End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +2694 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
# train_copa_456_1760637759
|
| 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: 1.0934
|
| 23 |
- Num Input Tokens Seen: 501440
|
|
|
|
| 17 |
|
| 18 |
# train_copa_456_1760637759
|
| 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 copa dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 1.0934
|
| 23 |
- Num Input Tokens Seen: 501440
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 1.093362808227539,
|
| 4 |
+
"eval_runtime": 0.9825,
|
| 5 |
+
"eval_samples_per_second": 81.421,
|
| 6 |
+
"eval_steps_per_second": 20.355,
|
| 7 |
+
"num_input_tokens_seen": 501440,
|
| 8 |
+
"total_flos": 2.257961656516608e+16,
|
| 9 |
+
"train_loss": 0.2722132059369324,
|
| 10 |
+
"train_runtime": 168.4339,
|
| 11 |
+
"train_samples_per_second": 37.997,
|
| 12 |
+
"train_steps_per_second": 9.499
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 1.093362808227539,
|
| 4 |
+
"eval_runtime": 0.9825,
|
| 5 |
+
"eval_samples_per_second": 81.421,
|
| 6 |
+
"eval_steps_per_second": 20.355,
|
| 7 |
+
"num_input_tokens_seen": 501440
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"num_input_tokens_seen": 501440,
|
| 4 |
+
"total_flos": 2.257961656516608e+16,
|
| 5 |
+
"train_loss": 0.2722132059369324,
|
| 6 |
+
"train_runtime": 168.4339,
|
| 7 |
+
"train_samples_per_second": 37.997,
|
| 8 |
+
"train_steps_per_second": 9.499
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2694 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 320,
|
| 3 |
+
"best_metric": 0.23625226318836212,
|
| 4 |
+
"best_model_checkpoint": "saves_multiple/prefix-tuning/llama-3-8b-instruct/train_copa_456_1760637759/checkpoint-320",
|
| 5 |
+
"epoch": 20.0,
|
| 6 |
+
"eval_steps": 160,
|
| 7 |
+
"global_step": 1600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0625,
|
| 14 |
+
"grad_norm": 200.03012084960938,
|
| 15 |
+
"learning_rate": 2.5000000000000004e-07,
|
| 16 |
+
"loss": 9.2806,
|
| 17 |
+
"num_input_tokens_seen": 1632,
|
| 18 |
+
"step": 5
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.125,
|
| 22 |
+
"grad_norm": 214.69956970214844,
|
| 23 |
+
"learning_rate": 5.625e-07,
|
| 24 |
+
"loss": 8.7836,
|
| 25 |
+
"num_input_tokens_seen": 3136,
|
| 26 |
+
"step": 10
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1875,
|
| 30 |
+
"grad_norm": 193.6563720703125,
|
| 31 |
+
"learning_rate": 8.75e-07,
|
| 32 |
+
"loss": 7.8812,
|
| 33 |
+
"num_input_tokens_seen": 4736,
|
| 34 |
+
"step": 15
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.25,
|
| 38 |
+
"grad_norm": 165.62625122070312,
|
| 39 |
+
"learning_rate": 1.1875e-06,
|
| 40 |
+
"loss": 6.7373,
|
| 41 |
+
"num_input_tokens_seen": 6304,
|
| 42 |
+
"step": 20
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.3125,
|
| 46 |
+
"grad_norm": 128.19512939453125,
|
| 47 |
+
"learning_rate": 1.5e-06,
|
| 48 |
+
"loss": 5.2865,
|
| 49 |
+
"num_input_tokens_seen": 7904,
|
| 50 |
+
"step": 25
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.375,
|
| 54 |
+
"grad_norm": 103.28244018554688,
|
| 55 |
+
"learning_rate": 1.8125e-06,
|
| 56 |
+
"loss": 3.977,
|
| 57 |
+
"num_input_tokens_seen": 9472,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4375,
|
| 62 |
+
"grad_norm": 80.52030944824219,
|
| 63 |
+
"learning_rate": 2.125e-06,
|
| 64 |
+
"loss": 2.8338,
|
| 65 |
+
"num_input_tokens_seen": 11008,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.5,
|
| 70 |
+
"grad_norm": 58.259361267089844,
|
| 71 |
+
"learning_rate": 2.4375e-06,
|
| 72 |
+
"loss": 1.5963,
|
| 73 |
+
"num_input_tokens_seen": 12512,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.5625,
|
| 78 |
+
"grad_norm": 41.13335418701172,
|
| 79 |
+
"learning_rate": 2.7500000000000004e-06,
|
| 80 |
+
"loss": 0.8379,
|
| 81 |
+
"num_input_tokens_seen": 14016,
|
| 82 |
+
"step": 45
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.625,
|
| 86 |
+
"grad_norm": 44.33089828491211,
|
| 87 |
+
"learning_rate": 3.0625000000000003e-06,
|
| 88 |
+
"loss": 0.4101,
|
| 89 |
+
"num_input_tokens_seen": 15584,
|
| 90 |
+
"step": 50
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.6875,
|
| 94 |
+
"grad_norm": 89.49345397949219,
|
| 95 |
+
"learning_rate": 3.3750000000000003e-06,
|
| 96 |
+
"loss": 0.3284,
|
| 97 |
+
"num_input_tokens_seen": 17152,
|
| 98 |
+
"step": 55
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.75,
|
| 102 |
+
"grad_norm": 22.057518005371094,
|
| 103 |
+
"learning_rate": 3.6875000000000007e-06,
|
| 104 |
+
"loss": 0.3562,
|
| 105 |
+
"num_input_tokens_seen": 18752,
|
| 106 |
+
"step": 60
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.8125,
|
| 110 |
+
"grad_norm": 15.762816429138184,
|
| 111 |
+
"learning_rate": 4.000000000000001e-06,
|
| 112 |
+
"loss": 0.2688,
|
| 113 |
+
"num_input_tokens_seen": 20320,
|
| 114 |
+
"step": 65
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.875,
|
| 118 |
+
"grad_norm": 25.10136604309082,
|
| 119 |
+
"learning_rate": 4.312500000000001e-06,
|
| 120 |
+
"loss": 0.2826,
|
| 121 |
+
"num_input_tokens_seen": 21888,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.9375,
|
| 126 |
+
"grad_norm": 22.71470832824707,
|
| 127 |
+
"learning_rate": 4.625000000000001e-06,
|
| 128 |
+
"loss": 0.2588,
|
| 129 |
+
"num_input_tokens_seen": 23424,
|
| 130 |
+
"step": 75
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.0,
|
| 134 |
+
"grad_norm": 8.763684272766113,
|
| 135 |
+
"learning_rate": 4.937500000000001e-06,
|
| 136 |
+
"loss": 0.2796,
|
| 137 |
+
"num_input_tokens_seen": 24960,
|
| 138 |
+
"step": 80
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.0625,
|
| 142 |
+
"grad_norm": 24.20291519165039,
|
| 143 |
+
"learning_rate": 5.2500000000000006e-06,
|
| 144 |
+
"loss": 0.2615,
|
| 145 |
+
"num_input_tokens_seen": 26464,
|
| 146 |
+
"step": 85
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.125,
|
| 150 |
+
"grad_norm": 5.363375663757324,
|
| 151 |
+
"learning_rate": 5.5625000000000005e-06,
|
| 152 |
+
"loss": 0.263,
|
| 153 |
+
"num_input_tokens_seen": 28064,
|
| 154 |
+
"step": 90
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.1875,
|
| 158 |
+
"grad_norm": 34.018043518066406,
|
| 159 |
+
"learning_rate": 5.8750000000000005e-06,
|
| 160 |
+
"loss": 0.2422,
|
| 161 |
+
"num_input_tokens_seen": 29664,
|
| 162 |
+
"step": 95
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.25,
|
| 166 |
+
"grad_norm": 35.1256217956543,
|
| 167 |
+
"learning_rate": 6.1875000000000005e-06,
|
| 168 |
+
"loss": 0.2233,
|
| 169 |
+
"num_input_tokens_seen": 31264,
|
| 170 |
+
"step": 100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.3125,
|
| 174 |
+
"grad_norm": 7.767378807067871,
|
| 175 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 176 |
+
"loss": 0.2181,
|
| 177 |
+
"num_input_tokens_seen": 32864,
|
| 178 |
+
"step": 105
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.375,
|
| 182 |
+
"grad_norm": 14.077888488769531,
|
| 183 |
+
"learning_rate": 6.8125e-06,
|
| 184 |
+
"loss": 0.3463,
|
| 185 |
+
"num_input_tokens_seen": 34432,
|
| 186 |
+
"step": 110
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.4375,
|
| 190 |
+
"grad_norm": 10.013070106506348,
|
| 191 |
+
"learning_rate": 7.125e-06,
|
| 192 |
+
"loss": 0.2563,
|
| 193 |
+
"num_input_tokens_seen": 35968,
|
| 194 |
+
"step": 115
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.5,
|
| 198 |
+
"grad_norm": 7.696524143218994,
|
| 199 |
+
"learning_rate": 7.437500000000001e-06,
|
| 200 |
+
"loss": 0.2045,
|
| 201 |
+
"num_input_tokens_seen": 37536,
|
| 202 |
+
"step": 120
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.5625,
|
| 206 |
+
"grad_norm": 12.979708671569824,
|
| 207 |
+
"learning_rate": 7.75e-06,
|
| 208 |
+
"loss": 0.432,
|
| 209 |
+
"num_input_tokens_seen": 39136,
|
| 210 |
+
"step": 125
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 1.625,
|
| 214 |
+
"grad_norm": 19.278409957885742,
|
| 215 |
+
"learning_rate": 8.062500000000001e-06,
|
| 216 |
+
"loss": 0.2754,
|
| 217 |
+
"num_input_tokens_seen": 40736,
|
| 218 |
+
"step": 130
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.6875,
|
| 222 |
+
"grad_norm": 7.332155704498291,
|
| 223 |
+
"learning_rate": 8.375e-06,
|
| 224 |
+
"loss": 0.2547,
|
| 225 |
+
"num_input_tokens_seen": 42240,
|
| 226 |
+
"step": 135
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.75,
|
| 230 |
+
"grad_norm": 4.9524245262146,
|
| 231 |
+
"learning_rate": 8.687500000000001e-06,
|
| 232 |
+
"loss": 0.2304,
|
| 233 |
+
"num_input_tokens_seen": 43808,
|
| 234 |
+
"step": 140
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.8125,
|
| 238 |
+
"grad_norm": 7.520391941070557,
|
| 239 |
+
"learning_rate": 9e-06,
|
| 240 |
+
"loss": 0.218,
|
| 241 |
+
"num_input_tokens_seen": 45344,
|
| 242 |
+
"step": 145
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.875,
|
| 246 |
+
"grad_norm": 15.684959411621094,
|
| 247 |
+
"learning_rate": 9.312500000000001e-06,
|
| 248 |
+
"loss": 0.2485,
|
| 249 |
+
"num_input_tokens_seen": 46912,
|
| 250 |
+
"step": 150
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.9375,
|
| 254 |
+
"grad_norm": 3.4786770343780518,
|
| 255 |
+
"learning_rate": 9.625e-06,
|
| 256 |
+
"loss": 0.2414,
|
| 257 |
+
"num_input_tokens_seen": 48480,
|
| 258 |
+
"step": 155
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.0,
|
| 262 |
+
"grad_norm": 2.7771852016448975,
|
| 263 |
+
"learning_rate": 9.937500000000001e-06,
|
| 264 |
+
"loss": 0.2344,
|
| 265 |
+
"num_input_tokens_seen": 50080,
|
| 266 |
+
"step": 160
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.0,
|
| 270 |
+
"eval_loss": 0.23884077370166779,
|
| 271 |
+
"eval_runtime": 0.9157,
|
| 272 |
+
"eval_samples_per_second": 87.369,
|
| 273 |
+
"eval_steps_per_second": 21.842,
|
| 274 |
+
"num_input_tokens_seen": 50080,
|
| 275 |
+
"step": 160
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.0625,
|
| 279 |
+
"grad_norm": 4.3142242431640625,
|
| 280 |
+
"learning_rate": 9.999809615320857e-06,
|
| 281 |
+
"loss": 0.2257,
|
| 282 |
+
"num_input_tokens_seen": 51616,
|
| 283 |
+
"step": 165
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 2.125,
|
| 287 |
+
"grad_norm": 3.4161956310272217,
|
| 288 |
+
"learning_rate": 9.999036202410324e-06,
|
| 289 |
+
"loss": 0.1899,
|
| 290 |
+
"num_input_tokens_seen": 53216,
|
| 291 |
+
"step": 170
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.1875,
|
| 295 |
+
"grad_norm": 10.080194473266602,
|
| 296 |
+
"learning_rate": 9.997667954183566e-06,
|
| 297 |
+
"loss": 0.3096,
|
| 298 |
+
"num_input_tokens_seen": 54784,
|
| 299 |
+
"step": 175
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.25,
|
| 303 |
+
"grad_norm": 3.2639670372009277,
|
| 304 |
+
"learning_rate": 9.995705033448435e-06,
|
| 305 |
+
"loss": 0.2305,
|
| 306 |
+
"num_input_tokens_seen": 56384,
|
| 307 |
+
"step": 180
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.3125,
|
| 311 |
+
"grad_norm": 9.221940040588379,
|
| 312 |
+
"learning_rate": 9.993147673772869e-06,
|
| 313 |
+
"loss": 0.2628,
