End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +9 -0
- trainer_state.json +2504 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
# train_cb_42_1760623596
|
| 19 |
|
| 20 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
-
- Loss: 0.
|
| 23 |
- Num Input Tokens Seen: 725992
|
| 24 |
|
| 25 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
# train_cb_42_1760623596
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the cb dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.1325
|
| 23 |
- Num Input Tokens Seen: 725992
|
| 24 |
|
| 25 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.13247625529766083,
|
| 4 |
+
"eval_runtime": 0.8406,
|
| 5 |
+
"eval_samples_per_second": 29.74,
|
| 6 |
+
"eval_steps_per_second": 8.327,
|
| 7 |
+
"num_input_tokens_seen": 725992,
|
| 8 |
+
"total_flos": 3.2691091634847744e+16,
|
| 9 |
+
"train_loss": 0.2356332610214227,
|
| 10 |
+
"train_runtime": 292.8644,
|
| 11 |
+
"train_samples_per_second": 15.365,
|
| 12 |
+
"train_steps_per_second": 3.893
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.13247625529766083,
|
| 4 |
+
"eval_runtime": 0.8406,
|
| 5 |
+
"eval_samples_per_second": 29.74,
|
| 6 |
+
"eval_steps_per_second": 8.327,
|
| 7 |
+
"num_input_tokens_seen": 725992
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"num_input_tokens_seen": 725992,
|
| 4 |
+
"total_flos": 3.2691091634847744e+16,
|
| 5 |
+
"train_loss": 0.2356332610214227,
|
| 6 |
+
"train_runtime": 292.8644,
|
| 7 |
+
"train_samples_per_second": 15.365,
|
| 8 |
+
"train_steps_per_second": 3.893
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2504 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 855,
|
| 3 |
+
"best_metric": 0.13247625529766083,
|
| 4 |
+
"best_model_checkpoint": "saves_multiple/prompt-tuning/llama-3-8b-instruct/train_cb_42_1760623596/checkpoint-855",
|
| 5 |
+
"epoch": 20.0,
|
| 6 |
+
"eval_steps": 57,
|
| 7 |
+
"global_step": 1140,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.08771929824561403,
|
| 14 |
+
"grad_norm": 5.84375,
|
| 15 |
+
"learning_rate": 0.0010526315789473684,
|
| 16 |
+
"loss": 0.7268,
|
| 17 |
+
"num_input_tokens_seen": 3232,
|
| 18 |
+
"step": 5,
|
| 19 |
+
"train_runtime": 2.9953,
|
| 20 |
+
"train_tokens_per_second": 1079.037
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.17543859649122806,
|
| 24 |
+
"grad_norm": 34.0,
|
| 25 |
+
"learning_rate": 0.0023684210526315787,
|
| 26 |
+
"loss": 0.2451,
|
| 27 |
+
"num_input_tokens_seen": 6432,
|
| 28 |
+
"step": 10,
|
| 29 |
+
"train_runtime": 4.132,
|
| 30 |
+
"train_tokens_per_second": 1556.624
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2631578947368421,
|
| 34 |
+
"grad_norm": 50.0,
|
| 35 |
+
"learning_rate": 0.003684210526315789,
|
| 36 |
+
"loss": 2.5574,
|
| 37 |
+
"num_input_tokens_seen": 9376,
|
| 38 |
+
"step": 15,
|
| 39 |
+
"train_runtime": 5.2153,
|
| 40 |
+
"train_tokens_per_second": 1797.799
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.3508771929824561,
|
| 44 |
+
"grad_norm": 73.5,
|
| 45 |
+
"learning_rate": 0.004999999999999999,
|
| 46 |
+
"loss": 1.6515,
|
| 47 |
+
"num_input_tokens_seen": 12448,
|
| 48 |
+
"step": 20,
|
| 49 |
+
"train_runtime": 6.3091,
|
| 50 |
+
"train_tokens_per_second": 1973.036
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.43859649122807015,
|
| 54 |
+
"grad_norm": 120.5,
|
| 55 |
+
"learning_rate": 0.0063157894736842095,
|
| 56 |
+
"loss": 1.7016,
|
| 57 |
+
"num_input_tokens_seen": 15456,
|
| 58 |
+
"step": 25,
|
| 59 |
+
"train_runtime": 7.4066,
|
| 60 |
+
"train_tokens_per_second": 2086.797
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.5263157894736842,
|
| 64 |
+
"grad_norm": 34.25,
|
| 65 |
+
"learning_rate": 0.0076315789473684215,
|
| 66 |
+
"loss": 5.4554,
|
| 67 |
+
"num_input_tokens_seen": 18592,
|
| 68 |
+
"step": 30,
|
| 69 |
+
"train_runtime": 8.5217,
|
| 70 |
+
"train_tokens_per_second": 2181.73
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.6140350877192983,
|
| 74 |
+
"grad_norm": 29.0,
|
| 75 |
+
"learning_rate": 0.008947368421052631,
|
| 76 |
+
"loss": 0.9526,
|
| 77 |
+
"num_input_tokens_seen": 23136,
|
| 78 |
+
"step": 35,
|
| 79 |
+
"train_runtime": 9.8883,
|
| 80 |
+
"train_tokens_per_second": 2339.737
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.7017543859649122,
|
| 84 |
+
"grad_norm": 32.25,
|
| 85 |
+
"learning_rate": 0.010263157894736842,
|
| 86 |
+
"loss": 2.461,
|
| 87 |
+
"num_input_tokens_seen": 26496,
|
| 88 |
+
"step": 40,
|
| 89 |
+
"train_runtime": 11.0339,
|
| 90 |
+
"train_tokens_per_second": 2401.324
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.7894736842105263,
|
| 94 |
+
"grad_norm": 23.375,
|
| 95 |
+
"learning_rate": 0.011578947368421052,
|
| 96 |
+
"loss": 2.1645,
|
| 97 |
+
"num_input_tokens_seen": 29728,
|
| 98 |
+
"step": 45,
|
| 99 |
+
"train_runtime": 12.1701,
|
| 100 |
+
"train_tokens_per_second": 2442.706
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8771929824561403,
|
| 104 |
+
"grad_norm": 8.1875,
|
| 105 |
+
"learning_rate": 0.012894736842105263,
|
| 106 |
+
"loss": 1.0525,
