End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +2048 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
# train_cb_789_1760637867
|
| 19 |
|
| 20 |
-
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
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
-
- Loss: 0.
|
| 23 |
- Num Input Tokens Seen: 711112
|
| 24 |
|
| 25 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
# train_cb_789_1760637867
|
| 19 |
|
| 20 |
+
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 cb dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.1673
|
| 23 |
- Num Input Tokens Seen: 711112
|
| 24 |
|
| 25 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.16730651259422302,
|
| 4 |
+
"eval_runtime": 0.8644,
|
| 5 |
+
"eval_samples_per_second": 28.922,
|
| 6 |
+
"eval_steps_per_second": 8.098,
|
| 7 |
+
"num_input_tokens_seen": 711112,
|
| 8 |
+
"total_flos": 3.2021051960131584e+16,
|
| 9 |
+
"train_loss": 0.23175576015195826,
|
| 10 |
+
"train_runtime": 283.4199,
|
| 11 |
+
"train_samples_per_second": 15.878,
|
| 12 |
+
"train_steps_per_second": 4.022
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.16730651259422302,
|
| 4 |
+
"eval_runtime": 0.8644,
|
| 5 |
+
"eval_samples_per_second": 28.922,
|
| 6 |
+
"eval_steps_per_second": 8.098,
|
| 7 |
+
"num_input_tokens_seen": 711112
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"num_input_tokens_seen": 711112,
|
| 4 |
+
"total_flos": 3.2021051960131584e+16,
|
| 5 |
+
"train_loss": 0.23175576015195826,
|
| 6 |
+
"train_runtime": 283.4199,
|
| 7 |
+
"train_samples_per_second": 15.878,
|
| 8 |
+
"train_steps_per_second": 4.022
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2048 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 228,
|
| 3 |
+
"best_metric": 0.16730651259422302,
|
| 4 |
+
"best_model_checkpoint": "saves_multiple/prompt-tuning/llama-3-8b-instruct/train_cb_789_1760637867/checkpoint-228",
|
| 5 |
+
"epoch": 20.0,
|
| 6 |
+
"eval_steps": 57,
|
| 7 |
+
"global_step": 1140,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 470.0,
|
| 15 |
+
"learning_rate": 0.0010526315789473684,
|
| 16 |
+
"loss": 0.6016,
|
| 17 |
+
"num_input_tokens_seen": 3136,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.17543859649122806,
|
| 22 |
+
"grad_norm": 30.0,
|
| 23 |
+
"learning_rate": 0.0023684210526315787,
|
| 24 |
+
"loss": 0.7503,
|
| 25 |
+
"num_input_tokens_seen": 6112,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.2631578947368421,
|
| 30 |
+
"grad_norm": 26.25,
|
| 31 |
+
"learning_rate": 0.003684210526315789,
|
| 32 |
+
"loss": 2.8137,
|
| 33 |
+
"num_input_tokens_seen": 10112,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.3508771929824561,
|
| 38 |
+
"grad_norm": 15.5,
|
| 39 |
+
"learning_rate": 0.004999999999999999,
|
| 40 |
+
"loss": 0.6094,
|
| 41 |
+
"num_input_tokens_seen": 13280,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.43859649122807015,
|
| 46 |
+
"grad_norm": 55.5,
|
| 47 |
+
"learning_rate": 0.0063157894736842095,
|
| 48 |
+
"loss": 0.7539,
|
| 49 |
+
"num_input_tokens_seen": 16288,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.5263157894736842,
|
| 54 |
+
"grad_norm": 15.25,
|
| 55 |
+
"learning_rate": 0.0076315789473684215,
|
| 56 |
+
"loss": 1.6404,
|
| 57 |
+
"num_input_tokens_seen": 19104,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.6140350877192983,
|
| 62 |
+
"grad_norm": 70.5,
|
| 63 |
+
"learning_rate": 0.008947368421052631,
|
| 64 |
+
"loss": 5.4717,
|
| 65 |
+
"num_input_tokens_seen": 22144,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.7017543859649122,
|
| 70 |
+
"grad_norm": 24.25,
|
| 71 |
+
"learning_rate": 0.010263157894736842,
|
| 72 |
+
"loss": 4.8847,
|
| 73 |
+
"num_input_tokens_seen": 25792,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.7894736842105263,
|
| 78 |
+
"grad_norm": 12.1875,
|
| 79 |
+
"learning_rate": 0.011578947368421052,
|
| 80 |
+
"loss": 2.5551,
|
| 81 |
+
"num_input_tokens_seen": 28576,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.8771929824561403,
|
| 86 |
+
"grad_norm": 3.296875,
|
| 87 |
+
"learning_rate": 0.012894736842105263,
|
| 88 |
+
"loss": 0.7962,
|
| 89 |
+
"num_input_tokens_seen": 31424,
|
| 90 |
+
"step": 50
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.9649122807017544,
|
| 94 |
+
"grad_norm": 1.8203125,
|
| 95 |
+
"learning_rate": 0.014210526315789472,
|
| 96 |
+
"loss": 1.353,
|
| 97 |
+
"num_input_tokens_seen": 34720,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"eval_loss": 1.0482252836227417,
|
| 103 |
+
"eval_runtime": 0.8373,
|
| 104 |
+
"eval_samples_per_second": 29.856,
|
| 105 |
+
"eval_steps_per_second": 8.36,
|
| 106 |
+
"num_input_tokens_seen": 35448,
|
| 107 |
+
"step": 57
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.0526315789473684,
|
| 111 |
+
"grad_norm": 2.46875,
|
| 112 |
+
"learning_rate": 0.015526315789473685,
|
| 113 |
+
"loss": 0.6558,
|
| 114 |
+
"num_input_tokens_seen": 37688,
|
| 115 |
+
"step": 60
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.1403508771929824,
|
| 119 |
+
"grad_norm": 1.140625,
|
| 120 |
+
"learning_rate": 0.016842105263157894,
|
| 121 |
+
"loss": 0.7116,
|
| 122 |
+
"num_input_tokens_seen": 40792,
|
| 123 |
+
"step": 65
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2280701754385965,
|
| 127 |
+
"grad_norm": 0.337890625,
|
| 128 |
+
"learning_rate": 0.018157894736842106,
|
| 129 |
+
"loss": 0.4407,
|
| 130 |
+
"num_input_tokens_seen": 43832,
|
| 131 |
+
"step": 70
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.3157894736842106,
|
| 135 |
+
"grad_norm": 0.09033203125,
|
| 136 |
+
"learning_rate": 0.019473684210526317,
|
| 137 |
+
"loss": 0.4844,
|
| 138 |
+
"num_input_tokens_seen": 46648,
|
| 139 |
+
"step": 75
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 1.4035087719298245,
|
| 143 |
+
"grad_norm": 0.2158203125,
|
| 144 |
+
"learning_rate": 0.020789473684210528,
|
| 145 |
+
"loss": 0.3423,
|
| 146 |
+
"num_input_tokens_seen": 49848,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.4912280701754386,
|
| 151 |
+
"grad_norm": 0.1962890625,
|
| 152 |
+
"learning_rate": 0.022105263157894735,
|
| 153 |
+
"loss": 0.3679,
|
| 154 |
+
"num_input_tokens_seen": 52504,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.5789473684210527,
|
| 159 |
+
"grad_norm": 0.287109375,
|
| 160 |
+
"learning_rate": 0.023421052631578947,
|
| 161 |
+
"loss": 0.4392,
|
| 162 |
+
"num_input_tokens_seen": 55448,
|
| 163 |
+
"step": 90
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.6666666666666665,
|
| 167 |
+
"grad_norm": 0.2373046875,
|
| 168 |
+
"learning_rate": 0.024736842105263158,
|
| 169 |
+
"loss": 0.1719,
|
| 170 |
+
"num_input_tokens_seen": 58776,
|
| 171 |
+
"step": 95
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.7543859649122808,
|
| 175 |
+
"grad_norm": 0.154296875,
|
| 176 |
+
"learning_rate": 0.02605263157894737,
|
| 177 |
+
"loss": 0.2579,
|
| 178 |
+
"num_input_tokens_seen": 61624,
|
| 179 |
+
"step": 100
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.8421052631578947,
|
| 183 |
+
"grad_norm": 0.10400390625,
|
| 184 |
+
"learning_rate": 0.027368421052631577,
|
| 185 |
+
"loss": 0.2097,
|
| 186 |
+
"num_input_tokens_seen": 65336,
|
| 187 |
+
"step": 105
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.9298245614035088,
|
| 191 |
+
"grad_norm": 3.5,
|
| 192 |
+
"learning_rate": 0.028684210526315788,
|
| 193 |
+
"loss": 0.4513,
|
| 194 |
+
"num_input_tokens_seen": 68280,
|
| 195 |
+
"step": 110
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_loss": 0.22055481374263763,
|
| 200 |
+
"eval_runtime": 0.849,
|
| 201 |
+
"eval_samples_per_second": 29.447,
|
| 202 |
+
"eval_steps_per_second": 8.245,
|
| 203 |
+
"num_input_tokens_seen": 70496,
|
| 204 |
+
"step": 114
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.017543859649123,
|
| 208 |
+
"grad_norm": 0.14453125,
|
| 209 |
+
"learning_rate": 0.03,
|
| 210 |
+
"loss": 0.2656,
|
| 211 |
+
"num_input_tokens_seen": 71008,
|
| 212 |
+
"step": 115
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.1052631578947367,
|
| 216 |
+
"grad_norm": 0.19921875,
|
| 217 |
+
"learning_rate": 0.02999824208523885,
|
| 218 |
+
"loss": 0.2642,
|
| 219 |
+
"num_input_tokens_seen": 74176,
|
| 220 |
+
"step": 120
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 2.192982456140351,
|
| 224 |
+
"grad_norm": 0.04248046875,
|
| 225 |
+
"learning_rate": 0.029992968752990647,
|
| 226 |
+
"loss": 0.2855,
|
| 227 |
+
"num_input_tokens_seen": 77472,
|
| 228 |
+
"step": 125
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.280701754385965,
|
| 232 |
+