|
| 314 |
+
"num_input_tokens_seen": 57920,
|
| 315 |
+
"step": 185
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.375,
|
| 319 |
+
"grad_norm": 3.5883162021636963,
|
| 320 |
+
"learning_rate": 9.9899961794571e-06,
|
| 321 |
+
"loss": 0.2506,
|
| 322 |
+
"num_input_tokens_seen": 59520,
|
| 323 |
+
"step": 190
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.4375,
|
| 327 |
+
"grad_norm": 14.086549758911133,
|
| 328 |
+
"learning_rate": 9.986250925497429e-06,
|
| 329 |
+
"loss": 0.2822,
|
| 330 |
+
"num_input_tokens_seen": 61088,
|
| 331 |
+
"step": 195
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.5,
|
| 335 |
+
"grad_norm": 7.669594764709473,
|
| 336 |
+
"learning_rate": 9.981912357541628e-06,
|
| 337 |
+
"loss": 0.2747,
|
| 338 |
+
"num_input_tokens_seen": 62656,
|
| 339 |
+
"step": 200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.5625,
|
| 343 |
+
"grad_norm": 2.2965734004974365,
|
| 344 |
+
"learning_rate": 9.976980991835896e-06,
|
| 345 |
+
"loss": 0.2442,
|
| 346 |
+
"num_input_tokens_seen": 64224,
|
| 347 |
+
"step": 205
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.625,
|
| 351 |
+
"grad_norm": 3.275341033935547,
|
| 352 |
+
"learning_rate": 9.971457415163435e-06,
|
| 353 |
+
"loss": 0.2616,
|
| 354 |
+
"num_input_tokens_seen": 65760,
|
| 355 |
+
"step": 210
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 2.6875,
|
| 359 |
+
"grad_norm": 2.5465989112854004,
|
| 360 |
+
"learning_rate": 9.965342284774633e-06,
|
| 361 |
+
"loss": 0.2425,
|
| 362 |
+
"num_input_tokens_seen": 67296,
|
| 363 |
+
"step": 215
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 2.75,
|
| 367 |
+
"grad_norm": 4.648524284362793,
|
| 368 |
+
"learning_rate": 9.958636328308852e-06,
|
| 369 |
+
"loss": 0.2764,
|
| 370 |
+
"num_input_tokens_seen": 68832,
|
| 371 |
+
"step": 220
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.8125,
|
| 375 |
+
"grad_norm": 10.161576271057129,
|
| 376 |
+
"learning_rate": 9.951340343707852e-06,
|
| 377 |
+
"loss": 0.255,
|
| 378 |
+
"num_input_tokens_seen": 70400,
|
| 379 |
+
"step": 225
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 2.875,
|
| 383 |
+
"grad_norm": 5.629575252532959,
|
| 384 |
+
"learning_rate": 9.943455199120836e-06,
|
| 385 |
+
"loss": 0.2356,
|
| 386 |
+
"num_input_tokens_seen": 72000,
|
| 387 |
+
"step": 230
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.9375,
|
| 391 |
+
"grad_norm": 4.087530612945557,
|
| 392 |
+
"learning_rate": 9.934981832801161e-06,
|
| 393 |
+
"loss": 0.2339,
|
| 394 |
+
"num_input_tokens_seen": 73536,
|
| 395 |
+
"step": 235
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 3.0,
|
| 399 |
+
"grad_norm": 1.1364092826843262,
|
| 400 |
+
"learning_rate": 9.925921252994677e-06,
|
| 401 |
+
"loss": 0.232,
|
| 402 |
+
"num_input_tokens_seen": 75104,
|
| 403 |
+
"step": 240
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 3.0625,
|
| 407 |
+
"grad_norm": 1.9829556941986084,
|
| 408 |
+
"learning_rate": 9.916274537819774e-06,
|
| 409 |
+
"loss": 0.2282,
|
| 410 |
+
"num_input_tokens_seen": 76640,
|
| 411 |
+
"step": 245
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 3.125,
|
| 415 |
+
"grad_norm": 1.830893635749817,
|
| 416 |
+
"learning_rate": 9.90604283513909e-06,
|
| 417 |
+
"loss": 0.2614,
|
| 418 |
+
"num_input_tokens_seen": 78208,
|
| 419 |
+
"step": 250
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.1875,
|
| 423 |
+
"grad_norm": 2.716417074203491,
|
| 424 |
+
"learning_rate": 9.89522736242292e-06,
|
| 425 |
+
"loss": 0.2393,
|
| 426 |
+
"num_input_tokens_seen": 79808,
|
| 427 |
+
"step": 255
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.25,
|
| 431 |
+
"grad_norm": 3.134875774383545,
|
| 432 |
+
"learning_rate": 9.883829406604363e-06,
|
| 433 |
+
"loss": 0.2281,
|
| 434 |
+
"num_input_tokens_seen": 81376,
|
| 435 |
+
"step": 260
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 3.3125,
|
| 439 |
+
"grad_norm": 12.125977516174316,
|
| 440 |
+
"learning_rate": 9.871850323926178e-06,
|
| 441 |
+
"loss": 0.2635,
|
| 442 |
+
"num_input_tokens_seen": 82912,
|
| 443 |
+
"step": 265
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 3.375,
|
| 447 |
+
"grad_norm": 1.7760015726089478,
|
| 448 |
+
"learning_rate": 9.859291539779407e-06,
|
| 449 |
+
"loss": 0.2369,
|
| 450 |
+
"num_input_tokens_seen": 84512,
|
| 451 |
+
"step": 270
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.4375,
|
| 455 |
+
"grad_norm": 1.7993685007095337,
|
| 456 |
+
"learning_rate": 9.846154548533773e-06,
|
| 457 |
+
"loss": 0.2255,
|
| 458 |
+
"num_input_tokens_seen": 86112,
|
| 459 |
+
"step": 275
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.5,
|
| 463 |
+
"grad_norm": 7.983632564544678,
|
| 464 |
+
"learning_rate": 9.83244091335986e-06,
|
| 465 |
+
"loss": 0.2446,
|
| 466 |
+
"num_input_tokens_seen": 87648,
|
| 467 |
+
"step": 280
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.5625,
|
| 471 |
+
"grad_norm": 4.111968517303467,
|
| 472 |
+
"learning_rate": 9.818152266043115e-06,
|
| 473 |
+
"loss": 0.2316,
|
| 474 |
+
"num_input_tokens_seen": 89280,
|
| 475 |
+
"step": 285
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 3.625,
|
| 479 |
+
"grad_norm": 4.560153007507324,
|
| 480 |
+
"learning_rate": 9.803290306789676e-06,
|
| 481 |
+
"loss": 0.2343,
|
| 482 |
+
"num_input_tokens_seen": 90816,
|
| 483 |
+
"step": 290
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 3.6875,
|
| 487 |
+
"grad_norm": 5.503631114959717,
|
| 488 |
+
"learning_rate": 9.787856804024073e-06,
|
| 489 |
+
"loss": 0.2224,
|
| 490 |
+
"num_input_tokens_seen": 92352,
|
| 491 |
+
"step": 295
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.75,
|
| 495 |
+
"grad_norm": 5.467092514038086,
|
| 496 |
+
"learning_rate": 9.771853594178791e-06,
|
| 497 |
+
"loss": 0.2373,
|
| 498 |
+
"num_input_tokens_seen": 93888,
|
| 499 |
+
"step": 300
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 3.8125,
|
| 503 |
+
"grad_norm": 4.816873550415039,
|
| 504 |
+
"learning_rate": 9.755282581475769e-06,
|
| 505 |
+
"loss": 0.2378,
|
| 506 |
+
"num_input_tokens_seen": 95424,
|
| 507 |
+
"step": 305
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 3.875,
|
| 511 |
+
"grad_norm": 0.8352380990982056,
|
| 512 |
+
"learning_rate": 9.7381457376998e-06,
|
| 513 |
+
"loss": 0.2402,
|
| 514 |
+
"num_input_tokens_seen": 96960,
|
| 515 |
+
"step": 310
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 3.9375,
|
| 519 |
+
"grad_norm": 1.431336760520935,
|
| 520 |
+
"learning_rate": 9.720445101963923e-06,
|
| 521 |
+
"loss": 0.2336,
|
| 522 |
+
"num_input_tokens_seen": 98528,
|
| 523 |
+
"step": 315
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 4.0,
|
| 527 |
+
"grad_norm": 0.8496450185775757,
|
| 528 |
+
"learning_rate": 9.702182780466775e-06,
|
| 529 |
+
"loss": 0.2424,
|
| 530 |
+
"num_input_tokens_seen": 100096,
|
| 531 |
+
"step": 320
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.0,
|
| 535 |
+
"eval_loss": 0.23625226318836212,
|
| 536 |
+
"eval_runtime": 0.9166,
|
| 537 |
+
"eval_samples_per_second": 87.277,
|
| 538 |
+
"eval_steps_per_second": 21.819,
|
| 539 |
+
"num_input_tokens_seen": 100096,
|
| 540 |
+
"step": 320
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 4.0625,
|
| 544 |
+
"grad_norm": 0.5525237321853638,
|
| 545 |
+
"learning_rate": 9.683360946241988e-06,
|
| 546 |
+
"loss": 0.2293,
|
| 547 |
+
"num_input_tokens_seen": 101664,
|
| 548 |
+
"step": 325
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 4.125,
|
| 552 |
+
"grad_norm": 1.4087265729904175,
|
| 553 |
+
"learning_rate": 9.663981838899612e-06,
|
| 554 |
+
"loss": 0.2239,
|
| 555 |
+
"num_input_tokens_seen": 103232,
|
| 556 |
+
"step": 330
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 4.1875,
|
| 560 |
+
"grad_norm": 0.8357235193252563,
|
| 561 |
+
"learning_rate": 9.644047764359623e-06,
|
| 562 |
+
"loss": 0.2219,
|
| 563 |
+
"num_input_tokens_seen": 104864,
|
| 564 |
+
"step": 335
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 4.25,
|
| 568 |
+
"grad_norm": 1.3893150091171265,
|
| 569 |
+
"learning_rate": 9.623561094577541e-06,
|
| 570 |
+
"loss": 0.2178,
|
| 571 |
+
"num_input_tokens_seen": 106496,
|
| 572 |
+
"step": 340
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 4.3125,
|
| 576 |
+
"grad_norm": 1.327685832977295,
|
| 577 |
+
"learning_rate": 9.602524267262202e-06,
|
| 578 |
+
"loss": 0.2373,
|
| 579 |
+
"num_input_tokens_seen": 108064,
|
| 580 |
+
"step": 345
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 4.375,
|
| 584 |
+
"grad_norm": 2.736804723739624,
|
| 585 |
+
"learning_rate": 9.58093978558568e-06,
|
| 586 |
+
"loss": 0.245,
|
| 587 |
+
"num_input_tokens_seen": 109600,
|
| 588 |
+
"step": 350
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 4.4375,
|
| 592 |
+
"grad_norm": 1.7261260747909546,
|
| 593 |
+
"learning_rate": 9.558810217885444e-06,
|
| 594 |
+
"loss": 0.2297,
|
| 595 |
+
"num_input_tokens_seen": 111200,
|
| 596 |
+
"step": 355
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 4.5,
|
| 600 |
+
"grad_norm": 1.1373929977416992,
|
| 601 |
+
"learning_rate": 9.536138197358747e-06,
|
| 602 |
+
"loss": 0.2371,
|
| 603 |
+
"num_input_tokens_seen": 112768,
|
| 604 |
+
"step": 360
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 4.5625,
|
| 608 |
+
"grad_norm": 0.8946912288665771,
|
| 609 |
+
"learning_rate": 9.512926421749305e-06,
|
| 610 |
+
"loss": 0.2346,
|
| 611 |
+
"num_input_tokens_seen": 114336,
|
| 612 |
+
"step": 365
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 4.625,
|
| 616 |
+
"grad_norm": 2.0754528045654297,
|
| 617 |
+
"learning_rate": 9.48917765302629e-06,
|
| 618 |
+
"loss": 0.2305,
|
| 619 |
+
"num_input_tokens_seen": 115872,
|
| 620 |
+
"step": 370
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 4.6875,
|
| 624 |
+
"grad_norm": 0.8747203350067139,
|
| 625 |
+
"learning_rate": 9.464894717055686e-06,
|
| 626 |
+
"loss": 0.2299,
|
| 627 |
+
"num_input_tokens_seen": 117440,
|
| 628 |
+
"step": 375
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 4.75,
|
| 632 |
+
"grad_norm": 1.415370225906372,
|
| 633 |
+
"learning_rate": 9.440080503264038e-06,
|
| 634 |
+
"loss": 0.2204,
|
| 635 |
+
"num_input_tokens_seen": 118976,
|
| 636 |
+
"step": 380
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 4.8125,
|
| 640 |
+
"grad_norm": 0.9907402396202087,
|
| 641 |
+
"learning_rate": 9.414737964294636e-06,
|
| 642 |
+
"loss": 0.2366,
|
| 643 |
+
"num_input_tokens_seen": 120544,
|
| 644 |
+
"step": 385
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 4.875,
|
| 648 |
+
"grad_norm": 2.0482561588287354,
|
| 649 |
+
"learning_rate": 9.388870115656185e-06,
|
| 650 |
+
"loss": 0.2541,
|
| 651 |
+
"num_input_tokens_seen": 122144,
|
| 652 |
+
"step": 390
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 4.9375,
|
| 656 |
+
"grad_norm": 1.3031115531921387,
|
| 657 |
+
"learning_rate": 9.362480035363987e-06,
|
| 658 |
+
"loss": 0.2257,
|
| 659 |
+
"num_input_tokens_seen": 123680,
|
| 660 |
+
"step": 395
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 5.0,