|
| 107 |
+
"num_input_tokens_seen": 33024,
|
| 108 |
+
"step": 50,
|
| 109 |
+
"train_runtime": 13.3281,
|
| 110 |
+
"train_tokens_per_second": 2477.775
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.9649122807017544,
|
| 114 |
+
"grad_norm": 1.8125,
|
| 115 |
+
"learning_rate": 0.014210526315789472,
|
| 116 |
+
"loss": 0.6057,
|
| 117 |
+
"num_input_tokens_seen": 35904,
|
| 118 |
+
"step": 55,
|
| 119 |
+
"train_runtime": 14.4008,
|
| 120 |
+
"train_tokens_per_second": 2493.189
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.0,
|
| 124 |
+
"eval_loss": 1.388181447982788,
|
| 125 |
+
"eval_runtime": 0.8101,
|
| 126 |
+
"eval_samples_per_second": 30.859,
|
| 127 |
+
"eval_steps_per_second": 8.64,
|
| 128 |
+
"num_input_tokens_seen": 36480,
|
| 129 |
+
"step": 57
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0526315789473684,
|
| 133 |
+
"grad_norm": 3.203125,
|
| 134 |
+
"learning_rate": 0.015526315789473685,
|
| 135 |
+
"loss": 0.6631,
|
| 136 |
+
"num_input_tokens_seen": 38016,
|
| 137 |
+
"step": 60,
|
| 138 |
+
"train_runtime": 17.0848,
|
| 139 |
+
"train_tokens_per_second": 2225.139
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 1.1403508771929824,
|
| 143 |
+
"grad_norm": 0.625,
|
| 144 |
+
"learning_rate": 0.016842105263157894,
|
| 145 |
+
"loss": 0.5022,
|
| 146 |
+
"num_input_tokens_seen": 40800,
|
| 147 |
+
"step": 65,
|
| 148 |
+
"train_runtime": 18.168,
|
| 149 |
+
"train_tokens_per_second": 2245.709
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.2280701754385965,
|
| 153 |
+
"grad_norm": 0.87890625,
|
| 154 |
+
"learning_rate": 0.018157894736842106,
|
| 155 |
+
"loss": 1.1335,
|
| 156 |
+
"num_input_tokens_seen": 43648,
|
| 157 |
+
"step": 70,
|
| 158 |
+
"train_runtime": 19.234,
|
| 159 |
+
"train_tokens_per_second": 2269.317
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.3157894736842106,
|
| 163 |
+
"grad_norm": 0.6796875,
|
| 164 |
+
"learning_rate": 0.019473684210526317,
|
| 165 |
+
"loss": 0.4496,
|
| 166 |
+
"num_input_tokens_seen": 46112,
|
| 167 |
+
"step": 75,
|
| 168 |
+
"train_runtime": 20.2339,
|
| 169 |
+
"train_tokens_per_second": 2278.953
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.4035087719298245,
|
| 173 |
+
"grad_norm": 0.302734375,
|
| 174 |
+
"learning_rate": 0.020789473684210528,
|
| 175 |
+
"loss": 0.4359,
|
| 176 |
+
"num_input_tokens_seen": 49632,
|
| 177 |
+
"step": 80,
|
| 178 |
+
"train_runtime": 21.4011,
|
| 179 |
+
"train_tokens_per_second": 2319.131
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.4912280701754386,
|
| 183 |
+
"grad_norm": 0.380859375,
|
| 184 |
+
"learning_rate": 0.022105263157894735,
|
| 185 |
+
"loss": 0.2075,
|
| 186 |
+
"num_input_tokens_seen": 53024,
|
| 187 |
+
"step": 85,
|
| 188 |
+
"train_runtime": 22.5602,
|
| 189 |
+
"train_tokens_per_second": 2350.331
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.5789473684210527,
|
| 193 |
+
"grad_norm": 0.6484375,
|
| 194 |
+
"learning_rate": 0.023421052631578947,
|
| 195 |
+
"loss": 0.3102,
|
| 196 |
+
"num_input_tokens_seen": 56672,
|
| 197 |
+
"step": 90,
|
| 198 |
+
"train_runtime": 23.7736,
|
| 199 |
+
"train_tokens_per_second": 2383.819
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.6666666666666665,
|
| 203 |
+
"grad_norm": 1.265625,
|
| 204 |
+
"learning_rate": 0.024736842105263158,
|
| 205 |
+
"loss": 0.3557,
|
| 206 |
+
"num_input_tokens_seen": 59296,
|
| 207 |
+
"step": 95,
|
| 208 |
+
"train_runtime": 24.8127,
|
| 209 |
+
"train_tokens_per_second": 2389.745
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.7543859649122808,
|
| 213 |
+
"grad_norm": 0.248046875,
|
| 214 |
+
"learning_rate": 0.02605263157894737,
|
| 215 |
+
"loss": 0.3327,
|
| 216 |
+
"num_input_tokens_seen": 62528,
|
| 217 |
+
"step": 100,
|
| 218 |
+
"train_runtime": 25.9465,
|
| 219 |
+
"train_tokens_per_second": 2409.879
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.8421052631578947,
|
| 223 |
+
"grad_norm": 0.10546875,
|
| 224 |
+
"learning_rate": 0.027368421052631577,
|
| 225 |
+
"loss": 0.2654,
|
| 226 |
+
"num_input_tokens_seen": 65856,
|
| 227 |
+
"step": 105,
|
| 228 |
+
"train_runtime": 27.0946,
|
| 229 |
+
"train_tokens_per_second": 2430.599
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.9298245614035088,
|
| 233 |
+
"grad_norm": 0.703125,
|
| 234 |
+
"learning_rate": 0.028684210526315788,
|
| 235 |
+
"loss": 0.3279,
|
| 236 |
+
"num_input_tokens_seen": 69664,
|
| 237 |
+
"step": 110,
|
| 238 |
+
"train_runtime": 28.3144,
|
| 239 |
+
"train_tokens_per_second": 2460.372
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.0,
|
| 243 |
+
"eval_loss": 0.32342329621315,
|
| 244 |
+
"eval_runtime": 0.8136,
|
| 245 |
+
"eval_samples_per_second": 30.729,
|
| 246 |
+
"eval_steps_per_second": 8.604,
|
| 247 |
+
"num_input_tokens_seen": 72112,
|
| 248 |
+
"step": 114
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.017543859649123,
|
| 252 |
+
"grad_norm": 0.09130859375,
|
| 253 |
+
"learning_rate": 0.03,
|
| 254 |
+
"loss": 0.2806,
|
| 255 |
+
"num_input_tokens_seen": 72848,
|
| 256 |
+
"step": 115,