"grad_norm": 0.06640625,
|
| 233 |
+
"learning_rate": 0.02998418123926453,
|
| 234 |
+
"loss": 0.3213,
|
| 235 |
+
"num_input_tokens_seen": 80192,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.3684210526315788,
|
| 240 |
+
"grad_norm": 1.4140625,
|
| 241 |
+
"learning_rate": 0.029971881603753848,
|
| 242 |
+
"loss": 0.4367,
|
| 243 |
+
"num_input_tokens_seen": 83424,
|
| 244 |
+
"step": 135
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.456140350877193,
|
| 248 |
+
"grad_norm": 0.06640625,
|
| 249 |
+
"learning_rate": 0.02995607272935338,
|
| 250 |
+
"loss": 0.4123,
|
| 251 |
+
"num_input_tokens_seen": 86080,
|
| 252 |
+
"step": 140
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.543859649122807,
|
| 256 |
+
"grad_norm": 0.0537109375,
|
| 257 |
+
"learning_rate": 0.02993675832148361,
|
| 258 |
+
"loss": 0.3135,
|
| 259 |
+
"num_input_tokens_seen": 89600,
|
| 260 |
+
"step": 145
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.6315789473684212,
|
| 264 |
+
"grad_norm": 0.08642578125,
|
| 265 |
+
"learning_rate": 0.029913942907222237,
|
| 266 |
+
"loss": 0.2866,
|
| 267 |
+
"num_input_tokens_seen": 92704,
|
| 268 |
+
"step": 150
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.719298245614035,
|
| 272 |
+
"grad_norm": 0.06201171875,
|
| 273 |
+
"learning_rate": 0.029887631834243058,
|
| 274 |
+
"loss": 0.2538,
|
| 275 |
+
"num_input_tokens_seen": 96160,
|
| 276 |
+
"step": 155
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.807017543859649,
|
| 280 |
+
"grad_norm": 0.2158203125,
|
| 281 |
+
"learning_rate": 0.029857831269562547,
|
| 282 |
+
"loss": 0.3449,
|
| 283 |
+
"num_input_tokens_seen": 99520,
|
| 284 |
+
"step": 160
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.8947368421052633,
|
| 288 |
+
"grad_norm": 0.052490234375,
|
| 289 |
+
"learning_rate": 0.029824548198094384,
|
| 290 |
+
"loss": 0.2438,
|
| 291 |
+
"num_input_tokens_seen": 102528,
|
| 292 |
+
"step": 165
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.982456140350877,
|
| 296 |
+
"grad_norm": 0.10595703125,
|
| 297 |
+
"learning_rate": 0.029787790421012244,
|
| 298 |
+
"loss": 0.2683,
|
| 299 |
+
"num_input_tokens_seen": 106304,
|
| 300 |
+
"step": 170
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.0,
|
| 304 |
+
"eval_loss": 0.18028998374938965,
|
| 305 |
+
"eval_runtime": 0.84,
|
| 306 |
+
"eval_samples_per_second": 29.763,
|
| 307 |
+
"eval_steps_per_second": 8.334,
|
| 308 |
+
"num_input_tokens_seen": 106416,
|
| 309 |
+
"step": 171
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.0701754385964914,
|
| 313 |
+
"grad_norm": 0.0283203125,
|
| 314 |
+
"learning_rate": 0.029747566553921325,
|
| 315 |
+
"loss": 0.2154,
|
| 316 |
+
"num_input_tokens_seen": 108880,
|
| 317 |
+
"step": 175
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.1578947368421053,
|
| 321 |
+
"grad_norm": 0.056640625,
|
| 322 |
+
"learning_rate": 0.029703886024838914,
|
| 323 |
+
"loss": 0.2558,
|
| 324 |
+
"num_input_tokens_seen": 111984,
|
| 325 |
+
"step": 180
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.245614035087719,
|
| 329 |
+
"grad_norm": 0.044189453125,
|
| 330 |
+
"learning_rate": 0.0296567590719846,
|
| 331 |
+
"loss": 0.2182,
|
| 332 |
+
"num_input_tokens_seen": 115248,
|
| 333 |
+
"step": 185
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.3333333333333335,
|
| 337 |
+
"grad_norm": 0.032958984375,
|
| 338 |
+
"learning_rate": 0.029606196741380517,
|
| 339 |
+
"loss": 0.1636,
|
| 340 |
+
"num_input_tokens_seen": 118384,
|
| 341 |
+
"step": 190
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 3.4210526315789473,
|
| 345 |
+
"grad_norm": 0.0693359375,
|
| 346 |
+
"learning_rate": 0.029552210884262308,
|
| 347 |
+
"loss": 0.2113,
|
| 348 |
+
"num_input_tokens_seen": 122224,
|
| 349 |
+
"step": 195
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 3.5087719298245617,
|
| 353 |
+
"grad_norm": 0.04150390625,
|
| 354 |
+
"learning_rate": 0.029494814154301326,
|
| 355 |
+
"loss": 0.2613,
|
| 356 |
+
"num_input_tokens_seen": 125584,
|
| 357 |
+
"step": 200
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.5964912280701755,
|
| 361 |
+
"grad_norm": 0.03466796875,
|
| 362 |
+
"learning_rate": 0.029434020004638753,
|
| 363 |
+
"loss": 0.1587,
|
| 364 |
+
"num_input_tokens_seen": 128816,
|
| 365 |
+
"step": 205
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.6842105263157894,
|
| 369 |
+
"grad_norm": 0.032958984375,
|
| 370 |
+
"learning_rate": 0.029369842684732334,
|
| 371 |
+
"loss": 0.4129,
|
| 372 |
+
"num_input_tokens_seen": 131952,
|
| 373 |
+
"step": 210
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 3.7719298245614032,
|
| 377 |
+
"grad_norm": 0.08056640625,
|
| 378 |
+
"learning_rate": 0.02930229723701646,
|
| 379 |
+
"loss": 0.2719,
|
| 380 |
+
"num_input_tokens_seen": 135248,
|
| 381 |
+
"step": 215
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.8596491228070176,
|
| 385 |
+
"grad_norm": 0.0703125,
|
| 386 |
+
"learning_rate": 0.029231399493376414,
|
| 387 |
+
"loss": 0.2218,
|
| 388 |
+
"num_input_tokens_seen": 138384,
|
| 389 |
+
"step": 220
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.9473684210526314,
|
| 393 |
+
"grad_norm": 0.059326171875,
|
| 394 |
+
"learning_rate": 0.02915716607143754,
|
| 395 |
+
"loss": 0.2404,
|
| 396 |
+
"num_input_tokens_seen": 141264,
|
| 397 |
+
"step": 225
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 4.0,
|
| 401 |
+
"eval_loss": 0.16730651259422302,
|
| 402 |
+
"eval_runtime": 0.8382,
|
| 403 |
+
"eval_samples_per_second": 29.825,
|
| 404 |
+
"eval_steps_per_second": 8.351,
|
| 405 |
+
"num_input_tokens_seen": 142480,
|
| 406 |
+
"step": 228
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.035087719298246,
|
| 410 |
+
"grad_norm": 0.1142578125,
|
| 411 |
+
"learning_rate": 0.029079614370670265,
|
| 412 |
+
"loss": 0.2303,
|
| 413 |
+
"num_input_tokens_seen": 143664,
|
| 414 |
+
"step": 230
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.12280701754386,
|
| 418 |
+
"grad_norm": 0.0294189453125,
|
| 419 |
+
"learning_rate": 0.028998762568311857,
|
| 420 |
+
"loss": 0.1414,
|
| 421 |
+
"num_input_tokens_seen": 146800,
|
| 422 |
+
"step": 235
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.2105263157894735,
|
| 426 |
+
"grad_norm": 0.0286865234375,
|
| 427 |
+
"learning_rate": 0.028914629615105897,
|
| 428 |
+
"loss": 0.1369,
|
| 429 |
+
"num_input_tokens_seen": 149456,
|
| 430 |
+
"step": 240
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.298245614035087,
|
| 434 |
+
"grad_norm": 0.041015625,
|
| 435 |
+
"learning_rate": 0.028827235230860424,
|
| 436 |
+
"loss": 0.196,
|
| 437 |
+
"num_input_tokens_seen": 152880,
|
| 438 |
+
"step": 245
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 4.385964912280702,
|
| 442 |
+
"grad_norm": 0.029296875,
|
| 443 |
+
"learning_rate": 0.02873659989982586,
|
| 444 |
+
"loss": 0.2025,
|
| 445 |
+
"num_input_tokens_seen": 156720,
|
| 446 |
+
"step": 250
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.473684210526316,
|
| 450 |
+
"grad_norm": 0.03564453125,
|
| 451 |
+
"learning_rate": 0.02864274486589371,
|
| 452 |
+
"loss": 0.311,
|
| 453 |
+
"num_input_tokens_seen": 159600,
|
| 454 |
+
"step": 255
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 4.56140350877193,
|
| 458 |
+
"grad_norm": 0.050537109375,
|
| 459 |
+
"learning_rate": 0.028545692127617244,
|
| 460 |
+
"loss": 0.2458,
|
| 461 |
+
"num_input_tokens_seen": 162224,
|
| 462 |
+
"step": 260
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.649122807017544,
|
| 466 |
+
"grad_norm": 0.0341796875,
|
| 467 |
+
"learning_rate": 0.0284454644330553,
|
| 468 |
+
"loss": 0.1756,
|
| 469 |
+
"num_input_tokens_seen": 165328,
|
| 470 |
+
"step": 265
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7368421052631575,
|
| 474 |
+
"grad_norm": 0.07666015625,
|
| 475 |
+
"learning_rate": 0.02834208527444037,
|
| 476 |
+
"loss": 0.2166,
|
| 477 |
+
"num_input_tokens_seen": 168592,
|
| 478 |
+
"step": 270
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 4.824561403508772,
|
| 482 |
+
"grad_norm": 0.025390625,
|
| 483 |
+
"learning_rate": 0.028235578882672318,
|
| 484 |
+
"loss": 0.2856,
|
| 485 |
+
"num_input_tokens_seen": 171696,
|
| 486 |
+
"step": 275
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 4.912280701754386,
|
| 490 |
+
"grad_norm": 0.09814453125,
|
| 491 |
+
"learning_rate": 0.028125970221638905,
|
| 492 |
+
"loss": 0.2382,
|
| 493 |
+
"num_input_tokens_seen": 174672,
|
| 494 |
+
"step": 280
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 5.0,
|
| 498 |
+