|
| 664 |
+
"grad_norm": 2.6485493183135986,
|
| 665 |
+
"learning_rate": 9.335570863573687e-06,
|
| 666 |
+
"loss": 0.235,
|
| 667 |
+
"num_input_tokens_seen": 125248,
|
| 668 |
+
"step": 400
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 5.0625,
|
| 672 |
+
"grad_norm": 2.2649688720703125,
|
| 673 |
+
"learning_rate": 9.30814580220763e-06,
|
| 674 |
+
"loss": 0.2452,
|
| 675 |
+
"num_input_tokens_seen": 126848,
|
| 676 |
+
"step": 405
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 5.125,
|
| 680 |
+
"grad_norm": 1.3141331672668457,
|
| 681 |
+
"learning_rate": 9.280208114573859e-06,
|
| 682 |
+
"loss": 0.2276,
|
| 683 |
+
"num_input_tokens_seen": 128448,
|
| 684 |
+
"step": 410
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 5.1875,
|
| 688 |
+
"grad_norm": 3.1463115215301514,
|
| 689 |
+
"learning_rate": 9.251761124977816e-06,
|
| 690 |
+
"loss": 0.2391,
|
| 691 |
+
"num_input_tokens_seen": 130048,
|
| 692 |
+
"step": 415
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 5.25,
|
| 696 |
+
"grad_norm": 1.473178744316101,
|
| 697 |
+
"learning_rate": 9.222808218326784e-06,
|
| 698 |
+
"loss": 0.2134,
|
| 699 |
+
"num_input_tokens_seen": 131616,
|
| 700 |
+
"step": 420
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 5.3125,
|
| 704 |
+
"grad_norm": 3.250842332839966,
|
| 705 |
+
"learning_rate": 9.193352839727122e-06,
|
| 706 |
+
"loss": 0.2114,
|
| 707 |
+
"num_input_tokens_seen": 133120,
|
| 708 |
+
"step": 425
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 5.375,
|
| 712 |
+
"grad_norm": 6.812272071838379,
|
| 713 |
+
"learning_rate": 9.163398494074314e-06,
|
| 714 |
+
"loss": 0.2555,
|
| 715 |
+
"num_input_tokens_seen": 134720,
|
| 716 |
+
"step": 430
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 5.4375,
|
| 720 |
+
"grad_norm": 5.645451068878174,
|
| 721 |
+
"learning_rate": 9.132948745635943e-06,
|
| 722 |
+
"loss": 0.2165,
|
| 723 |
+
"num_input_tokens_seen": 136288,
|
| 724 |
+
"step": 435
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 5.5,
|
| 728 |
+
"grad_norm": 2.791454315185547,
|
| 729 |
+
"learning_rate": 9.102007217627568e-06,
|
| 730 |
+
"loss": 0.2247,
|
| 731 |
+
"num_input_tokens_seen": 137920,
|
| 732 |
+
"step": 440
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 5.5625,
|
| 736 |
+
"grad_norm": 3.0683376789093018,
|
| 737 |
+
"learning_rate": 9.070577591781598e-06,
|
| 738 |
+
"loss": 0.2556,
|
| 739 |
+
"num_input_tokens_seen": 139488,
|
| 740 |
+
"step": 445
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 5.625,
|
| 744 |
+
"grad_norm": 1.0806782245635986,
|
| 745 |
+
"learning_rate": 9.038663607909198e-06,
|
| 746 |
+
"loss": 0.2418,
|
| 747 |
+
"num_input_tokens_seen": 141056,
|
| 748 |
+
"step": 450
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 5.6875,
|
| 752 |
+
"grad_norm": 3.1237070560455322,
|
| 753 |
+
"learning_rate": 9.006269063455305e-06,
|
| 754 |
+
"loss": 0.2406,
|
| 755 |
+
"num_input_tokens_seen": 142624,
|
| 756 |
+
"step": 455
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 5.75,
|
| 760 |
+
"grad_norm": 1.0548875331878662,
|
| 761 |
+
"learning_rate": 8.97339781304675e-06,
|
| 762 |
+
"loss": 0.2277,
|
| 763 |
+
"num_input_tokens_seen": 144128,
|
| 764 |
+
"step": 460
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 5.8125,
|
| 768 |
+
"grad_norm": 1.0157082080841064,
|
| 769 |
+
"learning_rate": 8.94005376803361e-06,
|
| 770 |
+
"loss": 0.2334,
|
| 771 |
+
"num_input_tokens_seen": 145696,
|
| 772 |
+
"step": 465
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 5.875,
|
| 776 |
+
"grad_norm": 0.5916211605072021,
|
| 777 |
+
"learning_rate": 8.906240896023794e-06,
|
| 778 |
+
"loss": 0.2352,
|
| 779 |
+
"num_input_tokens_seen": 147264,
|
| 780 |
+
"step": 470
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 5.9375,
|
| 784 |
+
"grad_norm": 1.234691858291626,
|
| 785 |
+
"learning_rate": 8.871963220410929e-06,
|
| 786 |
+
"loss": 0.2199,
|
| 787 |
+
"num_input_tokens_seen": 148864,
|
| 788 |
+
"step": 475
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 6.0,
|
| 792 |
+
"grad_norm": 1.6203759908676147,
|
| 793 |
+
"learning_rate": 8.837224819895627e-06,
|
| 794 |
+
"loss": 0.226,
|
| 795 |
+
"num_input_tokens_seen": 150400,
|
| 796 |
+
"step": 480
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 6.0,
|
| 800 |
+
"eval_loss": 0.24168923497200012,
|
| 801 |
+
"eval_runtime": 0.9165,
|
| 802 |
+
"eval_samples_per_second": 87.29,
|
| 803 |
+
"eval_steps_per_second": 21.823,
|
| 804 |
+
"num_input_tokens_seen": 150400,
|
| 805 |
+
"step": 480
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 6.0625,
|
| 809 |
+
"grad_norm": 2.004993438720703,
|
| 810 |
+
"learning_rate": 8.802029828000157e-06,
|
| 811 |
+
"loss": 0.2223,
|
| 812 |
+
"num_input_tokens_seen": 152000,
|
| 813 |
+
"step": 485
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 6.125,
|
| 817 |
+
"grad_norm": 3.092801332473755,
|
| 818 |
+
"learning_rate": 8.766382432576589e-06,
|
| 819 |
+
"loss": 0.237,
|
| 820 |
+
"num_input_tokens_seen": 153568,
|
| 821 |
+
"step": 490
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 6.1875,
|
| 825 |
+
"grad_norm": 1.6768417358398438,
|
| 826 |
+
"learning_rate": 8.730286875308498e-06,
|
| 827 |
+
"loss": 0.2362,
|
| 828 |
+
"num_input_tokens_seen": 155168,
|
| 829 |
+
"step": 495
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 6.25,
|
| 833 |
+
"grad_norm": 2.1835248470306396,
|
| 834 |
+
"learning_rate": 8.693747451206231e-06,
|
| 835 |
+
"loss": 0.2231,
|
| 836 |
+
"num_input_tokens_seen": 156704,
|
| 837 |
+
"step": 500
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 6.3125,
|
| 841 |
+
"grad_norm": 2.2564566135406494,
|
| 842 |
+
"learning_rate": 8.656768508095853e-06,
|
| 843 |
+
"loss": 0.2062,
|
| 844 |
+
"num_input_tokens_seen": 158176,
|
| 845 |
+
"step": 505
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 6.375,
|
| 849 |
+
"grad_norm": 1.0601657629013062,
|
| 850 |
+
"learning_rate": 8.61935444610179e-06,
|
| 851 |
+
"loss": 0.2769,
|
| 852 |
+
"num_input_tokens_seen": 159776,
|
| 853 |
+
"step": 510
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 6.4375,
|
| 857 |
+
"grad_norm": 0.9735612869262695,
|
| 858 |
+
"learning_rate": 8.581509717123272e-06,
|
| 859 |
+
"loss": 0.2179,
|
| 860 |
+
"num_input_tokens_seen": 161280,
|
| 861 |
+
"step": 515
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 6.5,
|
| 865 |
+
"grad_norm": 1.7658456563949585,
|
| 866 |
+
"learning_rate": 8.543238824304585e-06,
|
| 867 |
+
"loss": 0.2378,
|
| 868 |
+
"num_input_tokens_seen": 162880,
|
| 869 |
+
"step": 520
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 6.5625,
|
| 873 |
+
"grad_norm": 0.9902642369270325,
|
| 874 |
+
"learning_rate": 8.504546321499255e-06,
|
| 875 |
+
"loss": 0.2241,
|
| 876 |
+
"num_input_tokens_seen": 164448,
|
| 877 |
+
"step": 525
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 6.625,
|
| 881 |
+
"grad_norm": 2.589862585067749,
|
| 882 |
+
"learning_rate": 8.465436812728181e-06,
|
| 883 |
+
"loss": 0.2337,
|
| 884 |
+
"num_input_tokens_seen": 166048,
|
| 885 |
+
"step": 530
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 6.6875,
|
| 889 |
+
"grad_norm": 1.0508556365966797,
|
| 890 |
+
"learning_rate": 8.425914951631796e-06,
|
| 891 |
+
"loss": 0.2182,
|
| 892 |
+
"num_input_tokens_seen": 167616,
|
| 893 |
+
"step": 535
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 6.75,
|
| 897 |
+
"grad_norm": 1.600427269935608,
|
| 898 |
+
"learning_rate": 8.385985440916344e-06,
|
| 899 |
+
"loss": 0.2315,
|
| 900 |
+
"num_input_tokens_seen": 169184,
|
| 901 |
+
"step": 540
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 6.8125,
|
| 905 |
+
"grad_norm": 3.7025911808013916,
|
| 906 |
+
"learning_rate": 8.345653031794292e-06,
|
| 907 |
+
"loss": 0.2306,
|
| 908 |
+
"num_input_tokens_seen": 170816,
|
| 909 |
+
"step": 545
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 6.875,
|
| 913 |
+
"grad_norm": 1.6951884031295776,
|
| 914 |
+
"learning_rate": 8.304922523418988e-06,
|
| 915 |
+
"loss": 0.2025,
|
| 916 |
+
"num_input_tokens_seen": 172416,
|
| 917 |
+
"step": 550
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 6.9375,
|
| 921 |
+
"grad_norm": 2.9659578800201416,
|
| 922 |
+
"learning_rate": 8.263798762313613e-06,
|
| 923 |
+
"loss": 0.211,
|
| 924 |
+
"num_input_tokens_seen": 173952,
|
| 925 |
+
"step": 555
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 7.0,
|
| 929 |
+
"grad_norm": 4.885127067565918,
|
| 930 |
+
"learning_rate": 8.222286641794488e-06,
|
| 931 |
+
"loss": 0.2373,
|
| 932 |
+
"num_input_tokens_seen": 175520,
|
| 933 |
+
"step": 560
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 7.0625,
|
| 937 |
+
"grad_norm": 12.386910438537598,
|
| 938 |
+
"learning_rate": 8.18039110138882e-06,
|
| 939 |
+
"loss": 0.2171,
|
| 940 |
+
"num_input_tokens_seen": 177120,
|
| 941 |
+
"step": 565
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 7.125,
|
| 945 |
+
"grad_norm": 4.289523124694824,
|
| 946 |
+
"learning_rate": 8.138117126246951e-06,
|
| 947 |
+
"loss": 0.2133,
|
| 948 |
+
"num_input_tokens_seen": 178720,
|
| 949 |
+
"step": 570
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 7.1875,
|
| 953 |
+
"grad_norm": 3.2546839714050293,
|
| 954 |
+
"learning_rate": 8.095469746549172e-06,
|
| 955 |
+
"loss": 0.2257,
|
| 956 |
+
"num_input_tokens_seen": 180288,
|
| 957 |
+
"step": 575
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 7.25,
|
| 961 |
+
"grad_norm": 3.431429862976074,
|
| 962 |
+
"learning_rate": 8.052454036907174e-06,
|
| 963 |
+
"loss": 0.1958,
|
| 964 |
+
"num_input_tokens_seen": 181888,
|
| 965 |
+
"step": 580
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 7.3125,
|
| 969 |
+
"grad_norm": 2.91300106048584,
|
| 970 |
+
"learning_rate": 8.009075115760243e-06,
|
| 971 |
+
"loss": 0.2011,
|
| 972 |
+
"num_input_tokens_seen": 183456,
|
| 973 |
+
"step": 585
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 7.375,
|
| 977 |
+
"grad_norm": 3.276212453842163,
|
| 978 |
+
"learning_rate": 7.965338144766186e-06,
|
| 979 |
+
"loss": 0.2053,
|
| 980 |
+
"num_input_tokens_seen": 185024,
|
| 981 |
+
"step": 590
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 7.4375,
|
| 985 |
+
"grad_norm": 7.24328088760376,
|
| 986 |
+
"learning_rate": 7.921248328187174e-06,
|
| 987 |
+
"loss": 0.2549,
|
| 988 |
+
"num_input_tokens_seen": 186592,
|
| 989 |
+
"step": 595
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 7.5,
|
| 993 |
+
"grad_norm": 6.893558979034424,
|
| 994 |
+
"learning_rate": 7.876810912270462e-06,
|
| 995 |
+
"loss": 0.237,
|
| 996 |
+
"num_input_tokens_seen": 188128,
|
| 997 |
+
"step": 600
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 7.5625,
|
| 1001 |
+
"grad_norm": 5.550233364105225,
|
| 1002 |
+
"learning_rate": 7.832031184624165e-06,
|
| 1003 |
+
"loss": 0.2381,
|
| 1004 |
+
"num_input_tokens_seen": 189664,
|
| 1005 |
+
"step": 605
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 7.625,
|
| 1009 |
+
"grad_norm": 3.1067495346069336,
|
| 1010 |
+