|
| 257 |
+
"train_runtime": 31.212,
|
| 258 |
+
"train_tokens_per_second": 2333.976
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.1052631578947367,
|
| 262 |
+
"grad_norm": 0.0693359375,
|
| 263 |
+
"learning_rate": 0.02999824208523885,
|
| 264 |
+
"loss": 0.2312,
|
| 265 |
+
"num_input_tokens_seen": 75888,
|
| 266 |
+
"step": 120,
|
| 267 |
+
"train_runtime": 32.3237,
|
| 268 |
+
"train_tokens_per_second": 2347.752
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.192982456140351,
|
| 272 |
+
"grad_norm": 0.228515625,
|
| 273 |
+
"learning_rate": 0.029992968752990647,
|
| 274 |
+
"loss": 0.2075,
|
| 275 |
+
"num_input_tokens_seen": 78896,
|
| 276 |
+
"step": 125,
|
| 277 |
+
"train_runtime": 33.4253,
|
| 278 |
+
"train_tokens_per_second": 2360.368
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.280701754385965,
|
| 282 |
+
"grad_norm": 0.08544921875,
|
| 283 |
+
"learning_rate": 0.02998418123926453,
|
| 284 |
+
"loss": 0.2473,
|
| 285 |
+
"num_input_tokens_seen": 82288,
|
| 286 |
+
"step": 130,
|
| 287 |
+
"train_runtime": 34.5726,
|
| 288 |
+
"train_tokens_per_second": 2380.148
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.3684210526315788,
|
| 292 |
+
"grad_norm": 0.203125,
|
| 293 |
+
"learning_rate": 0.029971881603753848,
|
| 294 |
+
"loss": 0.2505,
|
| 295 |
+
"num_input_tokens_seen": 85936,
|
| 296 |
+
"step": 135,
|
| 297 |
+
"train_runtime": 35.7819,
|
| 298 |
+
"train_tokens_per_second": 2401.659
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.456140350877193,
|
| 302 |
+
"grad_norm": 0.205078125,
|
| 303 |
+
"learning_rate": 0.02995607272935338,
|
| 304 |
+
"loss": 0.4384,
|
| 305 |
+
"num_input_tokens_seen": 88784,
|
| 306 |
+
"step": 140,
|
| 307 |
+
"train_runtime": 36.8488,
|
| 308 |
+
"train_tokens_per_second": 2409.411
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.543859649122807,
|
| 312 |
+
"grad_norm": 0.10302734375,
|
| 313 |
+
"learning_rate": 0.02993675832148361,
|
| 314 |
+
"loss": 0.3659,
|
| 315 |
+
"num_input_tokens_seen": 91696,
|
| 316 |
+
"step": 145,
|
| 317 |
+
"train_runtime": 37.9323,
|
| 318 |
+
"train_tokens_per_second": 2417.361
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 2.6315789473684212,
|
| 322 |
+
"grad_norm": 0.484375,
|
| 323 |
+
"learning_rate": 0.029913942907222237,
|
| 324 |
+
"loss": 0.3397,
|
| 325 |
+
"num_input_tokens_seen": 95024,
|
| 326 |
+
"step": 150,
|
| 327 |
+
"train_runtime": 39.0718,
|
| 328 |
+
"train_tokens_per_second": 2432.033
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 2.719298245614035,
|
| 332 |
+
"grad_norm": 0.040283203125,
|
| 333 |
+
"learning_rate": 0.029887631834243058,
|
| 334 |
+
"loss": 0.2108,
|
| 335 |
+
"num_input_tokens_seen": 98512,
|
| 336 |
+
"step": 155,
|
| 337 |
+
"train_runtime": 40.2495,
|
| 338 |
+
"train_tokens_per_second": 2447.532
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.807017543859649,
|
| 342 |
+
"grad_norm": 0.1435546875,
|
| 343 |
+
"learning_rate": 0.029857831269562547,
|
| 344 |
+
"loss": 0.2795,
|
| 345 |
+
"num_input_tokens_seen": 101744,
|
| 346 |
+
"step": 160,
|
| 347 |
+
"train_runtime": 41.3805,
|
| 348 |
+
"train_tokens_per_second": 2458.745
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 2.8947368421052633,
|
| 352 |
+
"grad_norm": 0.1689453125,
|
| 353 |
+
"learning_rate": 0.029824548198094384,
|
| 354 |
+
"loss": 0.1927,
|
| 355 |
+
"num_input_tokens_seen": 105264,
|
| 356 |
+
"step": 165,
|
| 357 |
+
"train_runtime": 42.549,
|
| 358 |
+
"train_tokens_per_second": 2473.946
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 2.982456140350877,
|
| 362 |
+
"grad_norm": 0.1376953125,
|
| 363 |
+
"learning_rate": 0.029787790421012244,
|
| 364 |
+
"loss": 0.2478,
|
| 365 |
+
"num_input_tokens_seen": 108592,
|
| 366 |
+
"step": 170,
|
| 367 |
+
"train_runtime": 43.6989,
|
| 368 |
+
"train_tokens_per_second": 2485.008
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.0,
|
| 372 |
+
"eval_loss": 0.25780415534973145,
|
| 373 |
+
"eval_runtime": 0.8133,
|
| 374 |
+
"eval_samples_per_second": 30.739,
|
| 375 |
+
"eval_steps_per_second": 8.607,
|
| 376 |
+
"num_input_tokens_seen": 108712,
|
| 377 |
+
"step": 171
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.0701754385964914,
|
| 381 |
+
"grad_norm": 0.10302734375,
|
| 382 |
+
"learning_rate": 0.029747566553921325,
|
| 383 |
+
"loss": 0.5776,
|
| 384 |
+
"num_input_tokens_seen": 111976,
|
| 385 |
+
"step": 175,
|
| 386 |
+
"train_runtime": 46.769,
|
| 387 |
+
"train_tokens_per_second": 2394.235
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.1578947368421053,
|
| 391 |
+
"grad_norm": 0.042724609375,
|
| 392 |
+
"learning_rate": 0.029703886024838914,
|
| 393 |
+
"loss": 0.2945,
|
| 394 |
+
"num_input_tokens_seen": 114568,
|
| 395 |
+
"step": 180,
|
| 396 |
+
"train_runtime": 47.7992,
|
| 397 |
+
"train_tokens_per_second": 2396.862
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 3.245614035087719,
|
| 401 |
+
"grad_norm": 0.08642578125,
|
| 402 |
+
"learning_rate": 0.0296567590719846,
|
| 403 |
+
"loss": 0.3428,
|
| 404 |
+