"grad_norm": 0.09521484375,
|
| 499 |
+
"learning_rate": 0.028013284982364554,
|
| 500 |
+
"loss": 0.2199,
|
| 501 |
+
"num_input_tokens_seen": 177224,
|
| 502 |
+
"step": 285
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.0,
|
| 506 |
+
"eval_loss": 0.252529114484787,
|
| 507 |
+
"eval_runtime": 0.8428,
|
| 508 |
+
"eval_samples_per_second": 29.661,
|
| 509 |
+
"eval_steps_per_second": 8.305,
|
| 510 |
+
"num_input_tokens_seen": 177224,
|
| 511 |
+
"step": 285
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 5.087719298245614,
|
| 515 |
+
"grad_norm": 0.06982421875,
|
| 516 |
+
"learning_rate": 0.027897549576988666,
|
| 517 |
+
"loss": 0.16,
|
| 518 |
+
"num_input_tokens_seen": 180840,
|
| 519 |
+
"step": 290
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.175438596491228,
|
| 523 |
+
"grad_norm": 0.150390625,
|
| 524 |
+
"learning_rate": 0.027778791132574907,
|
| 525 |
+
"loss": 0.1788,
|
| 526 |
+
"num_input_tokens_seen": 183976,
|
| 527 |
+
"step": 295
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.2631578947368425,
|
| 531 |
+
"grad_norm": 0.07763671875,
|
| 532 |
+
"learning_rate": 0.02765703748475293,
|
| 533 |
+
"loss": 0.3165,
|
| 534 |
+
"num_input_tokens_seen": 187016,
|
| 535 |
+
"step": 300
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 5.350877192982456,
|
| 539 |
+
"grad_norm": 0.0849609375,
|
| 540 |
+
"learning_rate": 0.027532317171194046,
|
| 541 |
+
"loss": 0.2463,
|
| 542 |
+
"num_input_tokens_seen": 189480,
|
| 543 |
+
"step": 305
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.43859649122807,
|
| 547 |
+
"grad_norm": 0.030029296875,
|
| 548 |
+
"learning_rate": 0.027404659424922272,
|
| 549 |
+
"loss": 0.1892,
|
| 550 |
+
"num_input_tokens_seen": 192712,
|
| 551 |
+
"step": 310
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 5.526315789473684,
|
| 555 |
+
"grad_norm": 0.0322265625,
|
| 556 |
+
"learning_rate": 0.027274094167462487,
|
| 557 |
+
"loss": 0.3672,
|
| 558 |
+
"num_input_tokens_seen": 195496,
|
| 559 |
+
"step": 315
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 5.614035087719298,
|
| 563 |
+
"grad_norm": 0.09423828125,
|
| 564 |
+
"learning_rate": 0.02714065200182714,
|
| 565 |
+
"loss": 0.1932,
|
| 566 |
+
"num_input_tokens_seen": 199240,
|
| 567 |
+
"step": 320
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.701754385964913,
|
| 571 |
+
"grad_norm": 0.11083984375,
|
| 572 |
+
"learning_rate": 0.02700436420534326,
|
| 573 |
+
"loss": 0.2354,
|
| 574 |
+
"num_input_tokens_seen": 201832,
|
| 575 |
+
"step": 325
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 5.7894736842105265,
|
| 579 |
+
"grad_norm": 0.095703125,
|
| 580 |
+
"learning_rate": 0.02686526272232141,
|
| 581 |
+
"loss": 0.1387,
|
| 582 |
+
"num_input_tokens_seen": 204968,
|
| 583 |
+
"step": 330
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 5.87719298245614,
|
| 587 |
+
"grad_norm": 0.050537109375,
|
| 588 |
+
"learning_rate": 0.026723380156568298,
|
| 589 |
+
"loss": 0.3058,
|
| 590 |
+
"num_input_tokens_seen": 208360,
|
| 591 |
+
"step": 335
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 5.964912280701754,
|
| 595 |
+
"grad_norm": 0.09228515625,
|
| 596 |
+
"learning_rate": 0.026578749763744813,
|
| 597 |
+
"loss": 0.226,
|
| 598 |
+
"num_input_tokens_seen": 211080,
|
| 599 |
+
"step": 340
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 6.0,
|
| 603 |
+
"eval_loss": 0.20502713322639465,
|
| 604 |
+
"eval_runtime": 0.8468,
|
| 605 |
+
"eval_samples_per_second": 29.521,
|
| 606 |
+
"eval_steps_per_second": 8.266,
|
| 607 |
+
"num_input_tokens_seen": 212000,
|
| 608 |
+
"step": 342
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 6.052631578947368,
|
| 612 |
+
"grad_norm": 0.036376953125,
|
| 613 |
+
"learning_rate": 0.026431405443571282,
|
| 614 |
+
"loss": 0.18,
|
| 615 |
+
"num_input_tokens_seen": 214176,
|
| 616 |
+
"step": 345
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.140350877192983,
|
| 620 |
+
"grad_norm": 0.028076171875,
|
| 621 |
+
"learning_rate": 0.02628138173188176,
|
| 622 |
+
"loss": 0.1775,
|
| 623 |
+
"num_input_tokens_seen": 216736,
|
| 624 |
+
"step": 350
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.228070175438597,
|
| 628 |
+
"grad_norm": 0.06396484375,
|
| 629 |
+
"learning_rate": 0.026128713792529224,
|
| 630 |
+
"loss": 0.2343,
|
| 631 |
+
"num_input_tokens_seen": 220224,
|
| 632 |
+
"step": 355
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 6.315789473684211,
|
| 636 |
+
"grad_norm": 0.05078125,
|
| 637 |
+
"learning_rate": 0.025973437409143554,
|
| 638 |
+
"loss": 0.1887,
|
| 639 |
+
"num_input_tokens_seen": 223136,
|
| 640 |
+
"step": 360
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.4035087719298245,
|
| 644 |
+
"grad_norm": 0.0634765625,
|
| 645 |
+
"learning_rate": 0.025815588976744273,
|
| 646 |
+
"loss": 0.1615,
|
| 647 |
+
"num_input_tokens_seen": 226944,
|
| 648 |
+
"step": 365
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 6.491228070175438,
|
| 652 |
+
"grad_norm": 0.031494140625,
|
| 653 |
+
"learning_rate": 0.02565520549320996,
|
| 654 |
+
"loss": 0.1746,
|
| 655 |
+
"num_input_tokens_seen": 230368,
|
| 656 |
+
"step": 370
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 6.578947368421053,
|
| 660 |
+
"grad_norm": 0.0791015625,
|
| 661 |
+
"learning_rate": 0.02549232455060637,
|
| 662 |
+
"loss": 0.2084,
|
| 663 |
+
"num_input_tokens_seen": 233376,
|
| 664 |
+
"step": 375
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 6.666666666666667,
|
| 668 |
+
"grad_norm": 0.056884765625,
|
| 669 |
+
"learning_rate": 0.025326984326375274,
|
| 670 |
+
"loss": 0.1668,
|
| 671 |
+
"num_input_tokens_seen": 236384,
|
| 672 |
+
"step": 380
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.754385964912281,
|
| 676 |
+
"grad_norm": 0.030517578125,
|
| 677 |
+
"learning_rate": 0.025159223574386116,
|
| 678 |
+
"loss": 0.1415,
|
| 679 |
+
"num_input_tokens_seen": 239584,
|
| 680 |
+
"step": 385
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 6.842105263157895,
|
| 684 |
+
"grad_norm": 0.0242919921875,
|
| 685 |
+
"learning_rate": 0.02498908161585253,
|
| 686 |
+
"loss": 0.3343,
|
| 687 |
+
"num_input_tokens_seen": 242624,
|
| 688 |
+
"step": 390
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 6.9298245614035086,
|
| 692 |
+
"grad_norm": 0.034423828125,
|
| 693 |
+
"learning_rate": 0.024816598330115895,
|
| 694 |
+
"loss": 0.0928,
|
| 695 |
+
"num_input_tokens_seen": 246304,
|
| 696 |
+
"step": 395
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 7.0,
|
| 700 |
+
"eval_loss": 0.2429497092962265,
|
| 701 |
+
"eval_runtime": 0.8455,
|
| 702 |
+
"eval_samples_per_second": 29.567,
|
| 703 |
+
"eval_steps_per_second": 8.279,
|
| 704 |
+
"num_input_tokens_seen": 248272,
|
| 705 |
+
"step": 399
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.017543859649122,
|
| 709 |
+
"grad_norm": 0.064453125,
|
| 710 |
+
"learning_rate": 0.024641814145298088,
|
| 711 |
+
"loss": 0.3043,
|
| 712 |
+
"num_input_tokens_seen": 248912,
|
| 713 |
+
"step": 400
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 7.105263157894737,
|
| 717 |
+
"grad_norm": 0.07568359375,
|
| 718 |
+
"learning_rate": 0.024464770028825585,
|
| 719 |
+
"loss": 0.2289,
|
| 720 |
+
"num_input_tokens_seen": 252016,
|
| 721 |
+
"step": 405
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 7.192982456140351,
|
| 725 |
+
"grad_norm": 0.03515625,
|
| 726 |
+
"learning_rate": 0.024285507477827137,
|
| 727 |
+
"loss": 0.2158,
|
| 728 |
+
"num_input_tokens_seen": 255856,
|
| 729 |
+
"step": 410
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.280701754385965,
|
| 733 |
+
"grad_norm": 0.05517578125,
|
| 734 |
+
"learning_rate": 0.02410406850940735,
|
| 735 |
+
"loss": 0.2326,
|
| 736 |
+
"num_input_tokens_seen": 258800,
|
| 737 |
+
"step": 415
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 7.368421052631579,
|
| 741 |
+
"grad_norm": 0.03857421875,
|
| 742 |
+
"learning_rate": 0.02392049565079834,
|
| 743 |
+
"loss": 0.1642,
|
| 744 |
+
"num_input_tokens_seen": 261584,
|
| 745 |
+
"step": 420
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 7.456140350877193,
|
| 749 |
+
"grad_norm": 0.03076171875,
|
| 750 |
+
"learning_rate": 0.023734831929391822,
|
| 751 |
+
"loss": 0.1351,
|
| 752 |
+
"num_input_tokens_seen": 265168,
|
| 753 |
+
"step": 425
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 7.543859649122807,
|
| 757 |
+
"grad_norm": 0.0830078125,
|
| 758 |
+
"learning_rate": 0.02354712086265399,
|
| 759 |
+
"loss": 0.2195,
|
| 760 |
+
"num_input_tokens_seen": 268240,
|