"learning_rate": 7.786914473588057e-06,
|
| 1011 |
+
"loss": 0.1874,
|
| 1012 |
+
"num_input_tokens_seen": 191232,
|
| 1013 |
+
"step": 610
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 7.6875,
|
| 1017 |
+
"grad_norm": 7.392305850982666,
|
| 1018 |
+
"learning_rate": 7.74146614759957e-06,
|
| 1019 |
+
"loss": 0.2148,
|
| 1020 |
+
"num_input_tokens_seen": 192800,
|
| 1021 |
+
"step": 615
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 7.75,
|
| 1025 |
+
"grad_norm": 6.367889404296875,
|
| 1026 |
+
"learning_rate": 7.695691614555002e-06,
|
| 1027 |
+
"loss": 0.1938,
|
| 1028 |
+
"num_input_tokens_seen": 194368,
|
| 1029 |
+
"step": 620
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 7.8125,
|
| 1033 |
+
"grad_norm": 3.6397085189819336,
|
| 1034 |
+
"learning_rate": 7.649596321166024e-06,
|
| 1035 |
+
"loss": 0.2583,
|
| 1036 |
+
"num_input_tokens_seen": 195936,
|
| 1037 |
+
"step": 625
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 7.875,
|
| 1041 |
+
"grad_norm": 4.538295269012451,
|
| 1042 |
+
"learning_rate": 7.603185752311587e-06,
|
| 1043 |
+
"loss": 0.2361,
|
| 1044 |
+
"num_input_tokens_seen": 197568,
|
| 1045 |
+
"step": 630
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 7.9375,
|
| 1049 |
+
"grad_norm": 2.7171273231506348,
|
| 1050 |
+
"learning_rate": 7.55646543038526e-06,
|
| 1051 |
+
"loss": 0.2024,
|
| 1052 |
+
"num_input_tokens_seen": 199136,
|
| 1053 |
+
"step": 635
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 8.0,
|
| 1057 |
+
"grad_norm": 3.4454898834228516,
|
| 1058 |
+
"learning_rate": 7.50944091463814e-06,
|
| 1059 |
+
"loss": 0.203,
|
| 1060 |
+
"num_input_tokens_seen": 200704,
|
| 1061 |
+
"step": 640
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 8.0,
|
| 1065 |
+
"eval_loss": 0.2646110951900482,
|
| 1066 |
+
"eval_runtime": 0.9235,
|
| 1067 |
+
"eval_samples_per_second": 86.628,
|
| 1068 |
+
"eval_steps_per_second": 21.657,
|
| 1069 |
+
"num_input_tokens_seen": 200704,
|
| 1070 |
+
"step": 640
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 8.0625,
|
| 1074 |
+
"grad_norm": 9.628674507141113,
|
| 1075 |
+
"learning_rate": 7.462117800517337e-06,
|
| 1076 |
+
"loss": 0.2106,
|
| 1077 |
+
"num_input_tokens_seen": 202272,
|
| 1078 |
+
"step": 645
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 8.125,
|
| 1082 |
+
"grad_norm": 4.698387622833252,
|
| 1083 |
+
"learning_rate": 7.414501719000187e-06,
|
| 1084 |
+
"loss": 0.1585,
|
| 1085 |
+
"num_input_tokens_seen": 203872,
|
| 1086 |
+
"step": 650
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 8.1875,
|
| 1090 |
+
"grad_norm": 14.213072776794434,
|
| 1091 |
+
"learning_rate": 7.3665983359242175e-06,
|
| 1092 |
+
"loss": 0.1878,
|
| 1093 |
+
"num_input_tokens_seen": 205408,
|
| 1094 |
+
"step": 655
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 8.25,
|
| 1098 |
+
"grad_norm": 18.2841854095459,
|
| 1099 |
+
"learning_rate": 7.318413351312965e-06,
|
| 1100 |
+
"loss": 0.2258,
|
| 1101 |
+
"num_input_tokens_seen": 207040,
|
| 1102 |
+
"step": 660
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 8.3125,
|
| 1106 |
+
"grad_norm": 12.26921558380127,
|
| 1107 |
+
"learning_rate": 7.269952498697734e-06,
|
| 1108 |
+
"loss": 0.213,
|
| 1109 |
+
"num_input_tokens_seen": 208608,
|
| 1110 |
+
"step": 665
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 8.375,
|
| 1114 |
+
"grad_norm": 14.23438549041748,
|
| 1115 |
+
"learning_rate": 7.221221544435364e-06,
|
| 1116 |
+
"loss": 0.2241,
|
| 1117 |
+
"num_input_tokens_seen": 210240,
|
| 1118 |
+
"step": 670
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 8.4375,
|
| 1122 |
+
"grad_norm": 6.5942487716674805,
|
| 1123 |
+
"learning_rate": 7.172226287022086e-06,
|
| 1124 |
+
"loss": 0.2082,
|
| 1125 |
+
"num_input_tokens_seen": 211744,
|
| 1126 |
+
"step": 675
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 8.5,
|
| 1130 |
+
"grad_norm": 6.377285957336426,
|
| 1131 |
+
"learning_rate": 7.1229725564035665e-06,
|
| 1132 |
+
"loss": 0.1792,
|
| 1133 |
+
"num_input_tokens_seen": 213280,
|
| 1134 |
+
"step": 680
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 8.5625,
|
| 1138 |
+
"grad_norm": 5.1635966300964355,
|
| 1139 |
+
"learning_rate": 7.073466213281196e-06,
|
| 1140 |
+
"loss": 0.1922,
|
| 1141 |
+
"num_input_tokens_seen": 214816,
|
| 1142 |
+
"step": 685
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 8.625,
|
| 1146 |
+
"grad_norm": 12.18478012084961,
|
| 1147 |
+
"learning_rate": 7.023713148414728e-06,
|
| 1148 |
+
"loss": 0.2235,
|
| 1149 |
+
"num_input_tokens_seen": 216416,
|
| 1150 |
+
"step": 690
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 8.6875,
|
| 1154 |
+
"grad_norm": 5.056797504425049,
|
| 1155 |
+
"learning_rate": 6.973719281921336e-06,
|
| 1156 |
+
"loss": 0.2177,
|
| 1157 |
+
"num_input_tokens_seen": 217952,
|
| 1158 |
+
"step": 695
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 8.75,
|
| 1162 |
+
"grad_norm": 6.05808162689209,
|
| 1163 |
+
"learning_rate": 6.9234905625711816e-06,
|
| 1164 |
+
"loss": 0.2071,
|
| 1165 |
+
"num_input_tokens_seen": 219552,
|
| 1166 |
+
"step": 700
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 8.8125,
|
| 1170 |
+
"grad_norm": 4.5069499015808105,
|
| 1171 |
+
"learning_rate": 6.873032967079562e-06,
|
| 1172 |
+
"loss": 0.177,
|
| 1173 |
+
"num_input_tokens_seen": 221120,
|
| 1174 |
+
"step": 705
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 8.875,
|
| 1178 |
+
"grad_norm": 9.237682342529297,
|
| 1179 |
+
"learning_rate": 6.822352499395751e-06,
|
| 1180 |
+
"loss": 0.188,
|
| 1181 |
+
"num_input_tokens_seen": 222656,
|
| 1182 |
+
"step": 710
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 8.9375,
|
| 1186 |
+
"grad_norm": 9.70540714263916,
|
| 1187 |
+
"learning_rate": 6.771455189988579e-06,
|
| 1188 |
+
"loss": 0.1792,
|
| 1189 |
+
"num_input_tokens_seen": 224160,
|
| 1190 |
+
"step": 715
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 9.0,
|
| 1194 |
+
"grad_norm": 29.520042419433594,
|
| 1195 |
+
"learning_rate": 6.720347095128884e-06,
|
| 1196 |
+
"loss": 0.248,
|
| 1197 |
+
"num_input_tokens_seen": 225728,
|
| 1198 |
+
"step": 720
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 9.0625,
|
| 1202 |
+
"grad_norm": 6.013643741607666,
|
| 1203 |
+
"learning_rate": 6.669034296168855e-06,
|
| 1204 |
+
"loss": 0.1972,
|
| 1205 |
+
"num_input_tokens_seen": 227296,
|
| 1206 |
+
"step": 725
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 9.125,
|
| 1210 |
+
"grad_norm": 8.753713607788086,
|
| 1211 |
+
"learning_rate": 6.617522898818426e-06,
|
| 1212 |
+
"loss": 0.1308,
|
| 1213 |
+
"num_input_tokens_seen": 228896,
|
| 1214 |
+
"step": 730
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 9.1875,
|
| 1218 |
+
"grad_norm": 32.677330017089844,
|
| 1219 |
+
"learning_rate": 6.565819032418748e-06,
|
| 1220 |
+
"loss": 0.2039,
|
| 1221 |
+
"num_input_tokens_seen": 230464,
|
| 1222 |
+
"step": 735
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 9.25,
|
| 1226 |
+
"grad_norm": 12.465975761413574,
|
| 1227 |
+
"learning_rate": 6.513928849212874e-06,
|
| 1228 |
+
"loss": 0.1759,
|
| 1229 |
+
"num_input_tokens_seen": 232032,
|
| 1230 |
+
"step": 740
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 9.3125,
|
| 1234 |
+
"grad_norm": 25.805383682250977,
|
| 1235 |
+
"learning_rate": 6.461858523613684e-06,
|
| 1236 |
+
"loss": 0.1213,
|
| 1237 |
+
"num_input_tokens_seen": 233632,
|
| 1238 |
+
"step": 745
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 9.375,
|
| 1242 |
+
"grad_norm": 10.997842788696289,
|
| 1243 |
+
"learning_rate": 6.4096142514692085e-06,
|
| 1244 |
+
"loss": 0.1275,
|
| 1245 |
+
"num_input_tokens_seen": 235200,
|
| 1246 |
+
"step": 750
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 9.4375,
|
| 1250 |
+
"grad_norm": 16.847126007080078,
|
| 1251 |
+
"learning_rate": 6.3572022493253715e-06,
|
| 1252 |
+
"loss": 0.2011,
|
| 1253 |
+
"num_input_tokens_seen": 236704,
|
| 1254 |
+
"step": 755
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 9.5,
|
| 1258 |
+
"grad_norm": 15.590519905090332,
|
| 1259 |
+
"learning_rate": 6.304628753686295e-06,
|
| 1260 |
+
"loss": 0.1765,
|
| 1261 |
+
"num_input_tokens_seen": 238240,
|
| 1262 |
+
"step": 760
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 9.5625,
|
| 1266 |
+
"grad_norm": 8.250570297241211,
|
| 1267 |
+
"learning_rate": 6.251900020272208e-06,
|
| 1268 |
+
"loss": 0.1771,
|
| 1269 |
+
"num_input_tokens_seen": 239776,
|
| 1270 |
+
"step": 765
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 9.625,
|
| 1274 |
+
"grad_norm": 11.40953254699707,
|
| 1275 |
+
"learning_rate": 6.199022323275083e-06,
|
| 1276 |
+
"loss": 0.1629,
|
| 1277 |
+
"num_input_tokens_seen": 241280,
|
| 1278 |
+
"step": 770
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 9.6875,
|
| 1282 |
+
"grad_norm": 10.189095497131348,
|
| 1283 |
+
"learning_rate": 6.146001954612072e-06,
|
| 1284 |
+
"loss": 0.1316,
|
| 1285 |
+
"num_input_tokens_seen": 242752,
|
| 1286 |
+
"step": 775
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 9.75,
|
| 1290 |
+
"grad_norm": 15.832845687866211,
|
| 1291 |
+
"learning_rate": 6.092845223176823e-06,
|
| 1292 |
+
"loss": 0.1367,
|
| 1293 |
+
"num_input_tokens_seen": 244352,
|
| 1294 |
+
"step": 780
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 9.8125,
|
| 1298 |
+
"grad_norm": 14.90906810760498,
|
| 1299 |
+
"learning_rate": 6.039558454088796e-06,
|
| 1300 |
+
"loss": 0.1887,
|
| 1301 |
+
"num_input_tokens_seen": 245984,
|
| 1302 |
+
"step": 785
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 9.875,
|
| 1306 |
+
"grad_norm": 10.249608039855957,
|
| 1307 |
+
"learning_rate": 5.986147987940632e-06,
|
| 1308 |
+
"loss": 0.1589,
|
| 1309 |
+
"num_input_tokens_seen": 247552,
|
| 1310 |
+
"step": 790
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 9.9375,
|
| 1314 |
+
"grad_norm": 5.563095569610596,
|
| 1315 |
+
"learning_rate": 5.932620180043674e-06,
|
| 1316 |
+
"loss": 0.1707,
|
| 1317 |
+
"num_input_tokens_seen": 249088,
|
| 1318 |
+
"step": 795
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 10.0,
|
| 1322 |
+
"grad_norm": 9.259960174560547,
|
| 1323 |
+
"learning_rate": 5.878981399671774e-06,
|
| 1324 |
+
"loss": 0.2158,
|
| 1325 |
+
"num_input_tokens_seen": 250592,
|
| 1326 |
+
"step": 800
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 10.0,
|
| 1330 |
+
"eval_loss": 0.3806591033935547,
|
| 1331 |
+
"eval_runtime": 0.9182,
|
| 1332 |
+
"eval_samples_per_second": 87.124,
|
| 1333 |
+
"eval_steps_per_second": 21.781,
|
| 1334 |
+
"num_input_tokens_seen": 250592,
|
| 1335 |
+
"step": 800
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 10.0625,
|
| 1339 |
+
"grad_norm": 10.863539695739746,
|
| 1340 |
+
"learning_rate": 5.825238029303388e-06,
|
| 1341 |
+
"loss": 0.0837,
|
| 1342 |
+
"num_input_tokens_seen": 252256,
|
| 1343 |
+
"step": 805
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 10.125,
|
| 1347 |
+
"grad_norm": 8.517024993896484,
|
| 1348 |
+
"learning_rate": 5.771396463862145e-06,
|
| 1349 |
+
"loss": 0.0796,