"num_input_tokens_seen": 117384,
|
| 405 |
+
"step": 185,
|
| 406 |
+
"train_runtime": 48.8647,
|
| 407 |
+
"train_tokens_per_second": 2402.226
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 3.3333333333333335,
|
| 411 |
+
"grad_norm": 0.0654296875,
|
| 412 |
+
"learning_rate": 0.029606196741380517,
|
| 413 |
+
"loss": 0.2289,
|
| 414 |
+
"num_input_tokens_seen": 120968,
|
| 415 |
+
"step": 190,
|
| 416 |
+
"train_runtime": 50.051,
|
| 417 |
+
"train_tokens_per_second": 2416.893
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 3.4210526315789473,
|
| 421 |
+
"grad_norm": 0.1689453125,
|
| 422 |
+
"learning_rate": 0.029552210884262308,
|
| 423 |
+
"loss": 0.2794,
|
| 424 |
+
"num_input_tokens_seen": 124616,
|
| 425 |
+
"step": 195,
|
| 426 |
+
"train_runtime": 51.2594,
|
| 427 |
+
"train_tokens_per_second": 2431.085
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.5087719298245617,
|
| 431 |
+
"grad_norm": 0.0732421875,
|
| 432 |
+
"learning_rate": 0.029494814154301326,
|
| 433 |
+
"loss": 0.2983,
|
| 434 |
+
"num_input_tokens_seen": 128104,
|
| 435 |
+
"step": 200,
|
| 436 |
+
"train_runtime": 52.4262,
|
| 437 |
+
"train_tokens_per_second": 2443.511
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 3.5964912280701755,
|
| 441 |
+
"grad_norm": 0.09228515625,
|
| 442 |
+
"learning_rate": 0.029434020004638753,
|
| 443 |
+
"loss": 0.2329,
|
| 444 |
+
"num_input_tokens_seen": 131528,
|
| 445 |
+
"step": 205,
|
| 446 |
+
"train_runtime": 53.6072,
|
| 447 |
+
"train_tokens_per_second": 2453.553
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 3.6842105263157894,
|
| 451 |
+
"grad_norm": 0.11572265625,
|
| 452 |
+
"learning_rate": 0.029369842684732334,
|
| 453 |
+
"loss": 0.2165,
|
| 454 |
+
"num_input_tokens_seen": 134600,
|
| 455 |
+
"step": 210,
|
| 456 |
+
"train_runtime": 54.7146,
|
| 457 |
+
"train_tokens_per_second": 2460.038
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.7719298245614032,
|
| 461 |
+
"grad_norm": 0.0751953125,
|
| 462 |
+
"learning_rate": 0.02930229723701646,
|
| 463 |
+
"loss": 0.2549,
|
| 464 |
+
"num_input_tokens_seen": 137896,
|
| 465 |
+
"step": 215,
|
| 466 |
+
"train_runtime": 55.8416,
|
| 467 |
+
"train_tokens_per_second": 2469.413
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.8596491228070176,
|
| 471 |
+
"grad_norm": 0.1064453125,
|
| 472 |
+
"learning_rate": 0.029231399493376414,
|
| 473 |
+
"loss": 0.2149,
|
| 474 |
+
"num_input_tokens_seen": 141064,
|
| 475 |
+
"step": 220,
|
| 476 |
+
"train_runtime": 56.9668,
|
| 477 |
+
"train_tokens_per_second": 2476.249
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 3.9473684210526314,
|
| 481 |
+
"grad_norm": 0.0810546875,
|
| 482 |
+
"learning_rate": 0.02915716607143754,
|
| 483 |
+
"loss": 0.1393,
|
| 484 |
+
"num_input_tokens_seen": 144104,
|
| 485 |
+
"step": 225,
|
| 486 |
+
"train_runtime": 58.0651,
|
| 487 |
+
"train_tokens_per_second": 2481.764
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 4.0,
|
| 491 |
+
"eval_loss": 0.21973547339439392,
|
| 492 |
+
"eval_runtime": 0.8236,
|
| 493 |
+
"eval_samples_per_second": 30.354,
|
| 494 |
+
"eval_steps_per_second": 8.499,
|
| 495 |
+
"num_input_tokens_seen": 145296,
|
| 496 |
+
"step": 228
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 4.035087719298246,
|
| 500 |
+
"grad_norm": 0.107421875,
|
| 501 |
+
"learning_rate": 0.029079614370670265,
|
| 502 |
+
"loss": 0.2439,
|
| 503 |
+
"num_input_tokens_seen": 146576,
|
| 504 |
+
"step": 230,
|
| 505 |
+
"train_runtime": 60.7932,
|
| 506 |
+
"train_tokens_per_second": 2411.059
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 4.12280701754386,
|
| 510 |
+
"grad_norm": 0.115234375,
|
| 511 |
+
"learning_rate": 0.028998762568311857,
|
| 512 |
+
"loss": 0.281,
|
| 513 |
+
"num_input_tokens_seen": 149456,
|
| 514 |
+
"step": 235,
|
| 515 |
+
"train_runtime": 61.8714,
|
| 516 |
+
"train_tokens_per_second": 2415.593
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 4.2105263157894735,
|
| 520 |
+
"grad_norm": 0.2470703125,
|
| 521 |
+
"learning_rate": 0.028914629615105897,
|
| 522 |
+
"loss": 0.2493,
|
| 523 |
+
"num_input_tokens_seen": 152496,
|
| 524 |
+
"step": 240,
|
| 525 |
+
"train_runtime": 62.9736,
|
| 526 |
+
"train_tokens_per_second": 2421.585
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 4.298245614035087,
|
| 530 |
+
"grad_norm": 0.1005859375,
|
| 531 |
+
"learning_rate": 0.028827235230860424,
|
| 532 |
+
"loss": 0.2495,
|
| 533 |
+
"num_input_tokens_seen": 155504,
|
| 534 |
+
"step": 245,
|
| 535 |
+
"train_runtime": 64.0584,
|
| 536 |
+
"train_tokens_per_second": 2427.536
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 4.385964912280702,
|
| 540 |
+
"grad_norm": 0.0732421875,
|
| 541 |
+
"learning_rate": 0.02873659989982586,
|
| 542 |
+
"loss": 0.1699,
|
| 543 |
+
"num_input_tokens_seen": 158640,
|
| 544 |
+
"step": 250,
|
| 545 |
+
"train_runtime": 65.1646,
|
| 546 |
+
"train_tokens_per_second": 2434.45
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 4.473684210526316,
|
| 550 |
+
"grad_norm": 0.037109375,
|
| 551 |
+
