 1408 |
+
"learning_rate": 2.7028687137384267e-06,
|
| 1409 |
+
"loss": 0.0073,
|
| 1410 |
+
"num_input_tokens_seen": 244064,
|
| 1411 |
+
"step": 780
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 8.722222222222221,
|
| 1415 |
+
"grad_norm": 0.0063558281399309635,
|
| 1416 |
+
"learning_rate": 2.487812288140945e-06,
|
| 1417 |
+
"loss": 0.0017,
|
| 1418 |
+
"num_input_tokens_seen": 245664,
|
| 1419 |
+
"step": 785
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 8.777777777777779,
|
| 1423 |
+
"grad_norm": 0.04126088321208954,
|
| 1424 |
+
"learning_rate": 2.281221775660894e-06,
|
| 1425 |
+
"loss": 0.0015,
|
| 1426 |
+
"num_input_tokens_seen": 247264,
|
| 1427 |
+
"step": 790
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 8.833333333333334,
|
| 1431 |
+
"grad_norm": 0.07367562502622604,
|
| 1432 |
+
"learning_rate": 2.0831748665410765e-06,
|
| 1433 |
+
"loss": 0.015,
|
| 1434 |
+
"num_input_tokens_seen": 248832,
|
| 1435 |
+
"step": 795
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 8.88888888888889,
|
| 1439 |
+
"grad_norm": 0.00970037654042244,
|
| 1440 |
+
"learning_rate": 1.893746038124497e-06,
|
| 1441 |
+
"loss": 0.0739,
|
| 1442 |
+
"num_input_tokens_seen": 250432,
|
| 1443 |
+
"step": 800
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 8.944444444444445,
|
| 1447 |
+
"grad_norm": 0.00786888599395752,
|
| 1448 |
+
"learning_rate": 1.713006526846439e-06,
|
| 1449 |
+
"loss": 0.0179,
|
| 1450 |
+
"num_input_tokens_seen": 252064,
|
| 1451 |
+
"step": 805
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 9.0,
|
| 1455 |
+
"grad_norm": 0.005709424614906311,
|
| 1456 |
+
"learning_rate": 1.541024301445404e-06,
|
| 1457 |
+
"loss": 0.0015,
|
| 1458 |
+
"num_input_tokens_seen": 253632,
|
| 1459 |
+
"step": 810
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 9.0,
|
| 1463 |
+
"eval_loss": 0.0697808712720871,
|
| 1464 |
+
"eval_runtime": 0.5791,
|
| 1465 |
+
"eval_samples_per_second": 69.068,
|
| 1466 |
+
"eval_steps_per_second": 17.267,
|
| 1467 |
+
"num_input_tokens_seen": 253632,
|
| 1468 |
+
"step": 810
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 9.055555555555555,
|
| 1472 |
+
"grad_norm": 0.16491225361824036,
|
| 1473 |
+
"learning_rate": 1.3778640374027985e-06,
|
| 1474 |
+
"loss": 0.0478,
|
| 1475 |
+
"num_input_tokens_seen": 255232,
|
| 1476 |
+
"step": 815
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 9.11111111111111,
|
| 1480 |
+
"grad_norm": 0.006427465938031673,
|
| 1481 |
+
"learning_rate": 1.2235870926211619e-06,
|
| 1482 |
+
"loss": 0.0011,
|
| 1483 |
+
"num_input_tokens_seen": 256800,
|
| 1484 |
+
"step": 820
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 9.166666666666666,
|
| 1488 |
+
"grad_norm": 0.00451459689065814,
|
| 1489 |
+
"learning_rate": 1.0782514843499653e-06,
|
| 1490 |
+
"loss": 0.0648,
|
| 1491 |
+
"num_input_tokens_seen": 258368,
|
| 1492 |
+
"step": 825
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 9.222222222222221,
|
| 1496 |
+
"grad_norm": 0.03183142468333244,
|
| 1497 |
+
"learning_rate": 9.419118673676924e-07,
|
| 1498 |
+
"loss": 0.0007,
|
| 1499 |
+
"num_input_tokens_seen": 259872,
|
| 1500 |
+
"step": 830
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 9.277777777777779,
|
| 1504 |
+
"grad_norm": 0.0683126449584961,
|
| 1505 |
+
"learning_rate": 8.146195134284052e-07,
|
| 1506 |
+
"loss": 0.0021,
|
| 1507 |
+
"num_input_tokens_seen": 261376,
|
| 1508 |
+
"step": 835
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 9.333333333333334,
|
| 1512 |
+
"grad_norm": 0.4682014286518097,
|
| 1513 |
+
"learning_rate": 6.964222919805391e-07,
|
| 1514 |
+
"loss": 0.0036,
|
| 1515 |
+
"num_input_tokens_seen": 262944,
|
| 1516 |
+
"step": 840
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 9.38888888888889,
|
| 1520 |
+
"grad_norm": 0.07767174392938614,
|
| 1521 |
+
"learning_rate": 5.87364652165176e-07,
|
| 1522 |
+
"loss": 0.0904,
|
| 1523 |
+
"num_input_tokens_seen": 264544,
|
| 1524 |
+
"step": 845
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 9.444444444444445,
|
| 1528 |
+
"grad_norm": 0.2937130331993103,
|
| 1529 |
+
"learning_rate": 4.874876061005173e-07,
|
| 1530 |
+
"loss": 0.002,
|
| 1531 |
+
"num_input_tokens_seen": 266080,
|
| 1532 |
+
"step": 850
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 9.5,
|
| 1536 |
+
"grad_norm": 0.15090857446193695,