| 761 |
+
"step": 430
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 7.631578947368421,
|
| 765 |
+
"grad_norm": 0.0703125,
|
| 766 |
+
"learning_rate": 0.023357406447925527,
|
| 767 |
+
"loss": 0.2902,
|
| 768 |
+
"num_input_tokens_seen": 271728,
|
| 769 |
+
"step": 435
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 7.719298245614035,
|
| 773 |
+
"grad_norm": 0.031494140625,
|
| 774 |
+
"learning_rate": 0.023165733152109094,
|
| 775 |
+
"loss": 0.129,
|
| 776 |
+
"num_input_tokens_seen": 274736,
|
| 777 |
+
"step": 440
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 7.807017543859649,
|
| 781 |
+
"grad_norm": 0.028076171875,
|
| 782 |
+
"learning_rate": 0.022972145901246837,
|
| 783 |
+
"loss": 0.1276,
|
| 784 |
+
"num_input_tokens_seen": 278096,
|
| 785 |
+
"step": 445
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 7.894736842105263,
|
| 789 |
+
"grad_norm": 0.0634765625,
|
| 790 |
+
"learning_rate": 0.022776690069990207,
|
| 791 |
+
"loss": 0.0985,
|
| 792 |
+
"num_input_tokens_seen": 280592,
|
| 793 |
+
"step": 450
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 7.982456140350877,
|
| 797 |
+
"grad_norm": 0.06591796875,
|
| 798 |
+
"learning_rate": 0.022579411470964646,
|
| 799 |
+
"loss": 0.1848,
|
| 800 |
+
"num_input_tokens_seen": 284144,
|
| 801 |
+
"step": 455
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 8.0,
|
| 805 |
+
"eval_loss": 0.17890170216560364,
|
| 806 |
+
"eval_runtime": 0.8461,
|
| 807 |
+
"eval_samples_per_second": 29.547,
|
| 808 |
+
"eval_steps_per_second": 8.273,
|
| 809 |
+
"num_input_tokens_seen": 284248,
|
| 810 |
+
"step": 456
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 8.070175438596491,
|
| 814 |
+
"grad_norm": 0.0771484375,
|
| 815 |
+
"learning_rate": 0.022380356344031675,
|
| 816 |
+
"loss": 0.2103,
|
| 817 |
+
"num_input_tokens_seen": 287128,
|
| 818 |
+
"step": 460
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 8.157894736842104,
|
| 822 |
+
"grad_norm": 0.03662109375,
|
| 823 |
+
"learning_rate": 0.02217957134545074,
|
| 824 |
+
"loss": 0.1396,
|
| 825 |
+
"num_input_tokens_seen": 290744,
|
| 826 |
+
"step": 465
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 8.24561403508772,
|
| 830 |
+
"grad_norm": 0.072265625,
|
| 831 |
+
"learning_rate": 0.02197710353694355,
|
| 832 |
+
"loss": 0.1304,
|
| 833 |
+
"num_input_tokens_seen": 293816,
|
| 834 |
+
"step": 470
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 8.333333333333334,
|
| 838 |
+
"grad_norm": 0.052978515625,
|
| 839 |
+
"learning_rate": 0.02177300037466334,
|
| 840 |
+
"loss": 0.1891,
|
| 841 |
+
"num_input_tokens_seen": 297080,
|
| 842 |
+
"step": 475
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 8.421052631578947,
|
| 846 |
+
"grad_norm": 0.056884765625,
|
| 847 |
+
"learning_rate": 0.02156730969807168,
|
| 848 |
+
"loss": 0.2151,
|
| 849 |
+
"num_input_tokens_seen": 300408,
|
| 850 |
+
"step": 480
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 8.508771929824562,
|
| 854 |
+
"grad_norm": 0.06494140625,
|
| 855 |
+
"learning_rate": 0.021360079718725448,
|
| 856 |
+
"loss": 0.1712,
|
| 857 |
+
"num_input_tokens_seen": 303384,
|
| 858 |
+
"step": 485
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 8.596491228070175,
|
| 862 |
+
"grad_norm": 0.03369140625,
|
| 863 |
+
"learning_rate": 0.021151359008976602,
|
| 864 |
+
"loss": 0.1259,
|
| 865 |
+
"num_input_tokens_seen": 306232,
|
| 866 |
+
"step": 490
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 8.68421052631579,
|
| 870 |
+
"grad_norm": 0.043212890625,
|
| 871 |
+
"learning_rate": 0.02094119649058735,
|
| 872 |
+
"loss": 0.189,
|
| 873 |
+
"num_input_tokens_seen": 309816,
|
| 874 |
+
"step": 495
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 8.771929824561404,
|
| 878 |
+
"grad_norm": 0.031982421875,
|
| 879 |
+
"learning_rate": 0.020729641423263476,
|
| 880 |
+
"loss": 0.1627,
|
| 881 |
+
"num_input_tokens_seen": 312408,
|
| 882 |
+
"step": 500
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 8.859649122807017,
|
| 886 |
+
"grad_norm": 0.056884765625,
|
| 887 |
+
"learning_rate": 0.0205167433931084,
|
| 888 |
+
"loss": 0.1801,
|
| 889 |
+
"num_input_tokens_seen": 315416,
|
| 890 |
+
"step": 505
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 8.947368421052632,
|
| 894 |
+
"grad_norm": 0.0498046875,
|
| 895 |
+
"learning_rate": 0.020302552301000754,
|
| 896 |
+
"loss": 0.271,
|
| 897 |
+
"num_input_tokens_seen": 318392,
|
| 898 |
+
"step": 510
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 9.0,
|
| 902 |
+
"eval_loss": 0.19974899291992188,
|
| 903 |
+
"eval_runtime": 0.8451,
|
| 904 |
+
"eval_samples_per_second": 29.583,
|
| 905 |
+
"eval_steps_per_second": 8.283,
|
| 906 |
+
"num_input_tokens_seen": 319488,
|
| 907 |
+
"step": 513
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 9.035087719298245,
|
| 911 |
+
"grad_norm": 0.055419921875,
|
| 912 |
+
"learning_rate": 0.02008711835089822,
|
| 913 |
+
"loss": 0.14,
|
| 914 |
+
"num_input_tokens_seen": 320832,
|
| 915 |
+
"step": 515
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 9.12280701754386,
|
| 919 |
+
"grad_norm": 0.041015625,
|
| 920 |
+
"learning_rate": 0.019870492038070255,
|
| 921 |
+
"loss": 0.1427,
|
| 922 |
+
"num_input_tokens_seen": 323840,
|
| 923 |
+
"step": 520
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 9.210526315789474,
|
| 927 |
+
"grad_norm": 0.0223388671875,
|
| 928 |
+
"learning_rate": 0.01965272413726258,
|
| 929 |
+
"loss": 0.0586,
|
| 930 |
+
"num_input_tokens_seen": 327200,
|
| 931 |
+
"step": 525
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 9.298245614035087,
|
| 935 |
+
"grad_norm": 0.07861328125,
|
| 936 |
+
"learning_rate": 0.01943386569079618,
|
| 937 |
+
"loss": 0.1931,
|
| 938 |
+
"num_input_tokens_seen": 330208,
|
| 939 |
+
"step": 530
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 9.385964912280702,
|
| 943 |
+
"grad_norm": 0.02197265625,
|
| 944 |
+
"learning_rate": 0.01921396799660354,
|
| 945 |
+
"loss": 0.0365,
|
| 946 |
+
"num_input_tokens_seen": 332864,
|
| 947 |
+
"step": 535
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 9.473684210526315,
|
| 951 |
+
"grad_norm": 0.04541015625,
|
| 952 |
+
"learning_rate": 0.018993082596205,
|
| 953 |
+
"loss": 0.305,
|
| 954 |
+
"num_input_tokens_seen": 336224,
|
| 955 |
+
"step": 540
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 9.56140350877193,
|
| 959 |
+
"grad_norm": 0.09326171875,
|
| 960 |
+
"learning_rate": 0.018771261262628014,
|
| 961 |
+
"loss": 0.1647,
|
| 962 |
+
"num_input_tokens_seen": 339392,
|
| 963 |
+
"step": 545
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 9.649122807017545,
|
| 967 |
+
"grad_norm": 0.036376953125,
|
| 968 |
+
"learning_rate": 0.018548555988272136,
|
| 969 |
+
"loss": 0.1295,
|
| 970 |
+
"num_input_tokens_seen": 342624,
|
| 971 |
+
"step": 550
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 9.736842105263158,
|
| 975 |
+
"grad_norm": 0.056884765625,
|
| 976 |
+
"learning_rate": 0.018325018972722578,
|
| 977 |
+
"loss": 0.1854,
|
| 978 |
+
"num_input_tokens_seen": 345472,
|
| 979 |
+
"step": 555
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 9.824561403508772,
|
| 983 |
+
"grad_norm": 0.0712890625,
|
| 984 |
+
"learning_rate": 0.01810070261051526,
|
| 985 |
+
"loss": 0.1956,
|
| 986 |
+
"num_input_tokens_seen": 348160,
|
| 987 |
+
"step": 560
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 9.912280701754385,
|
| 991 |
+
"grad_norm": 0.0751953125,
|
| 992 |
+
"learning_rate": 0.01787565947885608,
|
| 993 |
+
"loss": 0.1919,
|
| 994 |
+
"num_input_tokens_seen": 351456,
|
| 995 |
+
"step": 565
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 10.0,
|
| 999 |
+
"grad_norm": 0.0830078125,
|
| 1000 |
+
"learning_rate": 0.01764994232529744,
|
| 1001 |
+
"loss": 0.1533,
|
| 1002 |
+
"num_input_tokens_seen": 354472,
|
| 1003 |
+
"step": 570
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 10.0,
|
| 1007 |
+
"eval_loss": 0.24398671090602875,
|
| 1008 |
+
"eval_runtime": 0.8453,
|
| 1009 |
+
"eval_samples_per_second": 29.576,
|
| 1010 |
+
"eval_steps_per_second": 8.281,
|
| 1011 |
+
"num_input_tokens_seen": 354472,
|
| 1012 |
+
"step": 570
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 10.087719298245615,
|
| 1016 |
+
"grad_norm": 0.0260009765625,
|
| 1017 |
+
"learning_rate": 0.01742360405537482,
|
| 1018 |
+
"loss": 0.1188,
|
| 1019 |
+
"num_input_tokens_seen": 358024,
|
| 1020 |
+
"step": 575
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 10.175438596491228,
|
| 1024 |
+
"grad_norm": 0.072265625,