"learning_rate": 0.02864274486589371,
|
| 552 |
+
"loss": 0.2308,
|
| 553 |
+
"num_input_tokens_seen": 161296,
|
| 554 |
+
"step": 255,
|
| 555 |
+
"train_runtime": 66.2052,
|
| 556 |
+
"train_tokens_per_second": 2436.303
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 4.56140350877193,
|
| 560 |
+
"grad_norm": 0.07177734375,
|
| 561 |
+
"learning_rate": 0.028545692127617244,
|
| 562 |
+
"loss": 0.2325,
|
| 563 |
+
"num_input_tokens_seen": 164656,
|
| 564 |
+
"step": 260,
|
| 565 |
+
"train_runtime": 67.3661,
|
| 566 |
+
"train_tokens_per_second": 2444.198
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 4.649122807017544,
|
| 570 |
+
"grad_norm": 16.875,
|
| 571 |
+
"learning_rate": 0.0284454644330553,
|
| 572 |
+
"loss": 0.3632,
|
| 573 |
+
"num_input_tokens_seen": 168240,
|
| 574 |
+
"step": 265,
|
| 575 |
+
"train_runtime": 68.5551,
|
| 576 |
+
"train_tokens_per_second": 2454.083
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 4.7368421052631575,
|
| 580 |
+
"grad_norm": 0.06689453125,
|
| 581 |
+
"learning_rate": 0.02834208527444037,
|
| 582 |
+
"loss": 0.1337,
|
| 583 |
+
"num_input_tokens_seen": 171568,
|
| 584 |
+
"step": 270,
|
| 585 |
+
"train_runtime": 69.6957,
|
| 586 |
+
"train_tokens_per_second": 2461.671
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 4.824561403508772,
|
| 590 |
+
"grad_norm": 0.0703125,
|
| 591 |
+
"learning_rate": 0.028235578882672318,
|
| 592 |
+
"loss": 0.1933,
|
| 593 |
+
"num_input_tokens_seen": 174320,
|
| 594 |
+
"step": 275,
|
| 595 |
+
"train_runtime": 70.7566,
|
| 596 |
+
"train_tokens_per_second": 2463.657
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 4.912280701754386,
|
| 600 |
+
"grad_norm": 0.103515625,
|
| 601 |
+
"learning_rate": 0.028125970221638905,
|
| 602 |
+
"loss": 0.1485,
|
| 603 |
+
"num_input_tokens_seen": 178448,
|
| 604 |
+
"step": 280,
|
| 605 |
+
"train_runtime": 72.0434,
|
| 606 |
+
"train_tokens_per_second": 2476.95
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 5.0,
|
| 610 |
+
"grad_norm": 0.201171875,
|
| 611 |
+
"learning_rate": 0.028013284982364554,
|
| 612 |
+
"loss": 0.2443,
|
| 613 |
+
"num_input_tokens_seen": 181408,
|
| 614 |
+
"step": 285,
|
| 615 |
+
"train_runtime": 73.1202,
|
| 616 |
+
"train_tokens_per_second": 2480.957
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 5.0,
|
| 620 |
+
"eval_loss": 0.19104143977165222,
|
| 621 |
+
"eval_runtime": 0.8198,
|
| 622 |
+
"eval_samples_per_second": 30.497,
|
| 623 |
+
"eval_steps_per_second": 8.539,
|
| 624 |
+
"num_input_tokens_seen": 181408,
|
| 625 |
+
"step": 285
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 5.087719298245614,
|
| 629 |
+
"grad_norm": 0.1494140625,
|
| 630 |
+
"learning_rate": 0.027897549576988666,
|
| 631 |
+
"loss": 0.2104,
|
| 632 |
+
"num_input_tokens_seen": 184352,
|
| 633 |
+
"step": 290,
|
| 634 |
+
"train_runtime": 75.914,
|
| 635 |
+
"train_tokens_per_second": 2428.431
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 5.175438596491228,
|
| 639 |
+
"grad_norm": 0.08203125,
|
| 640 |
+
"learning_rate": 0.027778791132574907,
|
| 641 |
+
"loss": 0.2249,
|
| 642 |
+
"num_input_tokens_seen": 187680,
|
| 643 |
+
"step": 295,
|
| 644 |
+
"train_runtime": 77.0713,
|
| 645 |
+
"train_tokens_per_second": 2435.146
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 5.2631578947368425,
|
| 649 |
+
"grad_norm": 0.0849609375,
|
| 650 |
+
"learning_rate": 0.02765703748475293,
|
| 651 |
+
"loss": 0.2072,
|
| 652 |
+
"num_input_tokens_seen": 190880,
|
| 653 |
+
"step": 300,
|
| 654 |
+
"train_runtime": 78.1954,
|
| 655 |
+
"train_tokens_per_second": 2441.065
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 5.350877192982456,
|
| 659 |
+
"grad_norm": 0.068359375,
|
| 660 |
+
"learning_rate": 0.027532317171194046,
|
| 661 |
+
"loss": 0.1397,
|
| 662 |
+
"num_input_tokens_seen": 194592,
|
| 663 |
+
"step": 305,
|
| 664 |
+
"train_runtime": 79.4127,
|
| 665 |
+
"train_tokens_per_second": 2450.388
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 5.43859649122807,
|
| 669 |
+
"grad_norm": 0.1142578125,
|
| 670 |
+
"learning_rate": 0.027404659424922272,
|
| 671 |
+
"loss": 0.1518,
|
| 672 |
+
"num_input_tokens_seen": 198720,
|
| 673 |
+
"step": 310,
|
| 674 |
+
"train_runtime": 80.6941,
|
| 675 |
+
"train_tokens_per_second": 2462.632
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 5.526315789473684,
|
| 679 |
+
"grad_norm": 0.052001953125,
|
| 680 |
+
"learning_rate": 0.027274094167462487,
|
| 681 |
+
"loss": 0.2501,
|
| 682 |
+
"num_input_tokens_seen": 201856,
|
| 683 |
+
"step": 315,
|
| 684 |
+
"train_runtime": 81.8118,
|
| 685 |
+
"train_tokens_per_second": 2467.322
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 5.614035087719298,
|
| 689 |
+
"grad_norm": 0.115234375,
|
| 690 |
+
"learning_rate": 0.02714065200182714,
|
| 691 |
+
"loss": 0.2199,
|
| 692 |
+
"num_input_tokens_seen": 204800,
|
| 693 |
+
"step": 320,
|
| 694 |
+
"train_runtime": 82.897,
|
| 695 |
+
"train_tokens_per_second": 2470.536
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 5.701754385964913,
|