|
| 1350 |
+
"num_input_tokens_seen": 253824,
|
| 1351 |
+
"step": 810
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 10.1875,
|
| 1355 |
+
"grad_norm": 9.531097412109375,
|
| 1356 |
+
"learning_rate": 5.717463109955896e-06,
|
| 1357 |
+
"loss": 0.0816,
|
| 1358 |
+
"num_input_tokens_seen": 255360,
|
| 1359 |
+
"step": 815
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 10.25,
|
| 1363 |
+
"grad_norm": 20.906747817993164,
|
| 1364 |
+
"learning_rate": 5.6634443851144115e-06,
|
| 1365 |
+
"loss": 0.0815,
|
| 1366 |
+
"num_input_tokens_seen": 256960,
|
| 1367 |
+
"step": 820
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 10.3125,
|
| 1371 |
+
"grad_norm": 21.023405075073242,
|
| 1372 |
+
"learning_rate": 5.609346717025738e-06,
|
| 1373 |
+
"loss": 0.1234,
|
| 1374 |
+
"num_input_tokens_seen": 258528,
|
| 1375 |
+
"step": 825
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 10.375,
|
| 1379 |
+
"grad_norm": 34.56439208984375,
|
| 1380 |
+
"learning_rate": 5.555176542771389e-06,
|
| 1381 |
+
"loss": 0.1462,
|
| 1382 |
+
"num_input_tokens_seen": 260096,
|
| 1383 |
+
"step": 830
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 10.4375,
|
| 1387 |
+
"grad_norm": 16.453866958618164,
|
| 1388 |
+
"learning_rate": 5.500940308060382e-06,
|
| 1389 |
+
"loss": 0.1031,
|
| 1390 |
+
"num_input_tokens_seen": 261632,
|
| 1391 |
+
"step": 835
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 10.5,
|
| 1395 |
+
"grad_norm": 20.025487899780273,
|
| 1396 |
+
"learning_rate": 5.446644466462269e-06,
|
| 1397 |
+
"loss": 0.1215,
|
| 1398 |
+
"num_input_tokens_seen": 263232,
|
| 1399 |
+
"step": 840
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 10.5625,
|
| 1403 |
+
"grad_norm": 21.75179672241211,
|
| 1404 |
+
"learning_rate": 5.392295478639226e-06,
|
| 1405 |
+
"loss": 0.2008,
|
| 1406 |
+
"num_input_tokens_seen": 264832,
|
| 1407 |
+
"step": 845
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 10.625,
|
| 1411 |
+
"grad_norm": 26.04926300048828,
|
| 1412 |
+
"learning_rate": 5.337899811577297e-06,
|
| 1413 |
+
"loss": 0.1439,
|
| 1414 |
+
"num_input_tokens_seen": 266432,
|
| 1415 |
+
"step": 850
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 10.6875,
|
| 1419 |
+
"grad_norm": 6.14185094833374,
|
| 1420 |
+
"learning_rate": 5.283463937816888e-06,
|
| 1421 |
+
"loss": 0.0927,
|
| 1422 |
+
"num_input_tokens_seen": 268000,
|
| 1423 |
+
"step": 855
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 10.75,
|
| 1427 |
+
"grad_norm": 7.8607635498046875,
|
| 1428 |
+
"learning_rate": 5.228994334682605e-06,
|
| 1429 |
+
"loss": 0.1166,
|
| 1430 |
+
"num_input_tokens_seen": 269568,
|
| 1431 |
+
"step": 860
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 10.8125,
|
| 1435 |
+
"grad_norm": 14.155070304870605,
|
| 1436 |
+
"learning_rate": 5.174497483512506e-06,
|
| 1437 |
+
"loss": 0.1066,
|
| 1438 |
+
"num_input_tokens_seen": 271104,
|
| 1439 |
+
"step": 865
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 10.875,
|
| 1443 |
+
"grad_norm": 14.761529922485352,
|
| 1444 |
+
"learning_rate": 5.1199798688868955e-06,
|
| 1445 |
+
"loss": 0.1839,
|
| 1446 |
+
"num_input_tokens_seen": 272640,
|
| 1447 |
+
"step": 870
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 10.9375,
|
| 1451 |
+
"grad_norm": 86.08441162109375,
|
| 1452 |
+
"learning_rate": 5.065447977856723e-06,
|
| 1453 |
+
"loss": 0.2267,
|
| 1454 |
+
"num_input_tokens_seen": 274208,
|
| 1455 |
+
"step": 875
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 11.0,
|
| 1459 |
+
"grad_norm": 6.295602798461914,
|
| 1460 |
+
"learning_rate": 5.010908299171685e-06,
|
| 1461 |
+
"loss": 0.195,
|
| 1462 |
+
"num_input_tokens_seen": 275776,
|
| 1463 |
+
"step": 880
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 11.0625,
|
| 1467 |
+
"grad_norm": 29.30211067199707,
|
| 1468 |
+
"learning_rate": 4.956367322508131e-06,
|
| 1469 |
+
"loss": 0.0878,
|
| 1470 |
+
"num_input_tokens_seen": 277344,
|
| 1471 |
+
"step": 885
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 11.125,
|
| 1475 |
+
"grad_norm": 36.51382827758789,
|
| 1476 |
+
"learning_rate": 4.90183153769686e-06,
|
| 1477 |
+
"loss": 0.097,
|
| 1478 |
+
"num_input_tokens_seen": 278944,
|
| 1479 |
+
"step": 890
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 11.1875,
|
| 1483 |
+
"grad_norm": 24.661712646484375,
|
| 1484 |
+
"learning_rate": 4.847307433950888e-06,
|
| 1485 |
+
"loss": 0.1409,
|
| 1486 |
+
"num_input_tokens_seen": 280480,
|
| 1487 |
+
"step": 895
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 11.25,
|
| 1491 |
+
"grad_norm": 15.945549011230469,
|
| 1492 |
+
"learning_rate": 4.792801499093305e-06,
|
| 1493 |
+
"loss": 0.0621,
|
| 1494 |
+
"num_input_tokens_seen": 282048,
|
| 1495 |
+
"step": 900
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 11.3125,
|
| 1499 |
+
"grad_norm": 6.951054573059082,
|
| 1500 |
+
"learning_rate": 4.738320218785281e-06,
|
| 1501 |
+
"loss": 0.1243,
|
| 1502 |
+
"num_input_tokens_seen": 283584,
|
| 1503 |
+
"step": 905
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 11.375,
|
| 1507 |
+
"grad_norm": 10.031401634216309,
|
| 1508 |
+
"learning_rate": 4.683870075754347e-06,
|
| 1509 |
+
"loss": 0.0546,
|
| 1510 |
+
"num_input_tokens_seen": 285216,
|
| 1511 |
+
"step": 910
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 11.4375,
|
| 1515 |
+
"grad_norm": 10.544556617736816,
|
| 1516 |
+
"learning_rate": 4.629457549023004e-06,
|
| 1517 |
+
"loss": 0.0756,
|
| 1518 |
+
"num_input_tokens_seen": 286784,
|
| 1519 |
+
"step": 915
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 11.5,
|
| 1523 |
+
"grad_norm": 13.169859886169434,
|
| 1524 |
+
"learning_rate": 4.575089113137792e-06,
|
| 1525 |
+
"loss": 0.0427,
|
| 1526 |
+
"num_input_tokens_seen": 288352,
|
| 1527 |
+
"step": 920
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 11.5625,
|
| 1531 |
+
"grad_norm": 9.484841346740723,
|
| 1532 |
+
"learning_rate": 4.52077123739888e-06,
|
| 1533 |
+
"loss": 0.1803,
|
| 1534 |
+
"num_input_tokens_seen": 289920,
|
| 1535 |
+
"step": 925
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 11.625,
|
| 1539 |
+
"grad_norm": 32.46955871582031,
|
| 1540 |
+
"learning_rate": 4.466510385090287e-06,
|
| 1541 |
+
"loss": 0.1306,
|
| 1542 |
+
"num_input_tokens_seen": 291520,
|
| 1543 |
+
"step": 930
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 11.6875,
|
| 1547 |
+
"grad_norm": 13.81043529510498,
|
| 1548 |
+
"learning_rate": 4.4123130127108125e-06,
|
| 1549 |
+
"loss": 0.0687,
|
| 1550 |
+
"num_input_tokens_seen": 293056,
|
| 1551 |
+
"step": 935
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 11.75,
|
| 1555 |
+
"grad_norm": 52.68048858642578,
|
| 1556 |
+
"learning_rate": 4.358185569205779e-06,
|
| 1557 |
+
"loss": 0.1427,
|
| 1558 |
+
"num_input_tokens_seen": 294624,
|
| 1559 |
+
"step": 940
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 11.8125,
|
| 1563 |
+
"grad_norm": 14.252890586853027,
|
| 1564 |
+
"learning_rate": 4.304134495199675e-06,
|
| 1565 |
+
"loss": 0.052,
|
| 1566 |
+
"num_input_tokens_seen": 296160,
|
| 1567 |
+
"step": 945
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 11.875,
|
| 1571 |
+
"grad_norm": 28.040077209472656,
|
| 1572 |
+
"learning_rate": 4.250166222229775e-06,
|
| 1573 |
+
"loss": 0.1382,
|
| 1574 |
+
"num_input_tokens_seen": 297696,
|
| 1575 |
+
"step": 950
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 11.9375,
|
| 1579 |
+
"grad_norm": 10.452444076538086,
|
| 1580 |
+
"learning_rate": 4.196287171980869e-06,
|
| 1581 |
+
"loss": 0.0832,
|
| 1582 |
+
"num_input_tokens_seen": 299296,
|
| 1583 |
+
"step": 955
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 12.0,
|
| 1587 |
+
"grad_norm": 0.764223575592041,
|
| 1588 |
+
"learning_rate": 4.142503755521129e-06,
|
| 1589 |
+
"loss": 0.013,
|
| 1590 |
+
"num_input_tokens_seen": 300832,
|
| 1591 |
+
"step": 960
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 12.0,
|
| 1595 |
+
"eval_loss": 0.5269124507904053,
|
| 1596 |
+
"eval_runtime": 0.9175,
|
| 1597 |
+
"eval_samples_per_second": 87.197,
|
| 1598 |
+
"eval_steps_per_second": 21.799,
|
| 1599 |
+
"num_input_tokens_seen": 300832,
|
| 1600 |
+
"step": 960
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 12.0625,
|
| 1604 |
+
"grad_norm": 31.75777816772461,
|
| 1605 |
+
"learning_rate": 4.088822372539263e-06,
|
| 1606 |
+
"loss": 0.0538,
|
| 1607 |
+
"num_input_tokens_seen": 302368,
|
| 1608 |
+
"step": 965
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 12.125,
|
| 1612 |
+
"grad_norm": 1.6154866218566895,
|
| 1613 |
+
"learning_rate": 4.0352494105830155e-06,
|
| 1614 |
+
"loss": 0.0165,
|
| 1615 |
+
"num_input_tokens_seen": 303936,
|
| 1616 |
+
"step": 970
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 12.1875,
|
| 1620 |
+
"grad_norm": 3.850043296813965,
|
| 1621 |
+
"learning_rate": 3.981791244299113e-06,
|
| 1622 |
+
"loss": 0.0048,
|
| 1623 |
+
"num_input_tokens_seen": 305536,
|
| 1624 |
+
"step": 975
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 12.25,
|
| 1628 |
+
"grad_norm": 48.39171600341797,
|
| 1629 |
+
"learning_rate": 3.928454234674748e-06,
|
| 1630 |
+
"loss": 0.0607,
|
| 1631 |
+
"num_input_tokens_seen": 307136,
|
| 1632 |
+
"step": 980
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 12.3125,
|
| 1636 |
+
"grad_norm": 2.30706787109375,
|
| 1637 |
+
"learning_rate": 3.875244728280676e-06,
|
| 1638 |
+
"loss": 0.0606,
|
| 1639 |
+
"num_input_tokens_seen": 308672,
|
| 1640 |
+
"step": 985
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 12.375,
|
| 1644 |
+
"grad_norm": 39.70808029174805,
|
| 1645 |
+
"learning_rate": 3.822169056516051e-06,
|
| 1646 |
+
"loss": 0.0663,
|
| 1647 |
+
"num_input_tokens_seen": 310272,
|
| 1648 |
+
"step": 990
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 12.4375,
|
| 1652 |
+
"grad_norm": 28.03505516052246,
|
| 1653 |
+
"learning_rate": 3.769233534855035e-06,
|
| 1654 |
+
"loss": 0.0959,
|
| 1655 |
+
"num_input_tokens_seen": 311840,
|
| 1656 |
+
"step": 995
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 12.5,
|
| 1660 |
+
"grad_norm": 47.5093879699707,
|
| 1661 |
+
"learning_rate": 3.7164444620953397e-06,
|
| 1662 |
+
"loss": 0.0328,
|
| 1663 |
+
"num_input_tokens_seen": 313376,
|
| 1664 |
+
"step": 1000
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 12.5625,
|
| 1668 |
+
"grad_norm": 73.91315460205078,
|
| 1669 |
+
"learning_rate": 3.663808119608716e-06,
|
| 1670 |
+
"loss": 0.1293,
|
| 1671 |
+
"num_input_tokens_seen": 314976,
|
| 1672 |
+
"step": 1005
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 12.625,
|
| 1676 |
+
"grad_norm": 60.9661865234375,
|
| 1677 |
+
"learning_rate": 3.6113307705935398e-06,
|
| 1678 |
+
"loss": 0.1331,
|
| 1679 |
+
"num_input_tokens_seen": 316608,
|
| 1680 |
+
"step": 1010
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 12.6875,
|
| 1684 |
+
"grad_norm": 34.23344421386719,
|
| 1685 |
+
"learning_rate": 3.559018659329554e-06,
|
| 1686 |
+
"loss": 0.0214,
|
| 1687 |
+
"num_input_tokens_seen": 318176,
|
| 1688 |
+