|
| 1025 |
+
"learning_rate": 0.017196697720206326,
|
| 1026 |
+
"loss": 0.1679,
|
| 1027 |
+
"num_input_tokens_seen": 361192,
|
| 1028 |
+
"step": 580
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 10.263157894736842,
|
| 1032 |
+
"grad_norm": 0.1083984375,
|
| 1033 |
+
"learning_rate": 0.01696927650405807,
|
| 1034 |
+
"loss": 0.119,
|
| 1035 |
+
"num_input_tokens_seen": 363816,
|
| 1036 |
+
"step": 585
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 10.350877192982455,
|
| 1040 |
+
"grad_norm": 0.056884765625,
|
| 1041 |
+
"learning_rate": 0.016741393711878452,
|
| 1042 |
+
"loss": 0.1062,
|
| 1043 |
+
"num_input_tokens_seen": 366440,
|
| 1044 |
+
"step": 590
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 10.43859649122807,
|
| 1048 |
+
"grad_norm": 0.0859375,
|
| 1049 |
+
"learning_rate": 0.016513102756804024,
|
| 1050 |
+
"loss": 0.1102,
|
| 1051 |
+
"num_input_tokens_seen": 369672,
|
| 1052 |
+
"step": 595
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 10.526315789473685,
|
| 1056 |
+
"grad_norm": 0.07763671875,
|
| 1057 |
+
"learning_rate": 0.016284457147640083,
|
| 1058 |
+
"loss": 0.1657,
|
| 1059 |
+
"num_input_tokens_seen": 372808,
|
| 1060 |
+
"step": 600
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 10.614035087719298,
|
| 1064 |
+
"grad_norm": 0.169921875,
|
| 1065 |
+
"learning_rate": 0.016055510476318827,
|
| 1066 |
+
"loss": 0.2843,
|
| 1067 |
+
"num_input_tokens_seen": 375560,
|
| 1068 |
+
"step": 605
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 10.701754385964913,
|
| 1072 |
+
"grad_norm": 0.0888671875,
|
| 1073 |
+
"learning_rate": 0.015826316405337982,
|
| 1074 |
+
"loss": 0.1846,
|
| 1075 |
+
"num_input_tokens_seen": 378792,
|
| 1076 |
+
"step": 610
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 10.789473684210526,
|
| 1080 |
+
"grad_norm": 0.031494140625,
|
| 1081 |
+
"learning_rate": 0.015596928655182963,
|
| 1082 |
+
"loss": 0.1775,
|
| 1083 |
+
"num_input_tokens_seen": 382312,
|
| 1084 |
+
"step": 615
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 10.87719298245614,
|
| 1088 |
+
"grad_norm": 0.037353515625,
|
| 1089 |
+
"learning_rate": 0.015367400991735372,
|
| 1090 |
+
"loss": 0.1141,
|
| 1091 |
+
"num_input_tokens_seen": 386088,
|
| 1092 |
+
"step": 620
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 10.964912280701755,
|
| 1096 |
+
"grad_norm": 0.07958984375,
|
| 1097 |
+
"learning_rate": 0.015137787213670897,
|
| 1098 |
+
"loss": 0.0974,
|
| 1099 |
+
"num_input_tokens_seen": 388840,
|
| 1100 |
+
"step": 625
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 11.0,
|
| 1104 |
+
"eval_loss": 0.2553347647190094,
|
| 1105 |
+
"eval_runtime": 0.8443,
|
| 1106 |
+
"eval_samples_per_second": 29.609,
|
| 1107 |
+
"eval_steps_per_second": 8.291,
|
| 1108 |
+
"num_input_tokens_seen": 389408,
|
| 1109 |
+
"step": 627
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 11.052631578947368,
|
| 1113 |
+
"grad_norm": 0.212890625,
|
| 1114 |
+
"learning_rate": 0.014908141139849508,
|
| 1115 |
+
"loss": 0.127,
|
| 1116 |
+
"num_input_tokens_seen": 391200,
|
| 1117 |
+
"step": 630
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 11.140350877192983,
|
| 1121 |
+
"grad_norm": 0.1181640625,
|
| 1122 |
+
"learning_rate": 0.014678516596700955,
|
| 1123 |
+
"loss": 0.222,
|
| 1124 |
+
"num_input_tokens_seen": 394880,
|
| 1125 |
+
"step": 635
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 11.228070175438596,
|
| 1129 |
+
"grad_norm": 0.11572265625,
|
| 1130 |
+
"learning_rate": 0.014448967405608415,
|
| 1131 |
+
"loss": 0.0825,
|
| 1132 |
+
"num_input_tokens_seen": 398336,
|
| 1133 |
+
"step": 640
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 11.31578947368421,
|
| 1137 |
+
"grad_norm": 0.083984375,
|
| 1138 |
+
"learning_rate": 0.014219547370293413,
|
| 1139 |
+
"loss": 0.0889,
|
| 1140 |
+
"num_input_tokens_seen": 401632,
|
| 1141 |
+
"step": 645
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 11.403508771929825,
|
| 1145 |
+
"grad_norm": 0.0869140625,
|
| 1146 |
+
"learning_rate": 0.013990310264204829,
|
| 1147 |
+
"loss": 0.0741,
|
| 1148 |
+
"num_input_tokens_seen": 405120,
|
| 1149 |
+
"step": 650
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 11.491228070175438,
|
| 1153 |
+
"grad_norm": 0.1201171875,
|
| 1154 |
+
"learning_rate": 0.013761309817915016,
|
| 1155 |
+
"loss": 0.1272,
|
| 1156 |
+
"num_input_tokens_seen": 408192,
|
| 1157 |
+
"step": 655
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 11.578947368421053,
|
| 1161 |
+
"grad_norm": 0.125,
|
| 1162 |
+
"learning_rate": 0.013532599706525941,
|
| 1163 |
+
"loss": 0.1003,
|
| 1164 |
+
"num_input_tokens_seen": 411360,
|
| 1165 |
+
"step": 660
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 11.666666666666666,
|
| 1169 |
+
"grad_norm": 0.1259765625,
|
| 1170 |
+
"learning_rate": 0.013304233537088392,
|
| 1171 |
+
"loss": 0.1163,
|
| 1172 |
+
"num_input_tokens_seen": 414144,
|
| 1173 |
+
"step": 665
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 11.75438596491228,
|
| 1177 |
+
"grad_norm": 0.02978515625,
|
| 1178 |
+
"learning_rate": 0.013076264836037051,
|
| 1179 |
+
"loss": 0.0978,
|
| 1180 |
+
"num_input_tokens_seen": 417760,
|
| 1181 |
+
"step": 670
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 11.842105263157894,
|
| 1185 |
+
"grad_norm": 0.10107421875,
|
| 1186 |
+
"learning_rate": 0.012848747036644558,
|
| 1187 |
+
"loss": 0.1583,
|
| 1188 |
+
"num_input_tokens_seen": 420640,
|
| 1189 |
+
"step": 675
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 11.929824561403509,
|
| 1193 |
+
"grad_norm": 0.1630859375,
|
| 1194 |
+
"learning_rate": 0.012621733466497287,
|
| 1195 |
+
"loss": 0.1862,
|
| 1196 |
+
"num_input_tokens_seen": 423360,
|
| 1197 |
+
"step": 680
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 12.0,
|
| 1201 |
+
"eval_loss": 0.1732322871685028,
|
| 1202 |
+
"eval_runtime": 0.8443,
|
| 1203 |
+
"eval_samples_per_second": 29.609,
|
| 1204 |
+
"eval_steps_per_second": 8.291,
|
| 1205 |
+
"num_input_tokens_seen": 425328,
|
| 1206 |
+
"step": 684
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 12.017543859649123,
|
| 1210 |
+
"grad_norm": 0.046875,
|
| 1211 |
+
"learning_rate": 0.012395277334996044,
|
| 1212 |
+
"loss": 0.078,
|
| 1213 |
+
"num_input_tokens_seen": 426288,
|
| 1214 |
+
"step": 685
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 12.105263157894736,
|
| 1218 |
+
"grad_norm": 0.11376953125,
|
| 1219 |
+
"learning_rate": 0.012169431720884335,
|
| 1220 |
+
"loss": 0.2284,
|
| 1221 |
+
"num_input_tokens_seen": 429136,
|
| 1222 |
+
"step": 690
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 12.192982456140351,
|
| 1226 |
+
"grad_norm": 0.0771484375,
|
| 1227 |
+
"learning_rate": 0.01194424955980734,
|
| 1228 |
+
"loss": 0.1374,
|
| 1229 |
+
"num_input_tokens_seen": 432272,
|
| 1230 |
+
"step": 695
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 12.280701754385966,
|
| 1234 |
+
"grad_norm": 0.166015625,
|
| 1235 |
+
"learning_rate": 0.011719783631904363,
|
| 1236 |
+
"loss": 0.1079,
|
| 1237 |
+
"num_input_tokens_seen": 435760,
|
| 1238 |
+
"step": 700
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 12.368421052631579,
|
| 1242 |
+
"grad_norm": 0.11083984375,
|
| 1243 |
+
"learning_rate": 0.01149608654943782,
|
| 1244 |
+
"loss": 0.1092,
|
| 1245 |
+
"num_input_tokens_seen": 439312,
|
| 1246 |
+
"step": 705
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 12.456140350877194,
|
| 1250 |
+
"grad_norm": 0.04443359375,
|
| 1251 |
+
"learning_rate": 0.011273210744461479,
|
| 1252 |
+
"loss": 0.065,
|
| 1253 |
+
"num_input_tokens_seen": 442640,
|
| 1254 |
+
"step": 710
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 12.543859649122806,
|
| 1258 |
+
"grad_norm": 0.0654296875,
|
| 1259 |
+
"learning_rate": 0.011051208456531014,
|
| 1260 |
+
"loss": 0.0943,
|
| 1261 |
+
"num_input_tokens_seen": 445648,
|
| 1262 |
+
"step": 715
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 12.631578947368421,
|
| 1266 |
+
"grad_norm": 0.08984375,
|
| 1267 |
+
"learning_rate": 0.010830131720459601,
|
| 1268 |
+
"loss": 0.0867,
|
| 1269 |
+
"num_input_tokens_seen": 448208,
|
| 1270 |
+
"step": 720
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 12.719298245614034,
|
| 1274 |
+
"grad_norm": 0.06103515625,
|
| 1275 |
+
"learning_rate": 0.010610032354121612,
|
| 1276 |
+
"loss": 0.0837,
|
| 1277 |
+
"num_input_tokens_seen": 451600,
|
| 1278 |
+
"step": 725
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 12.807017543859649,
|
| 1282 |
+
"grad_norm": 0.26953125,