| 699 |
+
"grad_norm": 0.064453125,
|
| 700 |
+
"learning_rate": 0.02700436420534326,
|
| 701 |
+
"loss": 0.2216,
|
| 702 |
+
"num_input_tokens_seen": 208064,
|
| 703 |
+
"step": 325,
|
| 704 |
+
"train_runtime": 84.0239,
|
| 705 |
+
"train_tokens_per_second": 2476.248
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 5.7894736842105265,
|
| 709 |
+
"grad_norm": 0.042236328125,
|
| 710 |
+
"learning_rate": 0.02686526272232141,
|
| 711 |
+
"loss": 0.2104,
|
| 712 |
+
"num_input_tokens_seen": 211168,
|
| 713 |
+
"step": 330,
|
| 714 |
+
"train_runtime": 85.1357,
|
| 715 |
+
"train_tokens_per_second": 2480.37
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 5.87719298245614,
|
| 719 |
+
"grad_norm": 0.030029296875,
|
| 720 |
+
"learning_rate": 0.026723380156568298,
|
| 721 |
+
"loss": 0.2176,
|
| 722 |
+
"num_input_tokens_seen": 214144,
|
| 723 |
+
"step": 335,
|
| 724 |
+
"train_runtime": 86.2206,
|
| 725 |
+
"train_tokens_per_second": 2483.675
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 5.964912280701754,
|
| 729 |
+
"grad_norm": 0.0262451171875,
|
| 730 |
+
"learning_rate": 0.026578749763744813,
|
| 731 |
+
"loss": 0.1904,
|
| 732 |
+
"num_input_tokens_seen": 217216,
|
| 733 |
+
"step": 340,
|
| 734 |
+
"train_runtime": 87.3287,
|
| 735 |
+
"train_tokens_per_second": 2487.339
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 6.0,
|
| 739 |
+
"eval_loss": 0.21254059672355652,
|
| 740 |
+
"eval_runtime": 0.8177,
|
| 741 |
+
"eval_samples_per_second": 30.574,
|
| 742 |
+
"eval_steps_per_second": 8.561,
|
| 743 |
+
"num_input_tokens_seen": 217760,
|
| 744 |
+
"step": 342
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 6.052631578947368,
|
| 748 |
+
"grad_norm": 0.0159912109375,
|
| 749 |
+
"learning_rate": 0.026431405443571282,
|
| 750 |
+
"loss": 0.1289,
|
| 751 |
+
"num_input_tokens_seen": 219808,
|
| 752 |
+
"step": 345,
|
| 753 |
+
"train_runtime": 90.1714,
|
| 754 |
+
"train_tokens_per_second": 2437.669
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 6.140350877192983,
|
| 758 |
+
"grad_norm": 0.09521484375,
|
| 759 |
+
"learning_rate": 0.02628138173188176,
|
| 760 |
+
"loss": 0.2063,
|
| 761 |
+
"num_input_tokens_seen": 223232,
|
| 762 |
+
"step": 350,
|
| 763 |
+
"train_runtime": 91.3323,
|
| 764 |
+
"train_tokens_per_second": 2444.173
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 6.228070175438597,
|
| 768 |
+
"grad_norm": 0.0294189453125,
|
| 769 |
+
"learning_rate": 0.026128713792529224,
|
| 770 |
+
"loss": 0.2161,
|
| 771 |
+
"num_input_tokens_seen": 227008,
|
| 772 |
+
"step": 355,
|
| 773 |
+
"train_runtime": 92.5629,
|
| 774 |
+
"train_tokens_per_second": 2452.473
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 6.315789473684211,
|
| 778 |
+
"grad_norm": 0.038818359375,
|
| 779 |
+
"learning_rate": 0.025973437409143554,
|
| 780 |
+
"loss": 0.2713,
|
| 781 |
+
"num_input_tokens_seen": 230016,
|
| 782 |
+
"step": 360,
|
| 783 |
+
"train_runtime": 93.6502,
|
| 784 |
+
"train_tokens_per_second": 2456.119
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 6.4035087719298245,
|
| 788 |
+
"grad_norm": 0.1220703125,
|
| 789 |
+
"learning_rate": 0.025815588976744273,
|
| 790 |
+
"loss": 0.234,
|
| 791 |
+
"num_input_tokens_seen": 233632,
|
| 792 |
+
"step": 365,
|
| 793 |
+
"train_runtime": 94.8403,
|
| 794 |
+
"train_tokens_per_second": 2463.425
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 6.491228070175438,
|
| 798 |
+
"grad_norm": 0.057861328125,
|
| 799 |
+
"learning_rate": 0.02565520549320996,
|
| 800 |
+
"loss": 0.1921,
|
| 801 |
+
"num_input_tokens_seen": 236960,
|
| 802 |
+
"step": 370,
|
| 803 |
+
"train_runtime": 95.9895,
|
| 804 |
+
"train_tokens_per_second": 2468.604
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 6.578947368421053,
|
| 808 |
+
"grad_norm": 0.068359375,
|
| 809 |
+
"learning_rate": 0.02549232455060637,
|
| 810 |
+
"loss": 0.1572,
|
| 811 |
+
"num_input_tokens_seen": 240160,
|
| 812 |
+
"step": 375,
|
| 813 |
+
"train_runtime": 97.1155,
|
| 814 |
+
"train_tokens_per_second": 2472.932
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 6.666666666666667,
|
| 818 |
+
"grad_norm": 0.055419921875,
|
| 819 |
+
"learning_rate": 0.025326984326375274,
|
| 820 |
+
"loss": 0.2613,
|
| 821 |
+
"num_input_tokens_seen": 243264,
|
| 822 |
+
"step": 380,
|
| 823 |
+
"train_runtime": 98.2242,
|
| 824 |
+
"train_tokens_per_second": 2476.619
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 6.754385964912281,
|
| 828 |
+
"grad_norm": 0.02587890625,
|
| 829 |
+
"learning_rate": 0.025159223574386116,
|
| 830 |
+
"loss": 0.2346,
|
| 831 |
+
"num_input_tokens_seen": 246560,
|
| 832 |
+
"step": 385,
|
| 833 |
+
"train_runtime": 99.3552,
|
| 834 |
+
"train_tokens_per_second": 2481.602
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 6.842105263157895,
|
| 838 |
+
"grad_norm": 0.034423828125,
|
| 839 |
+
"learning_rate": 0.02498908161585253,
|
| 840 |
+
"loss": 0.1977,
|
| 841 |
+
"num_input_tokens_seen": 249408,
|
| 842 |
+
"step": 390,
|
| 843 |
+
"train_runtime": 100.4363,
|
| 844 |
+