"step": 1015
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 12.75,
|
| 1692 |
+
"grad_norm": 68.91053771972656,
|
| 1693 |
+
"learning_rate": 3.5068780104348632e-06,
|
| 1694 |
+
"loss": 0.2041,
|
| 1695 |
+
"num_input_tokens_seen": 319744,
|
| 1696 |
+
"step": 1020
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 12.8125,
|
| 1700 |
+
"grad_norm": 12.771360397338867,
|
| 1701 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1702 |
+
"loss": 0.0461,
|
| 1703 |
+
"num_input_tokens_seen": 321312,
|
| 1704 |
+
"step": 1025
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 12.875,
|
| 1708 |
+
"grad_norm": 11.279967308044434,
|
| 1709 |
+
"learning_rate": 3.403135895476004e-06,
|
| 1710 |
+
"loss": 0.0526,
|
| 1711 |
+
"num_input_tokens_seen": 322816,
|
| 1712 |
+
"step": 1030
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 12.9375,
|
| 1716 |
+
"grad_norm": 36.483665466308594,
|
| 1717 |
+
"learning_rate": 3.351546773686065e-06,
|
| 1718 |
+
"loss": 0.0837,
|
| 1719 |
+
"num_input_tokens_seen": 324352,
|
| 1720 |
+
"step": 1035
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 13.0,
|
| 1724 |
+
"grad_norm": 5.231341361999512,
|
| 1725 |
+
"learning_rate": 3.3001538013450285e-06,
|
| 1726 |
+
"loss": 0.0695,
|
| 1727 |
+
"num_input_tokens_seen": 325888,
|
| 1728 |
+
"step": 1040
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 13.0625,
|
| 1732 |
+
"grad_norm": 0.3779163956642151,
|
| 1733 |
+
"learning_rate": 3.248963093702663e-06,
|
| 1734 |
+
"loss": 0.0415,
|
| 1735 |
+
"num_input_tokens_seen": 327456,
|
| 1736 |
+
"step": 1045
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 13.125,
|
| 1740 |
+
"grad_norm": 2.168973207473755,
|
| 1741 |
+
"learning_rate": 3.1979807419412523e-06,
|
| 1742 |
+
"loss": 0.013,
|
| 1743 |
+
"num_input_tokens_seen": 329056,
|
| 1744 |
+
"step": 1050
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 13.1875,
|
| 1748 |
+
"grad_norm": 9.570158958435059,
|
| 1749 |
+
"learning_rate": 3.147212812450819e-06,
|
| 1750 |
+
"loss": 0.0101,
|
| 1751 |
+
"num_input_tokens_seen": 330624,
|
| 1752 |
+
"step": 1055
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 13.25,
|
| 1756 |
+
"grad_norm": 13.025811195373535,
|
| 1757 |
+
"learning_rate": 3.0966653461072778e-06,
|
| 1758 |
+
"loss": 0.005,
|
| 1759 |
+
"num_input_tokens_seen": 332192,
|
| 1760 |
+
"step": 1060
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 13.3125,
|
| 1764 |
+
"grad_norm": 0.06026485562324524,
|
| 1765 |
+
"learning_rate": 3.0463443575536324e-06,
|
| 1766 |
+
"loss": 0.0222,
|
| 1767 |
+
"num_input_tokens_seen": 333760,
|
| 1768 |
+
"step": 1065
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 13.375,
|
| 1772 |
+
"grad_norm": 2.276689052581787,
|
| 1773 |
+
"learning_rate": 2.9962558344842963e-06,
|
| 1774 |
+
"loss": 0.05,
|
| 1775 |
+
"num_input_tokens_seen": 335328,
|
| 1776 |
+
"step": 1070
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 13.4375,
|
| 1780 |
+
"grad_norm": 0.11900363117456436,
|
| 1781 |
+
"learning_rate": 2.946405736932615e-06,
|
| 1782 |
+
"loss": 0.0244,
|
| 1783 |
+
"num_input_tokens_seen": 336864,
|
| 1784 |
+
"step": 1075
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 13.5,
|
| 1788 |
+
"grad_norm": 2.8280792236328125,
|
| 1789 |
+
"learning_rate": 2.8967999965616815e-06,
|
| 1790 |
+
"loss": 0.0222,
|
| 1791 |
+
"num_input_tokens_seen": 338400,
|
| 1792 |
+
"step": 1080
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 13.5625,
|
| 1796 |
+
"grad_norm": 15.377069473266602,
|
| 1797 |
+
"learning_rate": 2.8474445159585235e-06,
|
| 1798 |
+
"loss": 0.0272,
|
| 1799 |
+
"num_input_tokens_seen": 339968,
|
| 1800 |
+
"step": 1085
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 13.625,
|
| 1804 |
+
"grad_norm": 7.599987983703613,
|
| 1805 |
+
"learning_rate": 2.798345167931771e-06,
|
| 1806 |
+
"loss": 0.0055,
|
| 1807 |
+
"num_input_tokens_seen": 341568,
|
| 1808 |
+
"step": 1090
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 13.6875,
|
| 1812 |
+
"grad_norm": 2.0365688800811768,
|
| 1813 |
+
"learning_rate": 2.7495077948128245e-06,
|
| 1814 |
+
"loss": 0.0058,
|
| 1815 |
+
"num_input_tokens_seen": 343168,
|
| 1816 |
+
"step": 1095
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 13.75,
|
| 1820 |
+
"grad_norm": 40.94605255126953,
|
| 1821 |
+
"learning_rate": 2.700938207760701e-06,
|
| 1822 |
+
"loss": 0.0186,
|
| 1823 |
+
"num_input_tokens_seen": 344704,
|
| 1824 |
+
"step": 1100
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 13.8125,
|
| 1828 |
+
"grad_norm": 0.5802567601203918,
|
| 1829 |
+
"learning_rate": 2.6526421860705474e-06,
|
| 1830 |
+
"loss": 0.002,
|
| 1831 |
+
"num_input_tokens_seen": 346272,
|
| 1832 |
+
"step": 1105
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 13.875,
|
| 1836 |
+
"grad_norm": 2.4557807445526123,
|
| 1837 |
+
"learning_rate": 2.6046254764859687e-06,
|
| 1838 |
+
"loss": 0.0125,
|
| 1839 |
+
"num_input_tokens_seen": 347808,
|
| 1840 |
+
"step": 1110
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 13.9375,
|
| 1844 |
+
"grad_norm": 103.32002258300781,
|
| 1845 |
+
"learning_rate": 2.5568937925152272e-06,
|
| 1846 |
+
"loss": 0.05,
|
| 1847 |
+
"num_input_tokens_seen": 349376,
|
| 1848 |
+
"step": 1115
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 14.0,
|
| 1852 |
+
"grad_norm": 0.0683021992444992,
|
| 1853 |
+
"learning_rate": 2.5094528137513797e-06,
|
| 1854 |
+
"loss": 0.0174,
|
| 1855 |
+
"num_input_tokens_seen": 350976,
|
| 1856 |
+
"step": 1120
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 14.0,
|
| 1860 |
+
"eval_loss": 0.8447664380073547,
|
| 1861 |
+
"eval_runtime": 0.9205,
|
| 1862 |
+
"eval_samples_per_second": 86.907,
|
| 1863 |
+
"eval_steps_per_second": 21.727,
|
| 1864 |
+
"num_input_tokens_seen": 350976,
|
| 1865 |
+
"step": 1120
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 14.0625,
|
| 1869 |
+
"grad_norm": 0.2838919758796692,
|
| 1870 |
+
"learning_rate": 2.462308185196481e-06,
|
| 1871 |
+
"loss": 0.0008,
|
| 1872 |
+
"num_input_tokens_seen": 352512,
|
| 1873 |
+
"step": 1125
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 14.125,
|
| 1877 |
+
"grad_norm": 0.25671952962875366,
|
| 1878 |
+
"learning_rate": 2.4154655165898626e-06,
|
| 1879 |
+
"loss": 0.0009,
|
| 1880 |
+
"num_input_tokens_seen": 354048,
|
| 1881 |
+
"step": 1130
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 14.1875,
|
| 1885 |
+
"grad_norm": 0.08236575126647949,
|
| 1886 |
+
"learning_rate": 2.3689303817406523e-06,
|
| 1887 |
+
"loss": 0.0019,
|
| 1888 |
+
"num_input_tokens_seen": 355584,
|
| 1889 |
+
"step": 1135
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 14.25,
|
| 1893 |
+
"grad_norm": 0.2644610106945038,
|
| 1894 |
+
"learning_rate": 2.3227083178645316e-06,
|
| 1895 |
+
"loss": 0.0053,
|
| 1896 |
+
"num_input_tokens_seen": 357152,
|
| 1897 |
+
"step": 1140
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 14.3125,
|
| 1901 |
+
"grad_norm": 0.06165740266442299,
|
| 1902 |
+
"learning_rate": 2.2768048249248648e-06,
|
| 1903 |
+
"loss": 0.0109,
|
| 1904 |
+
"num_input_tokens_seen": 358752,
|
| 1905 |
+
"step": 1145
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 14.375,
|
| 1909 |
+
"grad_norm": 0.31678006052970886,
|
| 1910 |
+
"learning_rate": 2.2312253649782655e-06,
|
| 1911 |
+
"loss": 0.0006,
|
| 1912 |
+
"num_input_tokens_seen": 360288,
|
| 1913 |
+
"step": 1150
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 14.4375,
|
| 1917 |
+
"grad_norm": 0.9465001225471497,
|
| 1918 |
+
"learning_rate": 2.185975361524657e-06,
|
| 1919 |
+
"loss": 0.0722,
|
| 1920 |
+
"num_input_tokens_seen": 361856,
|
| 1921 |
+
"step": 1155
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 14.5,
|
| 1925 |
+
"grad_norm": 2.73948335647583,
|
| 1926 |
+
"learning_rate": 2.1410601988619394e-06,
|
| 1927 |
+
"loss": 0.0038,
|
| 1928 |
+
"num_input_tokens_seen": 363392,
|
| 1929 |
+
"step": 1160
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 14.5625,
|
| 1933 |
+
"grad_norm": 0.18431881070137024,
|
| 1934 |
+
"learning_rate": 2.096485221445301e-06,
|
| 1935 |
+
"loss": 0.0063,
|
| 1936 |
+
"num_input_tokens_seen": 364960,
|
| 1937 |
+
"step": 1165
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 14.625,
|
| 1941 |
+
"grad_norm": 6.978606700897217,
|
| 1942 |
+
"learning_rate": 2.0522557332512953e-06,
|
| 1943 |
+
"loss": 0.0038,
|
| 1944 |
+
"num_input_tokens_seen": 366528,
|
| 1945 |
+
"step": 1170
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 14.6875,
|
| 1949 |
+
"grad_norm": 0.9169700145721436,
|
| 1950 |
+
"learning_rate": 2.008376997146705e-06,
|
| 1951 |
+
"loss": 0.0268,
|
| 1952 |
+
"num_input_tokens_seen": 368032,
|
| 1953 |
+
"step": 1175
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 14.75,
|
| 1957 |
+
"grad_norm": 0.3377976417541504,
|
| 1958 |
+
"learning_rate": 1.9648542342623276e-06,
|
| 1959 |
+
"loss": 0.0022,
|
| 1960 |
+
"num_input_tokens_seen": 369632,
|
| 1961 |
+
"step": 1180
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 14.8125,
|
| 1965 |
+
"grad_norm": 0.23450040817260742,
|
| 1966 |
+
"learning_rate": 1.9216926233717087e-06,
|
| 1967 |
+
"loss": 0.0003,
|
| 1968 |
+
"num_input_tokens_seen": 371232,
|
| 1969 |
+
"step": 1185
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 14.875,
|
| 1973 |
+
"grad_norm": 17.118005752563477,
|
| 1974 |
+
"learning_rate": 1.8788973002749112e-06,
|
| 1975 |
+
"loss": 0.0058,
|
| 1976 |
+
"num_input_tokens_seen": 372800,
|
| 1977 |
+
"step": 1190
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 14.9375,
|
| 1981 |
+
"grad_norm": 23.535619735717773,
|
| 1982 |
+
"learning_rate": 1.83647335718742e-06,
|
| 1983 |
+
"loss": 0.008,
|
| 1984 |
+
"num_input_tokens_seen": 374368,
|
| 1985 |
+
"step": 1195
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 15.0,
|
| 1989 |
+
"grad_norm": 0.2620501220226288,
|
| 1990 |
+
"learning_rate": 1.7944258421342097e-06,
|
| 1991 |
+
"loss": 0.0006,
|
| 1992 |
+
"num_input_tokens_seen": 376000,
|
| 1993 |
+
"step": 1200
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 15.0625,
|
| 1997 |
+
"grad_norm": 0.06305874139070511,
|
| 1998 |
+
"learning_rate": 1.7527597583490825e-06,
|
| 1999 |
+
"loss": 0.0076,
|
| 2000 |
+
"num_input_tokens_seen": 377568,
|
| 2001 |
+
"step": 1205
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 15.125,
|
| 2005 |
+
"grad_norm": 0.17243485152721405,
|
| 2006 |
+
"learning_rate": 1.7114800636793378e-06,
|
| 2007 |
+
"loss": 0.0003,
|
| 2008 |
+
"num_input_tokens_seen": 379072,
|
| 2009 |
+
"step": 1210
|
| 2010 |
+
},
|
| 2011 |
+
{
|
| 2012 |
+
"epoch": 15.1875,
|
| 2013 |
+
"grad_norm": 109.47997283935547,
|
| 2014 |
+
"learning_rate": 1.6705916699958292e-06,
|
| 2015 |
+
"loss": 0.0422,
|
| 2016 |
+
"num_input_tokens_seen": 380640,
|
| 2017 |
+
"step": 1215
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 15.25,
|
| 2021 |
+
"grad_norm": 13.874452590942383,
|
| 2022 |
+
"learning_rate": 1.6300994426085103e-06,
|
| 2023 |
+
"loss": 0.0015,
|
| 2024 |
+
"num_input_tokens_seen": 382240,