|
| 1283 |
+
"learning_rate": 0.01039096194630704,
|
| 1284 |
+
"loss": 0.0844,
|
| 1285 |
+
"num_input_tokens_seen": 454992,
|
| 1286 |
+
"step": 730
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 12.894736842105264,
|
| 1290 |
+
"grad_norm": 0.11669921875,
|
| 1291 |
+
"learning_rate": 0.010172971844629716,
|
| 1292 |
+
"loss": 0.0553,
|
| 1293 |
+
"num_input_tokens_seen": 458032,
|
| 1294 |
+
"step": 735
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 12.982456140350877,
|
| 1298 |
+
"grad_norm": 0.06884765625,
|
| 1299 |
+
"learning_rate": 0.009956113143491957,
|
| 1300 |
+
"loss": 0.0931,
|
| 1301 |
+
"num_input_tokens_seen": 461104,
|
| 1302 |
+
"step": 740
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 13.0,
|
| 1306 |
+
"eval_loss": 0.23760755360126495,
|
| 1307 |
+
"eval_runtime": 0.845,
|
| 1308 |
+
"eval_samples_per_second": 29.585,
|
| 1309 |
+
"eval_steps_per_second": 8.284,
|
| 1310 |
+
"num_input_tokens_seen": 461216,
|
| 1311 |
+
"step": 741
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 13.070175438596491,
|
| 1315 |
+
"grad_norm": 0.173828125,
|
| 1316 |
+
"learning_rate": 0.009740436672108685,
|
| 1317 |
+
"loss": 0.043,
|
| 1318 |
+
"num_input_tokens_seen": 463904,
|
| 1319 |
+
"step": 745
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 13.157894736842104,
|
| 1323 |
+
"grad_norm": 0.0245361328125,
|
| 1324 |
+
"learning_rate": 0.009525992982593583,
|
| 1325 |
+
"loss": 0.1013,
|
| 1326 |
+
"num_input_tokens_seen": 467328,
|
| 1327 |
+
"step": 750
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 13.24561403508772,
|
| 1331 |
+
"grad_norm": 0.419921875,
|
| 1332 |
+
"learning_rate": 0.009312832338110292,
|
| 1333 |
+
"loss": 0.1884,
|
| 1334 |
+
"num_input_tokens_seen": 469792,
|
| 1335 |
+
"step": 755
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 13.333333333333334,
|
| 1339 |
+
"grad_norm": 0.09033203125,
|
| 1340 |
+
"learning_rate": 0.009101004701091252,
|
| 1341 |
+
"loss": 0.0553,
|
| 1342 |
+
"num_input_tokens_seen": 472896,
|
| 1343 |
+
"step": 760
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 13.421052631578947,
|
| 1347 |
+
"grad_norm": 0.1123046875,
|
| 1348 |
+
"learning_rate": 0.008890559721527138,
|
| 1349 |
+
"loss": 0.0876,
|
| 1350 |
+
"num_input_tokens_seen": 475904,
|
| 1351 |
+
"step": 765
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 13.508771929824562,
|
| 1355 |
+
"grad_norm": 0.1416015625,
|
| 1356 |
+
"learning_rate": 0.008681546725329408,
|
| 1357 |
+
"loss": 0.1052,
|
| 1358 |
+
"num_input_tokens_seen": 478848,
|
| 1359 |
+
"step": 770
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 13.596491228070175,
|
| 1363 |
+
"grad_norm": 0.09423828125,
|
| 1364 |
+
"learning_rate": 0.008474014702768904,
|
| 1365 |
+
"loss": 0.065,
|
| 1366 |
+
"num_input_tokens_seen": 482080,
|
| 1367 |
+
"step": 775
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 13.68421052631579,
|
| 1371 |
+
"grad_norm": 0.00860595703125,
|
| 1372 |
+
"learning_rate": 0.008268012296993067,
|
| 1373 |
+
"loss": 0.0326,
|
| 1374 |
+
"num_input_tokens_seen": 485440,
|
| 1375 |
+
"step": 780
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 13.771929824561404,
|
| 1379 |
+
"grad_norm": 0.1484375,
|
| 1380 |
+
"learning_rate": 0.008063587792624567,
|
| 1381 |
+
"loss": 0.0849,
|
| 1382 |
+
"num_input_tokens_seen": 488640,
|
| 1383 |
+
"step": 785
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 13.859649122807017,
|
| 1387 |
+
"grad_norm": 0.1455078125,
|
| 1388 |
+
"learning_rate": 0.007860789104443896,
|
| 1389 |
+
"loss": 0.1001,
|
| 1390 |
+
"num_input_tokens_seen": 492032,
|
| 1391 |
+
"step": 790
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 13.947368421052632,
|
| 1395 |
+
"grad_norm": 0.2197265625,
|
| 1396 |
+
"learning_rate": 0.0076596637661587325,
|
| 1397 |
+
"loss": 0.0754,
|
| 1398 |
+
"num_input_tokens_seen": 495264,
|
| 1399 |
+
"step": 795
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 14.0,
|
| 1403 |
+
"eval_loss": 0.21240630745887756,
|
| 1404 |
+
"eval_runtime": 0.8448,
|
| 1405 |
+
"eval_samples_per_second": 29.594,
|
| 1406 |
+
"eval_steps_per_second": 8.286,
|
| 1407 |
+
"num_input_tokens_seen": 496704,
|
| 1408 |
+
"step": 798
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 14.035087719298245,
|
| 1412 |
+
"grad_norm": 0.05908203125,
|
| 1413 |
+
"learning_rate": 0.007460258919262529,
|
| 1414 |
+
"loss": 0.0359,
|
| 1415 |
+
"num_input_tokens_seen": 498464,
|
| 1416 |
+
"step": 800
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 14.12280701754386,
|
| 1420 |
+
"grad_norm": 0.1103515625,
|
| 1421 |
+
"learning_rate": 0.007262621301985144,
|
| 1422 |
+
"loss": 0.0507,
|
| 1423 |
+
"num_input_tokens_seen": 501632,
|
| 1424 |
+
"step": 805
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 14.210526315789474,
|
| 1428 |
+
"grad_norm": 0.0111083984375,
|
| 1429 |
+
"learning_rate": 0.007066797238337862,
|
| 1430 |
+
"loss": 0.0363,
|
| 1431 |
+
"num_input_tokens_seen": 504544,
|
| 1432 |
+
"step": 810
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 14.298245614035087,
|
| 1436 |
+
"grad_norm": 0.064453125,
|
| 1437 |
+
"learning_rate": 0.006872832627255643,
|
| 1438 |
+
"loss": 0.065,
|
| 1439 |
+
"num_input_tokens_seen": 507456,
|
| 1440 |
+
"step": 815
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 14.385964912280702,
|
| 1444 |
+
"grad_norm": 0.038330078125,
|
| 1445 |
+
"learning_rate": 0.006680772931838868,
|
| 1446 |
+
"loss": 0.0311,
|
| 1447 |
+
"num_input_tokens_seen": 511136,
|
| 1448 |
+
"step": 820
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 14.473684210526315,
|
| 1452 |
+
"grad_norm": 0.142578125,
|
| 1453 |
+
"learning_rate": 0.00649066316869736,
|
| 1454 |
+
"loss": 0.0323,
|
| 1455 |
+
"num_input_tokens_seen": 514368,
|
| 1456 |
+
"step": 825
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 14.56140350877193,
|
| 1460 |
+
"grad_norm": 0.2109375,
|
| 1461 |
+
"learning_rate": 0.0063025478973989585,
|
| 1462 |
+
"loss": 0.084,
|
| 1463 |
+
"num_input_tokens_seen": 517248,
|
| 1464 |
+
"step": 830
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 14.649122807017545,
|
| 1468 |
+
"grad_norm": 0.134765625,
|
| 1469 |
+
"learning_rate": 0.006116471210025301,
|
| 1470 |
+
"loss": 0.0384,
|
| 1471 |
+
"num_input_tokens_seen": 520160,
|
| 1472 |
+
"step": 835
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 14.736842105263158,
|
| 1476 |
+
"grad_norm": 0.0751953125,
|
| 1477 |
+
"learning_rate": 0.005932476720837105,
|
| 1478 |
+
"loss": 0.0341,
|
| 1479 |
+
"num_input_tokens_seen": 523424,
|
| 1480 |
+
"step": 840
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 14.824561403508772,
|
| 1484 |
+
"grad_norm": 0.11279296875,
|
| 1485 |
+
"learning_rate": 0.005750607556051514,
|
| 1486 |
+
"loss": 0.1026,
|
| 1487 |
+
"num_input_tokens_seen": 526240,
|
| 1488 |
+
"step": 845
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 14.912280701754385,
|
| 1492 |
+
"grad_norm": 0.080078125,
|
| 1493 |
+
"learning_rate": 0.0055709063437337685,
|
| 1494 |
+
"loss": 0.0325,
|
| 1495 |
+
"num_input_tokens_seen": 529440,
|
| 1496 |
+
"step": 850
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 15.0,
|
| 1500 |
+
"grad_norm": 0.007049560546875,
|
| 1501 |
+
"learning_rate": 0.005393415203805707,
|
| 1502 |
+
"loss": 0.0246,
|
| 1503 |
+
"num_input_tokens_seen": 532504,
|
| 1504 |
+
"step": 855
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 15.0,
|
| 1508 |
+
"eval_loss": 0.2732408344745636,
|
| 1509 |
+
"eval_runtime": 0.8445,
|
| 1510 |
+
"eval_samples_per_second": 29.605,
|
| 1511 |
+
"eval_steps_per_second": 8.289,
|
| 1512 |
+
"num_input_tokens_seen": 532504,
|
| 1513 |
+
"step": 855
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 15.087719298245615,
|
| 1517 |
+
"grad_norm": 0.0230712890625,
|
| 1518 |
+
"learning_rate": 0.005218175738173303,
|
| 1519 |
+
"loss": 0.0257,
|
| 1520 |
+
"num_input_tokens_seen": 535544,
|
| 1521 |
+
"step": 860
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 15.175438596491228,
|
| 1525 |
+
"grad_norm": 0.02587890625,
|
| 1526 |
+
"learning_rate": 0.005045229020975681,
|
| 1527 |
+
"loss": 0.0155,
|
| 1528 |
+
"num_input_tokens_seen": 538744,
|
| 1529 |
+
"step": 865
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 15.263157894736842,
|
| 1533 |
+
"grad_norm": 0.0296630859375,
|
| 1534 |
+
"learning_rate": 0.004874615588957773,
|
| 1535 |
+
"loss": 0.0222,
|
| 1536 |
+
"num_input_tokens_seen": 542040,
|
| 1537 |
+
"step": 870
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 15.350877192982455,