"train_tokens_per_second": 2483.246
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 6.9298245614035086,
|
| 848 |
+
"grad_norm": 0.0634765625,
|
| 849 |
+
"learning_rate": 0.024816598330115895,
|
| 850 |
+
"loss": 0.092,
|
| 851 |
+
"num_input_tokens_seen": 252192,
|
| 852 |
+
"step": 395,
|
| 853 |
+
"train_runtime": 101.4958,
|
| 854 |
+
"train_tokens_per_second": 2484.752
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 7.0,
|
| 858 |
+
"eval_loss": 0.199934720993042,
|
| 859 |
+
"eval_runtime": 0.8143,
|
| 860 |
+
"eval_samples_per_second": 30.7,
|
| 861 |
+
"eval_steps_per_second": 8.596,
|
| 862 |
+
"num_input_tokens_seen": 254568,
|
| 863 |
+
"step": 399
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 7.017543859649122,
|
| 867 |
+
"grad_norm": 0.0380859375,
|
| 868 |
+
"learning_rate": 0.024641814145298088,
|
| 869 |
+
"loss": 0.1226,
|
| 870 |
+
"num_input_tokens_seen": 255432,
|
| 871 |
+
"step": 400,
|
| 872 |
+
"train_runtime": 104.2731,
|
| 873 |
+
"train_tokens_per_second": 2449.645
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 7.105263157894737,
|
| 877 |
+
"grad_norm": 0.041259765625,
|
| 878 |
+
"learning_rate": 0.024464770028825585,
|
| 879 |
+
"loss": 0.2957,
|
| 880 |
+
"num_input_tokens_seen": 258408,
|
| 881 |
+
"step": 405,
|
| 882 |
+
"train_runtime": 105.3703,
|
| 883 |
+
"train_tokens_per_second": 2452.38
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 7.192982456140351,
|
| 887 |
+
"grad_norm": 0.048828125,
|
| 888 |
+
"learning_rate": 0.024285507477827137,
|
| 889 |
+
"loss": 0.1356,
|
| 890 |
+
"num_input_tokens_seen": 262024,
|
| 891 |
+
"step": 410,
|
| 892 |
+
"train_runtime": 106.5773,
|
| 893 |
+
"train_tokens_per_second": 2458.534
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 7.280701754385965,
|
| 897 |
+
"grad_norm": 0.042236328125,
|
| 898 |
+
"learning_rate": 0.02410406850940735,
|
| 899 |
+
"loss": 0.1575,
|
| 900 |
+
"num_input_tokens_seen": 265032,
|
| 901 |
+
"step": 415,
|
| 902 |
+
"train_runtime": 107.6678,
|
| 903 |
+
"train_tokens_per_second": 2461.571
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 7.368421052631579,
|
| 907 |
+
"grad_norm": 0.058349609375,
|
| 908 |
+
"learning_rate": 0.02392049565079834,
|
| 909 |
+
"loss": 0.2333,
|
| 910 |
+
"num_input_tokens_seen": 267656,
|
| 911 |
+
"step": 420,
|
| 912 |
+
"train_runtime": 108.7096,
|
| 913 |
+
"train_tokens_per_second": 2462.12
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 7.456140350877193,
|
| 917 |
+
"grad_norm": 0.033447265625,
|
| 918 |
+
"learning_rate": 0.023734831929391822,
|
| 919 |
+
"loss": 0.152,
|
| 920 |
+
"num_input_tokens_seen": 271592,
|
| 921 |
+
"step": 425,
|
| 922 |
+
"train_runtime": 109.9651,
|
| 923 |
+
"train_tokens_per_second": 2469.802
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 7.543859649122807,
|
| 927 |
+
"grad_norm": 0.026611328125,
|
| 928 |
+
"learning_rate": 0.02354712086265399,
|
| 929 |
+
"loss": 0.0956,
|
| 930 |
+
"num_input_tokens_seen": 275016,
|
| 931 |
+
"step": 430,
|
| 932 |
+
"train_runtime": 111.13,
|
| 933 |
+
"train_tokens_per_second": 2474.723
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 7.631578947368421,
|
| 937 |
+
"grad_norm": 0.031982421875,
|
| 938 |
+
"learning_rate": 0.023357406447925527,
|
| 939 |
+
"loss": 0.1738,
|
| 940 |
+
"num_input_tokens_seen": 278248,
|
| 941 |
+
"step": 435,
|
| 942 |
+
"train_runtime": 112.259,
|
| 943 |
+
"train_tokens_per_second": 2478.625
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 7.719298245614035,
|
| 947 |
+
"grad_norm": 0.02197265625,
|
| 948 |
+
"learning_rate": 0.023165733152109094,
|
| 949 |
+
"loss": 0.0827,
|
| 950 |
+
"num_input_tokens_seen": 281320,
|
| 951 |
+
"step": 440,
|
| 952 |
+
"train_runtime": 113.3482,
|
| 953 |
+
"train_tokens_per_second": 2481.909
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 7.807017543859649,
|
| 957 |
+
"grad_norm": 0.046142578125,
|
| 958 |
+
"learning_rate": 0.022972145901246837,
|
| 959 |
+
"loss": 0.173,
|
| 960 |
+
"num_input_tokens_seen": 284360,
|
| 961 |
+
"step": 445,
|
| 962 |
+
"train_runtime": 114.445,
|
| 963 |
+
"train_tokens_per_second": 2484.687
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 7.894736842105263,
|
| 967 |
+
"grad_norm": 0.0703125,
|
| 968 |
+
"learning_rate": 0.022776690069990207,
|
| 969 |
+
"loss": 0.2527,
|
| 970 |
+
"num_input_tokens_seen": 287944,
|
| 971 |
+
"step": 450,
|
| 972 |
+
"train_runtime": 115.6396,
|
| 973 |
+
"train_tokens_per_second": 2490.013
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 7.982456140350877,
|
| 977 |
+
"grad_norm": 0.055419921875,
|
| 978 |
+
"learning_rate": 0.022579411470964646,
|
| 979 |
+
"loss": 0.285,
|
| 980 |
+
"num_input_tokens_seen": 290696,
|
| 981 |
+
"step": 455,
|
| 982 |
+
"train_runtime": 116.7001,
|
| 983 |
+
"train_tokens_per_second": 2490.966
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 8.0,
|
| 987 |
+
"eval_loss": 0.18686185777187347,
|
| 988 |
+
"eval_runtime": 0.8093,
|
| 989 |
+
"eval_samples_per_second": 30.89,
|
| 990 |
+
"eval_steps_per_second": 8.649,