|
| 1537 |
+
"learning_rate": 3.9682871345891883e-07,
|
| 1538 |
+
"loss": 0.0069,
|
| 1539 |
+
"num_input_tokens_seen": 267680,
|
| 1540 |
+
"step": 855
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 9.5,
|
| 1544 |
+
"eval_loss": 0.06901142001152039,
|
| 1545 |
+
"eval_runtime": 0.5858,
|
| 1546 |
+
"eval_samples_per_second": 68.285,
|
| 1547 |
+
"eval_steps_per_second": 17.071,
|
| 1548 |
+
"num_input_tokens_seen": 267680,
|
| 1549 |
+
"step": 855
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 9.555555555555555,
|
| 1553 |
+
"grad_norm": 5.201120376586914,
|
| 1554 |
+
"learning_rate": 3.1542206734221924e-07,
|
| 1555 |
+
"loss": 0.1035,
|
| 1556 |
+
"num_input_tokens_seen": 269280,
|
| 1557 |
+
"step": 860
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 9.61111111111111,
|
| 1561 |
+
"grad_norm": 0.4644763171672821,
|
| 1562 |
+
"learning_rate": 2.4329828146074095e-07,
|
| 1563 |
+
"loss": 0.0035,
|
| 1564 |
+
"num_input_tokens_seen": 270912,
|
| 1565 |
+
"step": 865
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 9.666666666666666,
|
| 1569 |
+
"grad_norm": 0.033365171402692795,
|
| 1570 |
+
"learning_rate": 1.8048447862070718e-07,
|
| 1571 |
+
"loss": 0.0094,
|
| 1572 |
+
"num_input_tokens_seen": 272512,
|
| 1573 |
+
"step": 870
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 9.722222222222221,
|
| 1577 |
+
"grad_norm": 0.2439211755990982,
|
| 1578 |
+
"learning_rate": 1.2700428052447033e-07,
|
| 1579 |
+
"loss": 0.0165,
|
| 1580 |
+
"num_input_tokens_seen": 274048,
|
| 1581 |
+
"step": 875
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 9.777777777777779,
|
| 1585 |
+
"grad_norm": 2.971935987472534,
|
| 1586 |
+
"learning_rate": 8.28777988873486e-08,
|
| 1587 |
+
"loss": 0.0171,
|
| 1588 |
+
"num_input_tokens_seen": 275648,
|
| 1589 |
+
"step": 880
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 9.833333333333334,
|
| 1593 |
+
"grad_norm": 0.022272281348705292,
|
| 1594 |
+
"learning_rate": 4.8121627874450625e-08,
|
| 1595 |
+
"loss": 0.0065,
|
| 1596 |
+
"num_input_tokens_seen": 277248,
|
| 1597 |
+
"step": 885
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 9.88888888888889,
|
| 1601 |
+
"grad_norm": 0.04621336981654167,
|
| 1602 |
+
"learning_rate": 2.2748837860270267e-08,
|
| 1603 |
+
"loss": 0.0028,
|
| 1604 |
+
"num_input_tokens_seen": 278816,
|
| 1605 |
+
"step": 890
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 9.944444444444445,
|
| 1609 |
+
"grad_norm": 0.011941147036850452,
|
| 1610 |
+
"learning_rate": 6.768970513457151e-09,
|
| 1611 |
+
"loss": 0.001,
|
| 1612 |
+
"num_input_tokens_seen": 280416,
|
| 1613 |
+
"step": 895
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 10.0,
|
| 1617 |
+
"grad_norm": 0.061786286532878876,
|
| 1618 |
+
"learning_rate": 1.8803520859811406e-10,
|
| 1619 |
+
"loss": 0.003,
|
| 1620 |
+
"num_input_tokens_seen": 281984,
|
| 1621 |
+
"step": 900
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 10.0,
|
| 1625 |
+
"eval_loss": 0.06747323274612427,
|
| 1626 |
+
"eval_runtime": 0.5823,
|
| 1627 |
+
"eval_samples_per_second": 68.688,
|
| 1628 |
+
"eval_steps_per_second": 17.172,
|
| 1629 |
+
"num_input_tokens_seen": 281984,
|
| 1630 |
+
"step": 900
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 10.0,
|
| 1634 |
+
"num_input_tokens_seen": 281984,
|
| 1635 |
+
"step": 900,
|
| 1636 |
+
"total_flos": 1.2698062524776448e+16,
|
| 1637 |
+
"train_loss": 0.06748742392027958,
|
| 1638 |
+
"train_runtime": 155.7099,
|
| 1639 |
+
"train_samples_per_second": 23.12,
|
| 1640 |
+
"train_steps_per_second": 5.78
|
| 1641 |
+
}
|
| 1642 |
+
],
|
| 1643 |
+
"logging_steps": 5,
|
| 1644 |
+
"max_steps": 900,
|
| 1645 |
+
"num_input_tokens_seen": 281984,
|
| 1646 |
+
"num_train_epochs": 10,
|
| 1647 |
+
"save_steps": 45,
|
| 1648 |
+
"stateful_callbacks": {
|
| 1649 |
+
"TrainerControl": {
|
| 1650 |
+
"args": {
|
| 1651 |
+
"should_epoch_stop": false,
|
| 1652 |
+
"should_evaluate": false,
|
| 1653 |
+
"should_log": false,
|
| 1654 |
+
"should_save": true,
|
| 1655 |
+
"should_training_stop": true
|
| 1656 |
+
},
|
| 1657 |
+
"attributes": {}
|
| 1658 |
+
}
|
| 1659 |
+
},
|
| 1660 |
+
"total_flos": 1.2698062524776448e+16,
|
| 1661 |
+
"train_batch_size": 4,
|
| 1662 |
+
"trial_name": null,
|
| 1663 |
+
"trial_params": null
|
| 1664 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|