|
| 1541 |
+
"grad_norm": 0.003753662109375,
|
| 1542 |
+
"learning_rate": 0.0047063754319689975,
|
| 1543 |
+
"loss": 0.0084,
|
| 1544 |
+
"num_input_tokens_seen": 544888,
|
| 1545 |
+
"step": 875
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 15.43859649122807,
|
| 1549 |
+
"grad_norm": 0.08154296875,
|
| 1550 |
+
"learning_rate": 0.004540547983590069,
|
| 1551 |
+
"loss": 0.0189,
|
| 1552 |
+
"num_input_tokens_seen": 547640,
|
| 1553 |
+
"step": 880
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 15.526315789473685,
|
| 1557 |
+
"grad_norm": 0.0791015625,
|
| 1558 |
+
"learning_rate": 0.0043771721118902335,
|
| 1559 |
+
"loss": 0.0149,
|
| 1560 |
+
"num_input_tokens_seen": 551224,
|
| 1561 |
+
"step": 885
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 15.614035087719298,
|
| 1565 |
+
"grad_norm": 0.0400390625,
|
| 1566 |
+
"learning_rate": 0.004216286110317013,
|
| 1567 |
+
"loss": 0.0107,
|
| 1568 |
+
"num_input_tokens_seen": 554264,
|
| 1569 |
+
"step": 890
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 15.701754385964913,
|
| 1573 |
+
"grad_norm": 0.0059814453125,
|
| 1574 |
+
"learning_rate": 0.00405792768872069,
|
| 1575 |
+
"loss": 0.0101,
|
| 1576 |
+
"num_input_tokens_seen": 557880,
|
| 1577 |
+
"step": 895
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 15.789473684210526,
|
| 1581 |
+
"grad_norm": 0.1923828125,
|
| 1582 |
+
"learning_rate": 0.003902133964515502,
|
| 1583 |
+
"loss": 0.041,
|
| 1584 |
+
"num_input_tokens_seen": 560376,
|
| 1585 |
+
"step": 900
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 15.87719298245614,
|
| 1589 |
+
"grad_norm": 0.1044921875,
|
| 1590 |
+
"learning_rate": 0.00374894145397979,
|
| 1591 |
+
"loss": 0.019,
|
| 1592 |
+
"num_input_tokens_seen": 563864,
|
| 1593 |
+
"step": 905
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 15.964912280701755,
|
| 1597 |
+
"grad_norm": 0.0216064453125,
|
| 1598 |
+
"learning_rate": 0.0035983860636969525,
|
| 1599 |
+
"loss": 0.0136,
|
| 1600 |
+
"num_input_tokens_seen": 567352,
|
| 1601 |
+
"step": 910
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 16.0,
|
| 1605 |
+
"eval_loss": 0.29675254225730896,
|
| 1606 |
+
"eval_runtime": 0.845,
|
| 1607 |
+
"eval_samples_per_second": 29.584,
|
| 1608 |
+
"eval_steps_per_second": 8.284,
|
| 1609 |
+
"num_input_tokens_seen": 567952,
|
| 1610 |
+
"step": 912
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 16.05263157894737,
|
| 1614 |
+
"grad_norm": 0.0224609375,
|
| 1615 |
+
"learning_rate": 0.003450503082139393,
|
| 1616 |
+
"loss": 0.0154,
|
| 1617 |
+
"num_input_tokens_seen": 570448,
|
| 1618 |
+
"step": 915
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 16.140350877192983,
|
| 1622 |
+
"grad_norm": 0.021484375,
|
| 1623 |
+
"learning_rate": 0.003305327171397263,
|
| 1624 |
+
"loss": 0.0102,
|
| 1625 |
+
"num_input_tokens_seen": 573808,
|
| 1626 |
+
"step": 920
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 16.228070175438596,
|
| 1630 |
+
"grad_norm": 0.033447265625,
|
| 1631 |
+
"learning_rate": 0.003162892359054098,
|
| 1632 |
+
"loss": 0.0104,
|
| 1633 |
+
"num_input_tokens_seen": 577200,
|
| 1634 |
+
"step": 925
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 16.31578947368421,
|
| 1638 |
+
"grad_norm": 0.04443359375,
|
| 1639 |
+
"learning_rate": 0.0030232320302111047,
|
| 1640 |
+
"loss": 0.0101,
|
| 1641 |
+
"num_input_tokens_seen": 580240,
|
| 1642 |
+
"step": 930
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 16.403508771929825,
|
| 1646 |
+
"grad_norm": 0.0103759765625,
|
| 1647 |
+
"learning_rate": 0.0028863789196621093,
|
| 1648 |
+
"loss": 0.0098,
|
| 1649 |
+
"num_input_tokens_seen": 583120,
|
| 1650 |
+
"step": 935
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 16.49122807017544,
|
| 1654 |
+
"grad_norm": 0.0247802734375,
|
| 1655 |
+
"learning_rate": 0.0027523651042208564,
|
| 1656 |
+
"loss": 0.0053,
|
| 1657 |
+
"num_input_tokens_seen": 586032,
|
| 1658 |
+
"step": 940
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 16.57894736842105,
|
| 1662 |
+
"grad_norm": 0.004425048828125,
|
| 1663 |
+
"learning_rate": 0.0026212219952026,
|
| 1664 |
+
"loss": 0.0064,
|
| 1665 |
+
"num_input_tokens_seen": 589392,
|
| 1666 |
+
"step": 945
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 16.666666666666668,
|
| 1670 |
+
"grad_norm": 0.028564453125,
|
| 1671 |
+
"learning_rate": 0.0024929803310616222,
|
| 1672 |
+
"loss": 0.0072,
|
| 1673 |
+
"num_input_tokens_seen": 592624,
|
| 1674 |
+
"step": 950
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 16.75438596491228,
|
| 1678 |
+
"grad_norm": 0.03173828125,
|
| 1679 |
+
"learning_rate": 0.002367670170186516,
|
| 1680 |
+
"loss": 0.0064,
|
| 1681 |
+
"num_input_tokens_seen": 595440,
|
| 1682 |
+
"step": 955
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 16.842105263157894,
|
| 1686 |
+
"grad_norm": 0.0196533203125,
|
| 1687 |
+
"learning_rate": 0.00224532088385481,
|
| 1688 |
+
"loss": 0.0094,
|
| 1689 |
+
"num_input_tokens_seen": 598640,
|
| 1690 |
+
"step": 960
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 16.92982456140351,
|
| 1694 |
+
"grad_norm": 0.025390625,
|
| 1695 |
+
"learning_rate": 0.002125961149348706,
|
| 1696 |
+
"loss": 0.0147,
|
| 1697 |
+
"num_input_tokens_seen": 601584,
|
| 1698 |
+
"step": 965
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 17.0,
|
| 1702 |
+
"eval_loss": 0.2885914742946625,
|
| 1703 |
+
"eval_runtime": 0.8445,
|
| 1704 |
+
"eval_samples_per_second": 29.604,
|
| 1705 |
+
"eval_steps_per_second": 8.289,
|
| 1706 |
+
"num_input_tokens_seen": 603760,
|
| 1707 |
+
"step": 969
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 17.017543859649123,
|
| 1711 |
+
"grad_norm": 0.01446533203125,
|
| 1712 |
+
"learning_rate": 0.0020096189432334192,
|
| 1713 |
+
"loss": 0.0054,
|
| 1714 |
+
"num_input_tokens_seen": 604464,
|
| 1715 |
+
"step": 970
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 17.105263157894736,
|
| 1719 |
+
"grad_norm": 0.014404296875,
|
| 1720 |
+
"learning_rate": 0.001896321534799823,
|
| 1721 |
+
"loss": 0.0125,
|
| 1722 |
+
"num_input_tokens_seen": 607600,
|
| 1723 |
+
"step": 975
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 17.19298245614035,
|
| 1727 |
+
"grad_norm": 0.0213623046875,
|
| 1728 |
+
"learning_rate": 0.0017860954796727994,
|
| 1729 |
+
"loss": 0.009,
|
| 1730 |
+
"num_input_tokens_seen": 611152,
|
| 1731 |
+
"step": 980
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 17.280701754385966,
|
| 1735 |
+
"grad_norm": 0.010009765625,
|
| 1736 |
+
"learning_rate": 0.0016789666135869374,
|
| 1737 |
+
"loss": 0.0081,
|
| 1738 |
+
"num_input_tokens_seen": 614768,
|
| 1739 |
+
"step": 985
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 17.36842105263158,
|
| 1743 |
+
"grad_norm": 0.016845703125,
|
| 1744 |
+
"learning_rate": 0.0015749600463309049,
|
| 1745 |
+
"loss": 0.0075,
|
| 1746 |
+
"num_input_tokens_seen": 618384,
|
| 1747 |
+
"step": 990
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 17.45614035087719,
|
| 1751 |
+
"grad_norm": 0.00616455078125,
|
| 1752 |
+
"learning_rate": 0.0014741001558620163,
|
| 1753 |
+
"loss": 0.0035,
|
| 1754 |
+
"num_input_tokens_seen": 621424,
|
| 1755 |
+
"step": 995
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 17.54385964912281,
|
| 1759 |
+
"grad_norm": 0.0038299560546875,
|
| 1760 |
+
"learning_rate": 0.0013764105825923066,
|
| 1761 |
+
"loss": 0.0093,
|
| 1762 |
+
"num_input_tokens_seen": 624464,
|
| 1763 |
+
"step": 1000
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 17.63157894736842,
|
| 1767 |
+
"grad_norm": 0.003570556640625,
|
| 1768 |
+
"learning_rate": 0.0012819142238474862,
|
| 1769 |
+
"loss": 0.0061,
|
| 1770 |
+
"num_input_tokens_seen": 627824,
|
| 1771 |
+
"step": 1005
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 17.719298245614034,
|
| 1775 |
+
"grad_norm": 0.008544921875,
|
| 1776 |
+
"learning_rate": 0.0011906332285000793,
|
| 1777 |
+
"loss": 0.0059,
|
| 1778 |
+
"num_input_tokens_seen": 631056,
|
| 1779 |
+
"step": 1010
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 17.80701754385965,
|
| 1783 |
+
"grad_norm": 0.021484375,
|
| 1784 |
+
"learning_rate": 0.0011025889917779736,
|
| 1785 |
+
"loss": 0.0057,
|
| 1786 |
+
"num_input_tokens_seen": 634128,
|
| 1787 |
+
"step": 1015
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 17.894736842105264,
|
| 1791 |
+
"grad_norm": 0.0303955078125,
|
| 1792 |
+
"learning_rate": 0.0010178021502496165,
|
| 1793 |
+
"loss": 0.0066,
|
| 1794 |
+
"num_input_tokens_seen": 636656,