|
| 2025 |
+
"step": 1220
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 15.3125,
|
| 2029 |
+
"grad_norm": 0.06081826239824295,
|
| 2030 |
+
"learning_rate": 1.5900081996875083e-06,
|
| 2031 |
+
"loss": 0.0003,
|
| 2032 |
+
"num_input_tokens_seen": 383872,
|
| 2033 |
+
"step": 1225
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 15.375,
|
| 2037 |
+
"grad_norm": 0.1609436422586441,
|
| 2038 |
+
"learning_rate": 1.5503227116898017e-06,
|
| 2039 |
+
"loss": 0.0004,
|
| 2040 |
+
"num_input_tokens_seen": 385472,
|
| 2041 |
+
"step": 1230
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 15.4375,
|
| 2045 |
+
"grad_norm": 4.266076564788818,
|
| 2046 |
+
"learning_rate": 1.5110477007916002e-06,
|
| 2047 |
+
"loss": 0.0012,
|
| 2048 |
+
"num_input_tokens_seen": 387072,
|
| 2049 |
+
"step": 1235
|
| 2050 |
+
},
|
| 2051 |
+
{
|
| 2052 |
+
"epoch": 15.5,
|
| 2053 |
+
"grad_norm": 0.06309761852025986,
|
| 2054 |
+
"learning_rate": 1.4721878403264344e-06,
|
| 2055 |
+
"loss": 0.0006,
|
| 2056 |
+
"num_input_tokens_seen": 388640,
|
| 2057 |
+
"step": 1240
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 15.5625,
|
| 2061 |
+
"grad_norm": 0.2793193459510803,
|
| 2062 |
+
"learning_rate": 1.433747754229093e-06,
|
| 2063 |
+
"loss": 0.0009,
|
| 2064 |
+
"num_input_tokens_seen": 390240,
|
| 2065 |
+
"step": 1245
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 15.625,
|
| 2069 |
+
"grad_norm": 0.028820207342505455,
|
| 2070 |
+
"learning_rate": 1.395732016485406e-06,
|
| 2071 |
+
"loss": 0.0006,
|
| 2072 |
+
"num_input_tokens_seen": 391808,
|
| 2073 |
+
"step": 1250
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 15.6875,
|
| 2077 |
+
"grad_norm": 0.11917386204004288,
|
| 2078 |
+
"learning_rate": 1.3581451505879995e-06,
|
| 2079 |
+
"loss": 0.0002,
|
| 2080 |
+
"num_input_tokens_seen": 393408,
|
| 2081 |
+
"step": 1255
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 15.75,
|
| 2085 |
+
"grad_norm": 0.14404775202274323,
|
| 2086 |
+
"learning_rate": 1.3209916289980336e-06,
|
| 2087 |
+
"loss": 0.0004,
|
| 2088 |
+
"num_input_tokens_seen": 394976,
|
| 2089 |
+
"step": 1260
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 15.8125,
|
| 2093 |
+
"grad_norm": 0.47015178203582764,
|
| 2094 |
+
"learning_rate": 1.2842758726130283e-06,
|
| 2095 |
+
"loss": 0.0003,
|
| 2096 |
+
"num_input_tokens_seen": 396448,
|
| 2097 |
+
"step": 1265
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 15.875,
|
| 2101 |
+
"grad_norm": 37.57123565673828,
|
| 2102 |
+
"learning_rate": 1.2480022502408306e-06,
|
| 2103 |
+
"loss": 0.0103,
|
| 2104 |
+
"num_input_tokens_seen": 398048,
|
| 2105 |
+
"step": 1270
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 15.9375,
|
| 2109 |
+
"grad_norm": 0.09548679739236832,
|
| 2110 |
+
"learning_rate": 1.2121750780797514e-06,
|
| 2111 |
+
"loss": 0.0006,
|
| 2112 |
+
"num_input_tokens_seen": 399648,
|
| 2113 |
+
"step": 1275
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 16.0,
|
| 2117 |
+
"grad_norm": 0.09692618250846863,
|
| 2118 |
+
"learning_rate": 1.1767986192049986e-06,
|
| 2119 |
+
"loss": 0.0003,
|
| 2120 |
+
"num_input_tokens_seen": 401184,
|
| 2121 |
+
"step": 1280
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 16.0,
|
| 2125 |
+
"eval_loss": 0.9965259432792664,
|
| 2126 |
+
"eval_runtime": 0.9197,
|
| 2127 |
+
"eval_samples_per_second": 86.981,
|
| 2128 |
+
"eval_steps_per_second": 21.745,
|
| 2129 |
+
"num_input_tokens_seen": 401184,
|
| 2130 |
+
"step": 1280
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 16.0625,
|
| 2134 |
+
"grad_norm": 0.06563537567853928,
|
| 2135 |
+
"learning_rate": 1.1418770830614012e-06,
|
| 2136 |
+
"loss": 0.0002,
|
| 2137 |
+
"num_input_tokens_seen": 402816,
|
| 2138 |
+
"step": 1285
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 16.125,
|
| 2142 |
+
"grad_norm": 25.29011344909668,
|
| 2143 |
+
"learning_rate": 1.1074146249625334e-06,
|
| 2144 |
+
"loss": 0.0035,
|
| 2145 |
+
"num_input_tokens_seen": 404384,
|
| 2146 |
+
"step": 1290
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 16.1875,
|
| 2150 |
+
"grad_norm": 0.15210725367069244,
|
| 2151 |
+
"learning_rate": 1.0734153455962765e-06,
|
| 2152 |
+
"loss": 0.0002,
|
| 2153 |
+
"num_input_tokens_seen": 405984,
|
| 2154 |
+
"step": 1295
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 16.25,
|
| 2158 |
+
"grad_norm": 1.038436770439148,
|
| 2159 |
+
"learning_rate": 1.0398832905368693e-06,
|
| 2160 |
+
"loss": 0.0003,
|
| 2161 |
+
"num_input_tokens_seen": 407456,
|
| 2162 |
+
"step": 1300
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 16.3125,
|
| 2166 |
+
"grad_norm": 0.04672938957810402,
|
| 2167 |
+
"learning_rate": 1.006822449763537e-06,
|
| 2168 |
+
"loss": 0.0002,
|
| 2169 |
+
"num_input_tokens_seen": 409056,
|
| 2170 |
+
"step": 1305
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 16.375,
|
| 2174 |
+
"grad_norm": 0.10642636567354202,
|
| 2175 |
+
"learning_rate": 9.742367571857092e-07,
|
| 2176 |
+
"loss": 0.0002,
|
| 2177 |
+
"num_input_tokens_seen": 410624,
|
| 2178 |
+
"step": 1310
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 16.4375,
|
| 2182 |
+
"grad_norm": 0.012557949870824814,
|
| 2183 |
+
"learning_rate": 9.421300901749386e-07,
|
| 2184 |
+
"loss": 0.0001,
|
| 2185 |
+
"num_input_tokens_seen": 412192,
|
| 2186 |
+
"step": 1315
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 16.5,
|
| 2190 |
+
"grad_norm": 0.02506287954747677,
|
| 2191 |
+
"learning_rate": 9.105062691035233e-07,
|
| 2192 |
+
"loss": 0.0002,
|
| 2193 |
+
"num_input_tokens_seen": 413728,
|
| 2194 |
+
"step": 1320
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 16.5625,
|
| 2198 |
+
"grad_norm": 0.19741126894950867,
|
| 2199 |
+
"learning_rate": 8.793690568899216e-07,
|
| 2200 |
+
"loss": 0.0002,
|
| 2201 |
+
"num_input_tokens_seen": 415264,
|
| 2202 |
+
"step": 1325
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 16.625,
|
| 2206 |
+
"grad_norm": 0.3180708587169647,
|
| 2207 |
+
"learning_rate": 8.487221585510075e-07,
|
| 2208 |
+
"loss": 0.0002,
|
| 2209 |
+
"num_input_tokens_seen": 416864,
|
| 2210 |
+
"step": 1330
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 16.6875,
|
| 2214 |
+
"grad_norm": 0.2431306093931198,
|
| 2215 |
+
"learning_rate": 8.185692207612023e-07,
|
| 2216 |
+
"loss": 0.0003,
|
| 2217 |
+
"num_input_tokens_seen": 418464,
|
| 2218 |
+
"step": 1335
|
| 2219 |
+
},
|
| 2220 |
+
{
|
| 2221 |
+
"epoch": 16.75,
|
| 2222 |
+
"grad_norm": 0.07357773184776306,
|
| 2223 |
+
"learning_rate": 7.88913831418568e-07,
|
| 2224 |
+
"loss": 0.0008,
|
| 2225 |
+
"num_input_tokens_seen": 420032,
|
| 2226 |
+
"step": 1340
|
| 2227 |
+
},
|
| 2228 |
+
{
|
| 2229 |
+
"epoch": 16.8125,
|
| 2230 |
+
"grad_norm": 0.5140334963798523,
|
| 2231 |
+
"learning_rate": 7.597595192178702e-07,
|
| 2232 |
+
"loss": 0.0002,
|
| 2233 |
+
"num_input_tokens_seen": 421568,
|
| 2234 |
+
"step": 1345
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 16.875,
|
| 2238 |
+
"grad_norm": 0.10227775573730469,
|
| 2239 |
+
"learning_rate": 7.311097532307121e-07,
|
| 2240 |
+
"loss": 0.0003,
|
| 2241 |
+
"num_input_tokens_seen": 423136,
|
| 2242 |
+
"step": 1350
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 16.9375,
|
| 2246 |
+
"grad_norm": 0.04345543682575226,
|
| 2247 |
+
"learning_rate": 7.029679424927366e-07,
|
| 2248 |
+
"loss": 0.0009,
|
| 2249 |
+
"num_input_tokens_seen": 424640,
|
| 2250 |
+
"step": 1355
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 17.0,
|
| 2254 |
+
"grad_norm": 0.2489359974861145,
|
| 2255 |
+
"learning_rate": 6.753374355979975e-07,
|
| 2256 |
+
"loss": 0.0002,
|
| 2257 |
+
"num_input_tokens_seen": 426208,
|
| 2258 |
+
"step": 1360
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 17.0625,
|
| 2262 |
+
"grad_norm": 0.05077870190143585,
|
| 2263 |
+
"learning_rate": 6.482215203005016e-07,
|
| 2264 |
+
"loss": 0.0002,
|
| 2265 |
+
"num_input_tokens_seen": 427776,
|
| 2266 |
+
"step": 1365
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 17.125,
|
| 2270 |
+
"grad_norm": 0.051710885018110275,
|
| 2271 |
+
"learning_rate": 6.216234231230012e-07,
|
| 2272 |
+
"loss": 0.0002,
|
| 2273 |
+
"num_input_tokens_seen": 429312,
|
| 2274 |
+
"step": 1370
|
| 2275 |
+
},
|
| 2276 |
+
{
|
| 2277 |
+
"epoch": 17.1875,
|
| 2278 |
+
"grad_norm": 0.08844368159770966,
|
| 2279 |
+
"learning_rate": 5.955463089730723e-07,
|
| 2280 |
+
"loss": 0.0001,
|
| 2281 |
+
"num_input_tokens_seen": 430880,
|
| 2282 |
+
"step": 1375
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 17.25,
|
| 2286 |
+
"grad_norm": 0.06657011061906815,
|
| 2287 |
+
"learning_rate": 5.699932807665198e-07,
|
| 2288 |
+
"loss": 0.0002,
|
| 2289 |
+
"num_input_tokens_seen": 432448,
|
| 2290 |
+
"step": 1380
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 17.3125,
|
| 2294 |
+
"grad_norm": 0.018467910587787628,
|
| 2295 |
+
"learning_rate": 5.449673790581611e-07,
|
| 2296 |
+
"loss": 0.0002,
|
| 2297 |
+
"num_input_tokens_seen": 434016,
|
| 2298 |
+
"step": 1385
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 17.375,
|
| 2302 |
+
"grad_norm": 0.0749938040971756,
|
| 2303 |
+
"learning_rate": 5.204715816800343e-07,
|
| 2304 |
+
"loss": 0.0017,
|
| 2305 |
+
"num_input_tokens_seen": 435520,
|
| 2306 |
+
"step": 1390
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 17.4375,
|
| 2310 |
+
"grad_norm": 0.09172087907791138,
|
| 2311 |
+
"learning_rate": 4.965088033870608e-07,
|
| 2312 |
+
"loss": 0.0002,
|
| 2313 |
+
"num_input_tokens_seen": 437056,
|
| 2314 |
+
"step": 1395
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 17.5,
|
| 2318 |
+
"grad_norm": 0.05389586091041565,
|
| 2319 |
+
"learning_rate": 4.730818955102234e-07,
|
| 2320 |
+
"loss": 0.0002,
|
| 2321 |
+
"num_input_tokens_seen": 438624,
|
| 2322 |
+
"step": 1400
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 17.5625,
|
| 2326 |
+
"grad_norm": 0.03451506793498993,
|
| 2327 |
+
"learning_rate": 4.501936456172845e-07,
|
| 2328 |
+
"loss": 0.0002,
|
| 2329 |
+
"num_input_tokens_seen": 440192,
|
| 2330 |
+
"step": 1405
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 17.625,
|
| 2334 |
+
"grad_norm": 0.05420379713177681,
|
| 2335 |
+
"learning_rate": 4.278467771810896e-07,
|
| 2336 |
+
"loss": 0.0002,
|
| 2337 |
+
"num_input_tokens_seen": 441760,
|
| 2338 |
+
"step": 1410
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 17.6875,
|
| 2342 |
+
"grad_norm": 0.013087287545204163,
|
| 2343 |
+
"learning_rate": 4.0604394925550906e-07,
|
| 2344 |
+
"loss": 0.0001,
|
| 2345 |
+
"num_input_tokens_seen": 443424,
|
| 2346 |
+
"step": 1415
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 17.75,
|
| 2350 |
+
"grad_norm": 0.0480063296854496,
|
| 2351 |
+
"learning_rate": 3.8478775615902965e-07,
|
| 2352 |
+
"loss": 0.0007,
|
| 2353 |
+
"num_input_tokens_seen": 444992,
|
| 2354 |
+
"step": 1420
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 17.8125,
|
| 2358 |
+
"grad_norm": 0.038323502987623215,
|
| 2359 |
+
"learning_rate": 3.6408072716606346e-07,
|
| 2360 |
+
"loss": 0.0002,