|
| 1795 |
+
"step": 1020
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 17.982456140350877,
|
| 1799 |
+
"grad_norm": 0.0361328125,
|
| 1800 |
+
"learning_rate": 0.0009362925769870394,
|
| 1801 |
+
"loss": 0.0111,
|
| 1802 |
+
"num_input_tokens_seen": 639696,
|
| 1803 |
+
"step": 1025
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 18.0,
|
| 1807 |
+
"eval_loss": 0.286989688873291,
|
| 1808 |
+
"eval_runtime": 0.8451,
|
| 1809 |
+
"eval_samples_per_second": 29.582,
|
| 1810 |
+
"eval_steps_per_second": 8.283,
|
| 1811 |
+
"num_input_tokens_seen": 639784,
|
| 1812 |
+
"step": 1026
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 18.07017543859649,
|
| 1816 |
+
"grad_norm": 0.0277099609375,
|
| 1817 |
+
"learning_rate": 0.0008580793769078487,
|
| 1818 |
+
"loss": 0.008,
|
| 1819 |
+
"num_input_tokens_seen": 642280,
|
| 1820 |
+
"step": 1030
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 18.157894736842106,
|
| 1824 |
+
"grad_norm": 0.023681640625,
|
| 1825 |
+
"learning_rate": 0.0007831808822972391,
|
| 1826 |
+
"loss": 0.0065,
|
| 1827 |
+
"num_input_tokens_seen": 645128,
|
| 1828 |
+
"step": 1035
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 18.24561403508772,
|
| 1832 |
+
"grad_norm": 0.0260009765625,
|
| 1833 |
+
"learning_rate": 0.0007116146485111063,
|
| 1834 |
+
"loss": 0.0066,
|
| 1835 |
+
"num_input_tokens_seen": 647528,
|
| 1836 |
+
"step": 1040
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 18.333333333333332,
|
| 1840 |
+
"grad_norm": 0.005828857421875,
|
| 1841 |
+
"learning_rate": 0.0006433974498612882,
|
| 1842 |
+
"loss": 0.004,
|
| 1843 |
+
"num_input_tokens_seen": 651560,
|
| 1844 |
+
"step": 1045
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 18.42105263157895,
|
| 1848 |
+
"grad_norm": 0.04248046875,
|
| 1849 |
+
"learning_rate": 0.0005785452756838482,
|
| 1850 |
+
"loss": 0.0122,
|
| 1851 |
+
"num_input_tokens_seen": 654568,
|
| 1852 |
+
"step": 1050
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 18.50877192982456,
|
| 1856 |
+
"grad_norm": 0.0213623046875,
|
| 1857 |
+
"learning_rate": 0.0005170733265913585,
|
| 1858 |
+
"loss": 0.0072,
|
| 1859 |
+
"num_input_tokens_seen": 657448,
|
| 1860 |
+
"step": 1055
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 18.596491228070175,
|
| 1864 |
+
"grad_norm": 0.0166015625,
|
| 1865 |
+
"learning_rate": 0.0004589960109100444,
|
| 1866 |
+
"loss": 0.0065,
|
| 1867 |
+
"num_input_tokens_seen": 660392,
|
| 1868 |
+
"step": 1060
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 18.68421052631579,
|
| 1872 |
+
"grad_norm": 0.042236328125,
|
| 1873 |
+
"learning_rate": 0.00040432694130264286,
|
| 1874 |
+
"loss": 0.0098,
|
| 1875 |
+
"num_input_tokens_seen": 663656,
|
| 1876 |
+
"step": 1065
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 18.771929824561404,
|
| 1880 |
+
"grad_norm": 0.00787353515625,
|
| 1881 |
+
"learning_rate": 0.0003530789315777466,
|
| 1882 |
+
"loss": 0.0055,
|
| 1883 |
+
"num_input_tokens_seen": 667208,
|
| 1884 |
+
"step": 1070
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 18.859649122807017,
|
| 1888 |
+
"grad_norm": 0.005645751953125,
|
| 1889 |
+
"learning_rate": 0.000305263993686391,
|
| 1890 |
+
"loss": 0.0049,
|
| 1891 |
+
"num_input_tokens_seen": 670568,
|
| 1892 |
+
"step": 1075
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 18.94736842105263,
|
| 1896 |
+
"grad_norm": 0.0218505859375,
|
| 1897 |
+
"learning_rate": 0.000260893334906595,
|
| 1898 |
+
"loss": 0.0129,
|
| 1899 |
+
"num_input_tokens_seen": 674024,
|
| 1900 |
+
"step": 1080
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 19.0,
|
| 1904 |
+
"eval_loss": 0.28917577862739563,
|
| 1905 |
+
"eval_runtime": 0.8471,
|
| 1906 |
+
"eval_samples_per_second": 29.514,
|
| 1907 |
+
"eval_steps_per_second": 8.264,
|
| 1908 |
+
"num_input_tokens_seen": 675800,
|
| 1909 |
+
"step": 1083
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 19.035087719298247,
|
| 1913 |
+
"grad_norm": 0.0157470703125,
|
| 1914 |
+
"learning_rate": 0.00021997735521649408,
|
| 1915 |
+
"loss": 0.0046,
|
| 1916 |
+
"num_input_tokens_seen": 676920,
|
| 1917 |
+
"step": 1085
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 19.12280701754386,
|
| 1921 |
+
"grad_norm": 0.0179443359375,
|
| 1922 |
+
"learning_rate": 0.00018252564485670973,
|
| 1923 |
+
"loss": 0.0069,
|
| 1924 |
+
"num_input_tokens_seen": 679896,
|
| 1925 |
+
"step": 1090
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 19.210526315789473,
|
| 1929 |
+
"grad_norm": 0.01556396484375,
|
| 1930 |
+
"learning_rate": 0.00014854698208250638,
|
| 1931 |
+
"loss": 0.0062,
|
| 1932 |
+
"num_input_tokens_seen": 683288,
|
| 1933 |
+
"step": 1095
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 19.29824561403509,
|
| 1937 |
+
"grad_norm": 0.01519775390625,
|
| 1938 |
+
"learning_rate": 0.00011804933110626359,
|
| 1939 |
+
"loss": 0.0067,
|
| 1940 |
+
"num_input_tokens_seen": 686168,
|
| 1941 |
+
"step": 1100
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 19.385964912280702,
|
| 1945 |
+
"grad_norm": 0.080078125,
|
| 1946 |
+
"learning_rate": 9.103984023075773e-05,
|
| 1947 |
+
"loss": 0.0107,
|
| 1948 |
+
"num_input_tokens_seen": 689016,
|
| 1949 |
+
"step": 1105
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 19.473684210526315,
|
| 1953 |
+
"grad_norm": 0.0240478515625,
|
| 1954 |
+
"learning_rate": 6.752484017368553e-05,
|
| 1955 |
+
"loss": 0.0062,
|
| 1956 |
+
"num_input_tokens_seen": 692376,
|
| 1957 |
+
"step": 1110
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 19.56140350877193,
|
| 1961 |
+
"grad_norm": 0.010498046875,
|
| 1962 |
+
"learning_rate": 4.750984258380608e-05,
|
| 1963 |
+
"loss": 0.0084,
|
| 1964 |
+
"num_input_tokens_seen": 695480,
|
| 1965 |
+
"step": 1115
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 19.649122807017545,
|
| 1969 |
+
"grad_norm": 0.01300048828125,
|
| 1970 |
+
"learning_rate": 3.099953874908079e-05,
|
| 1971 |
+
"loss": 0.0044,
|
| 1972 |
+
"num_input_tokens_seen": 698424,
|
| 1973 |
+
"step": 1120
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 19.736842105263158,
|
| 1977 |
+
"grad_norm": 0.01708984375,
|
| 1978 |
+
"learning_rate": 1.7997798497084714e-05,
|
| 1979 |
+
"loss": 0.0061,
|
| 1980 |
+
"num_input_tokens_seen": 701592,
|
| 1981 |
+
"step": 1125
|
| 1982 |
+
},
|
| 1983 |
+
{
|
| 1984 |
+
"epoch": 19.82456140350877,
|
| 1985 |
+
"grad_norm": 0.0174560546875,
|
| 1986 |
+
"learning_rate": 8.50766928796709e-06,
|
| 1987 |
+
"loss": 0.0073,
|
| 1988 |
+
"num_input_tokens_seen": 704504,
|
| 1989 |
+
"step": 1130
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 19.912280701754387,
|
| 1993 |
+
"grad_norm": 0.024658203125,
|
| 1994 |
+
"learning_rate": 2.5313755001593604e-06,
|
| 1995 |
+
"loss": 0.009,
|
| 1996 |
+
"num_input_tokens_seen": 707992,
|
| 1997 |
+
"step": 1135
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 20.0,
|
| 2001 |
+
"grad_norm": 0.00469970703125,
|
| 2002 |
+
"learning_rate": 7.03179089989181e-08,
|
| 2003 |
+
"loss": 0.0085,
|
| 2004 |
+
"num_input_tokens_seen": 711112,
|
| 2005 |
+
"step": 1140
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 20.0,
|
| 2009 |
+
"eval_loss": 0.28609830141067505,
|
| 2010 |
+
"eval_runtime": 0.845,
|
| 2011 |
+
"eval_samples_per_second": 29.585,
|
| 2012 |
+
"eval_steps_per_second": 8.284,
|
| 2013 |
+
"num_input_tokens_seen": 711112,
|
| 2014 |
+
"step": 1140
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 20.0,
|
| 2018 |
+
"num_input_tokens_seen": 711112,
|
| 2019 |
+
"step": 1140,
|
| 2020 |
+
"total_flos": 3.2021051960131584e+16,
|
| 2021 |
+
"train_loss": 0.23175576015195826,
|
| 2022 |
+
"train_runtime": 283.4199,
|
| 2023 |
+
"train_samples_per_second": 15.878,
|
| 2024 |
+
"train_steps_per_second": 4.022
|
| 2025 |
+
}
|
| 2026 |
+
],
|
| 2027 |
+
"logging_steps": 5,
|
| 2028 |
+
"max_steps": 1140,
|
| 2029 |
+
"num_input_tokens_seen": 711112,
|
| 2030 |
+
"num_train_epochs": 20,
|
| 2031 |
+
"save_steps": 57,
|
| 2032 |
+
"stateful_callbacks": {
|
| 2033 |
+
"TrainerControl": {
|
| 2034 |
+
"args": {
|
| 2035 |
+
"should_epoch_stop": false,
|
| 2036 |
+
"should_evaluate": false,
|
| 2037 |
+
"should_log": false,
|
| 2038 |
+
"should_save": true,
|
| 2039 |
+
"should_training_stop": true
|
| 2040 |
+
},
|
| 2041 |
+
"attributes": {}
|
| 2042 |
+
}
|
| 2043 |
+
},
|
| 2044 |
+
"total_flos": 3.2021051960131584e+16,
|
| 2045 |
+
"train_batch_size": 4,
|
| 2046 |
+
"trial_name": null,
|
| 2047 |
+
"trial_params": null
|
| 2048 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|