|
| 2361 |
+
"num_input_tokens_seen": 446560,
|
| 2362 |
+
"step": 1425
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 17.875,
|
| 2366 |
+
"grad_norm": 0.22902554273605347,
|
| 2367 |
+
"learning_rate": 3.439253262059822e-07,
|
| 2368 |
+
"loss": 0.0002,
|
| 2369 |
+
"num_input_tokens_seen": 448192,
|
| 2370 |
+
"step": 1430
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 17.9375,
|
| 2374 |
+
"grad_norm": 0.026656942442059517,
|
| 2375 |
+
"learning_rate": 3.24323951569942e-07,
|
| 2376 |
+
"loss": 0.0002,
|
| 2377 |
+
"num_input_tokens_seen": 449792,
|
| 2378 |
+
"step": 1435
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 18.0,
|
| 2382 |
+
"grad_norm": 0.025560805574059486,
|
| 2383 |
+
"learning_rate": 3.052789356255037e-07,
|
| 2384 |
+
"loss": 0.0002,
|
| 2385 |
+
"num_input_tokens_seen": 451328,
|
| 2386 |
+
"step": 1440
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 18.0,
|
| 2390 |
+
"eval_loss": 1.080618143081665,
|
| 2391 |
+
"eval_runtime": 0.9186,
|
| 2392 |
+
"eval_samples_per_second": 87.091,
|
| 2393 |
+
"eval_steps_per_second": 21.773,
|
| 2394 |
+
"num_input_tokens_seen": 451328,
|
| 2395 |
+
"step": 1440
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 18.0625,
|
| 2399 |
+
"grad_norm": 0.03586863726377487,
|
| 2400 |
+
"learning_rate": 2.867925445391079e-07,
|
| 2401 |
+
"loss": 0.0004,
|
| 2402 |
+
"num_input_tokens_seen": 452896,
|
| 2403 |
+
"step": 1445
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 18.125,
|
| 2407 |
+
"grad_norm": 0.012726732529699802,
|
| 2408 |
+
"learning_rate": 2.688669780064268e-07,
|
| 2409 |
+
"loss": 0.0002,
|
| 2410 |
+
"num_input_tokens_seen": 454368,
|
| 2411 |
+
"step": 1450
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 18.1875,
|
| 2415 |
+
"grad_norm": 0.16940614581108093,
|
| 2416 |
+
"learning_rate": 2.5150436899061494e-07,
|
| 2417 |
+
"loss": 0.0002,
|
| 2418 |
+
"num_input_tokens_seen": 455936,
|
| 2419 |
+
"step": 1455
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 18.25,
|
| 2423 |
+
"grad_norm": 0.10751322656869888,
|
| 2424 |
+
"learning_rate": 2.3470678346851517e-07,
|
| 2425 |
+
"loss": 0.0002,
|
| 2426 |
+
"num_input_tokens_seen": 457536,
|
| 2427 |
+
"step": 1460
|
| 2428 |
+
},
|
| 2429 |
+
{
|
| 2430 |
+
"epoch": 18.3125,
|
| 2431 |
+
"grad_norm": 0.01832233928143978,
|
| 2432 |
+
"learning_rate": 2.1847622018482283e-07,
|
| 2433 |
+
"loss": 0.0001,
|
| 2434 |
+
"num_input_tokens_seen": 459072,
|
| 2435 |
+
"step": 1465
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 18.375,
|
| 2439 |
+
"grad_norm": 0.1913994550704956,
|
| 2440 |
+
"learning_rate": 2.028146104142581e-07,
|
| 2441 |
+
"loss": 0.0001,
|
| 2442 |
+
"num_input_tokens_seen": 460640,
|
| 2443 |
+
"step": 1470
|
| 2444 |
+
},
|
| 2445 |
+
{
|
| 2446 |
+
"epoch": 18.4375,
|
| 2447 |
+
"grad_norm": 0.17769768834114075,
|
| 2448 |
+
"learning_rate": 1.8772381773176417e-07,
|
| 2449 |
+
"loss": 0.0002,
|
| 2450 |
+
"num_input_tokens_seen": 462208,
|
| 2451 |
+
"step": 1475
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 18.5,
|
| 2455 |
+
"grad_norm": 0.1154564693570137,
|
| 2456 |
+
"learning_rate": 1.7320563779075595e-07,
|
| 2457 |
+
"loss": 0.0001,
|
| 2458 |
+
"num_input_tokens_seen": 463744,
|
| 2459 |
+
"step": 1480
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 18.5625,
|
| 2463 |
+
"grad_norm": 0.03802482411265373,
|
| 2464 |
+
"learning_rate": 1.5926179810946185e-07,
|
| 2465 |
+
"loss": 0.0002,
|
| 2466 |
+
"num_input_tokens_seen": 465344,
|
| 2467 |
+
"step": 1485
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 18.625,
|
| 2471 |
+
"grad_norm": 0.16934075951576233,
|
| 2472 |
+
"learning_rate": 1.4589395786535954e-07,
|
| 2473 |
+
"loss": 0.0001,
|
| 2474 |
+
"num_input_tokens_seen": 466880,
|
| 2475 |
+
"step": 1490
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 18.6875,
|
| 2479 |
+
"grad_norm": 0.045956190675497055,
|
| 2480 |
+
"learning_rate": 1.331037076977576e-07,
|
| 2481 |
+
"loss": 0.0002,
|
| 2482 |
+
"num_input_tokens_seen": 468448,
|
| 2483 |
+
"step": 1495
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 18.75,
|
| 2487 |
+
"grad_norm": 0.04867981746792793,
|
| 2488 |
+
"learning_rate": 1.2089256951851923e-07,
|
| 2489 |
+
"loss": 0.0002,
|
| 2490 |
+
"num_input_tokens_seen": 470048,
|
| 2491 |
+
"step": 1500
|
| 2492 |
+
},
|
| 2493 |
+
{
|
| 2494 |
+
"epoch": 18.8125,
|
| 2495 |
+
"grad_norm": 0.04806559160351753,
|
| 2496 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 2497 |
+
"loss": 0.0001,
|
| 2498 |
+
"num_input_tokens_seen": 471584,
|
| 2499 |
+
"step": 1505
|
| 2500 |
+
},
|
| 2501 |
+
{
|
| 2502 |
+
"epoch": 18.875,
|
| 2503 |
+
"grad_norm": 0.04029763862490654,
|
| 2504 |
+
"learning_rate": 9.821337205701664e-08,
|
| 2505 |
+
"loss": 0.0002,
|
| 2506 |
+
"num_input_tokens_seen": 473152,
|
| 2507 |
+
"step": 1510
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 18.9375,
|
| 2511 |
+
"grad_norm": 0.07741861045360565,
|
| 2512 |
+
"learning_rate": 8.77480113724516e-08,
|
| 2513 |
+
"loss": 0.0004,
|
| 2514 |
+
"num_input_tokens_seen": 474720,
|
| 2515 |
+
"step": 1515
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 19.0,
|
| 2519 |
+
"grad_norm": 0.011278778314590454,
|
| 2520 |
+
"learning_rate": 7.786715955054202e-08,
|
| 2521 |
+
"loss": 0.0001,
|
| 2522 |
+
"num_input_tokens_seen": 476320,
|
| 2523 |
+
"step": 1520
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 19.0625,
|
| 2527 |
+
"grad_norm": 0.34853649139404297,
|
| 2528 |
+
"learning_rate": 6.857199231384282e-08,
|
| 2529 |
+
"loss": 0.0002,
|
| 2530 |
+
"num_input_tokens_seen": 477824,
|
| 2531 |
+
"step": 1525
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 19.125,
|
| 2535 |
+
"grad_norm": 0.017191864550113678,
|
| 2536 |
+
"learning_rate": 5.986361569430166e-08,
|
| 2537 |
+
"loss": 0.0002,
|
| 2538 |
+
"num_input_tokens_seen": 479360,
|
| 2539 |
+
"step": 1530
|
| 2540 |
+
},
|
| 2541 |
+
{
|
| 2542 |
+
"epoch": 19.1875,
|
| 2543 |
+
"grad_norm": 0.08832144737243652,
|
| 2544 |
+
"learning_rate": 5.174306590164879e-08,
|
| 2545 |
+
"loss": 0.0003,
|
| 2546 |
+
"num_input_tokens_seen": 480896,
|
| 2547 |
+
"step": 1535
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 19.25,
|
| 2551 |
+
"grad_norm": 0.17200443148612976,
|
| 2552 |
+
"learning_rate": 4.42113092001023e-08,
|
| 2553 |
+
"loss": 0.0002,
|
| 2554 |
+
"num_input_tokens_seen": 482464,
|
| 2555 |
+
"step": 1540
|
| 2556 |
+
},
|
| 2557 |
+
{
|
| 2558 |
+
"epoch": 19.3125,
|
| 2559 |
+
"grad_norm": 0.1417255848646164,
|
| 2560 |
+
"learning_rate": 3.726924179339009e-08,
|
| 2561 |
+
"loss": 0.0001,
|
| 2562 |
+
"num_input_tokens_seen": 484032,
|
| 2563 |
+
"step": 1545
|
| 2564 |
+
},
|
| 2565 |
+
{
|
| 2566 |
+
"epoch": 19.375,
|
| 2567 |
+
"grad_norm": 0.040031641721725464,
|
| 2568 |
+
"learning_rate": 3.09176897181096e-08,
|
| 2569 |
+
"loss": 0.0001,
|
| 2570 |
+
"num_input_tokens_seen": 485632,
|
| 2571 |
+
"step": 1550
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 19.4375,
|
| 2575 |
+
"grad_norm": 0.024730442091822624,
|
| 2576 |
+
"learning_rate": 2.515740874544148e-08,
|
| 2577 |
+
"loss": 0.0002,
|
| 2578 |
+
"num_input_tokens_seen": 487232,
|
| 2579 |
+
"step": 1555
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 19.5,
|
| 2583 |
+
"grad_norm": 0.14058633148670197,
|
| 2584 |
+
"learning_rate": 1.9989084291216487e-08,
|
| 2585 |
+
"loss": 0.0002,
|
| 2586 |
+
"num_input_tokens_seen": 488832,
|
| 2587 |
+
"step": 1560
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 19.5625,
|
| 2591 |
+
"grad_norm": 0.15317003428936005,
|
| 2592 |
+
"learning_rate": 1.541333133436018e-08,
|
| 2593 |
+
"loss": 0.0001,
|
| 2594 |
+
"num_input_tokens_seen": 490400,
|
| 2595 |
+
"step": 1565
|
| 2596 |
+
},
|
| 2597 |
+
{
|
| 2598 |
+
"epoch": 19.625,
|
| 2599 |
+
"grad_norm": 0.05648527294397354,
|
| 2600 |
+
"learning_rate": 1.1430694343715354e-08,
|
| 2601 |
+
"loss": 0.0001,
|
| 2602 |
+
"num_input_tokens_seen": 491936,
|
| 2603 |
+
"step": 1570
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 19.6875,
|
| 2607 |
+
"grad_norm": 0.10134085267782211,
|
| 2608 |
+
"learning_rate": 8.041647213256066e-09,
|
| 2609 |
+
"loss": 0.0002,
|
| 2610 |
+
"num_input_tokens_seen": 493504,
|
| 2611 |
+
"step": 1575
|
| 2612 |
+
},
|
| 2613 |
+
{
|
| 2614 |
+
"epoch": 19.75,
|
| 2615 |
+
"grad_norm": 0.022655094042420387,
|
| 2616 |
+
"learning_rate": 5.246593205699424e-09,
|
| 2617 |
+
"loss": 0.0001,
|
| 2618 |
+
"num_input_tokens_seen": 495104,
|
| 2619 |
+
"step": 1580
|
| 2620 |
+
},
|
| 2621 |
+
{
|
| 2622 |
+
"epoch": 19.8125,
|
| 2623 |
+
"grad_norm": 0.009513732977211475,
|
| 2624 |
+
"learning_rate": 3.0458649045211897e-09,
|
| 2625 |
+
"loss": 0.0002,
|
| 2626 |
+
"num_input_tokens_seen": 496672,
|
| 2627 |
+
"step": 1585
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 19.875,
|
| 2631 |
+
"grad_norm": 0.013120757415890694,
|
| 2632 |
+
"learning_rate": 1.4397241743813185e-09,
|
| 2633 |
+
"loss": 0.0001,
|
| 2634 |
+
"num_input_tokens_seen": 498240,
|
| 2635 |
+
"step": 1590
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 19.9375,
|
| 2639 |
+
"grad_norm": 0.06909758597612381,
|
| 2640 |
+
"learning_rate": 4.283621299649987e-10,
|
| 2641 |
+
"loss": 0.0002,
|
| 2642 |
+
"num_input_tokens_seen": 499808,
|
| 2643 |
+
"step": 1595
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 20.0,
|
| 2647 |
+
"grad_norm": 0.05095551162958145,
|
| 2648 |
+
"learning_rate": 1.189911324084303e-11,
|
| 2649 |
+
"loss": 0.0003,
|
| 2650 |
+
"num_input_tokens_seen": 501440,
|
| 2651 |
+
"step": 1600
|
| 2652 |
+
},
|
| 2653 |
+
{
|
| 2654 |
+
"epoch": 20.0,
|
| 2655 |
+
"eval_loss": 1.093362808227539,
|
| 2656 |
+
"eval_runtime": 0.919,
|
| 2657 |
+
"eval_samples_per_second": 87.054,
|
| 2658 |
+
"eval_steps_per_second": 21.763,
|
| 2659 |
+
"num_input_tokens_seen": 501440,
|
| 2660 |
+
"step": 1600
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 20.0,
|
| 2664 |
+
"num_input_tokens_seen": 501440,
|
| 2665 |
+
"step": 1600,
|
| 2666 |
+
"total_flos": 2.257961656516608e+16,
|
| 2667 |
+
"train_loss": 0.2722132059369324,
|
| 2668 |
+
"train_runtime": 168.4339,
|
| 2669 |
+
"train_samples_per_second": 37.997,
|
| 2670 |
+
"train_steps_per_second": 9.499
|
| 2671 |
+
}
|
| 2672 |
+
],
|
| 2673 |
+
"logging_steps": 5,
|
| 2674 |
+
"max_steps": 1600,
|
| 2675 |
+
"num_input_tokens_seen": 501440,
|
| 2676 |
+
"num_train_epochs": 20,
|
| 2677 |
+
"save_steps": 160,
|
| 2678 |
+
"stateful_callbacks": {
|
| 2679 |
+
"TrainerControl": {
|
| 2680 |
+
"args": {
|
| 2681 |
+
"should_epoch_stop": false,
|
| 2682 |
+
"should_evaluate": false,
|
| 2683 |
+
"should_log": false,
|
| 2684 |
+
"should_save": true,
|
| 2685 |
+
"should_training_stop": true
|
| 2686 |
+
},
|
| 2687 |
+
"attributes": {}
|
| 2688 |
+
}
|
| 2689 |
+
},
|
| 2690 |
+
"total_flos": 2.257961656516608e+16,
|
| 2691 |
+
"train_batch_size": 4,
|
| 2692 |
+
"trial_name": null,
|
| 2693 |
+
"trial_params": null
|
| 2694 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|