|
| 991 |
+
"num_input_tokens_seen": 291000,
|
| 992 |
+
"step": 456
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 8.070175438596491,
|
| 996 |
+
"grad_norm": 0.1279296875,
|
| 997 |
+
"learning_rate": 0.022380356344031675,
|
| 998 |
+
"loss": 0.1509,
|
| 999 |
+
"num_input_tokens_seen": 293432,
|
| 1000 |
+
"step": 460,
|
| 1001 |
+
"train_runtime": 119.4256,
|
| 1002 |
+
"train_tokens_per_second": 2457.027
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 8.157894736842104,
|
| 1006 |
+
"grad_norm": 0.038330078125,
|
| 1007 |
+
"learning_rate": 0.02217957134545074,
|
| 1008 |
+
"loss": 0.2371,
|
| 1009 |
+
"num_input_tokens_seen": 296888,
|
| 1010 |
+
"step": 465,
|
| 1011 |
+
"train_runtime": 120.5995,
|
| 1012 |
+
"train_tokens_per_second": 2461.768
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 8.24561403508772,
|
| 1016 |
+
"grad_norm": 0.0927734375,
|
| 1017 |
+
"learning_rate": 0.02197710353694355,
|
| 1018 |
+
"loss": 0.201,
|
| 1019 |
+
"num_input_tokens_seen": 300312,
|
| 1020 |
+
"step": 470,
|
| 1021 |
+
"train_runtime": 121.761,
|
| 1022 |
+
"train_tokens_per_second": 2466.405
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 8.333333333333334,
|
| 1026 |
+
"grad_norm": 0.0712890625,
|
| 1027 |
+
"learning_rate": 0.02177300037466334,
|
| 1028 |
+
"loss": 0.1655,
|
| 1029 |
+
"num_input_tokens_seen": 303544,
|
| 1030 |
+
"step": 475,
|
| 1031 |
+
"train_runtime": 122.8907,
|
| 1032 |
+
"train_tokens_per_second": 2470.031
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 8.421052631578947,
|
| 1036 |
+
"grad_norm": 0.09716796875,
|
| 1037 |
+
"learning_rate": 0.02156730969807168,
|
| 1038 |
+
"loss": 0.1945,
|
| 1039 |
+
"num_input_tokens_seen": 306872,
|
| 1040 |
+
"step": 480,
|
| 1041 |
+
"train_runtime": 124.0358,
|
| 1042 |
+
"train_tokens_per_second": 2474.059
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 8.508771929824562,
|
| 1046 |
+
"grad_norm": 0.0257568359375,
|
| 1047 |
+
"learning_rate": 0.021360079718725448,
|
| 1048 |
+
"loss": 0.1618,
|
| 1049 |
+
"num_input_tokens_seen": 309592,
|
| 1050 |
+
"step": 485,
|
| 1051 |
+
"train_runtime": 125.0818,
|
| 1052 |
+
"train_tokens_per_second": 2475.117
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 8.596491228070175,
|
| 1056 |
+
"grad_norm": 0.041259765625,
|
| 1057 |
+
"learning_rate": 0.021151359008976602,
|
| 1058 |
+
"loss": 0.1631,
|
| 1059 |
+
"num_input_tokens_seen": 312728,
|
| 1060 |
+
"step": 490,
|
| 1061 |
+
"train_runtime": 126.2006,
|
| 1062 |
+
"train_tokens_per_second": 2478.024
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 8.68421052631579,
|
| 1066 |
+
"grad_norm": 0.06396484375,
|
| 1067 |
+
"learning_rate": 0.02094119649058735,
|
| 1068 |
+
"loss": 0.2375,
|
| 1069 |
+
"num_input_tokens_seen": 316088,
|
| 1070 |
+
"step": 495,
|
| 1071 |
+
"train_runtime": 127.3629,
|
| 1072 |
+
"train_tokens_per_second": 2481.79
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 8.771929824561404,
|
| 1076 |
+
"grad_norm": 0.05126953125,
|
| 1077 |
+
"learning_rate": 0.020729641423263476,
|
| 1078 |
+
"loss": 0.1928,
|
| 1079 |
+
"num_input_tokens_seen": 319352,
|
| 1080 |
+
"step": 500,
|
| 1081 |
+
"train_runtime": 128.493,
|
| 1082 |
+
"train_tokens_per_second": 2485.365
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 8.859649122807017,
|
| 1086 |
+
"grad_norm": 0.05810546875,
|
| 1087 |
+
"learning_rate": 0.0205167433931084,
|
| 1088 |
+
"loss": 0.199,
|
| 1089 |
+
"num_input_tokens_seen": 323256,
|
| 1090 |
+
"step": 505,
|
| 1091 |
+
"train_runtime": 129.7371,
|
| 1092 |
+
"train_tokens_per_second": 2491.624
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 8.947368421052632,
|
| 1096 |
+
"grad_norm": 0.08056640625,
|
| 1097 |
+
"learning_rate": 0.020302552301000754,
|
| 1098 |
+
"loss": 0.1165,
|
| 1099 |
+
"num_input_tokens_seen": 326200,
|
| 1100 |
+
"step": 510,
|
| 1101 |
+
"train_runtime": 130.8307,
|
| 1102 |
+
"train_tokens_per_second": 2493.298
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 9.0,
|
| 1106 |
+
"eval_loss": 0.20220531523227692,
|
| 1107 |
+
"eval_runtime": 0.813,
|
| 1108 |
+
"eval_samples_per_second": 30.752,
|
| 1109 |
+
"eval_steps_per_second": 8.61,
|
| 1110 |
+
"num_input_tokens_seen": 327792,
|
| 1111 |
+
"step": 513
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 9.035087719298245,
|
| 1115 |
+
"grad_norm": 0.1767578125,
|
| 1116 |
+
"learning_rate": 0.02008711835089822,
|
| 1117 |
+
"loss": 0.2465,
|
| 1118 |
+
"num_input_tokens_seen": 329008,
|
| 1119 |
+
"step": 515,
|
| 1120 |
+
"train_runtime": 133.5444,
|
| 1121 |
+
"train_tokens_per_second": 2463.659
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 9.12280701754386,
|
| 1125 |
+
"grad_norm": 0.0400390625,
|
| 1126 |
+
"learning_rate": 0.019870492038070255,
|
| 1127 |
+
"loss": 0.1526,
|
| 1128 |
+
"num_input_tokens_seen": 332304,
|
| 1129 |
+
"step": 520,
|
| 1130 |
+
"train_runtime": 134.6906,
|
| 1131 |
+
"train_tokens_per_second": 2467.166
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 9.210526315789474,
|
| 1135 |
+
"grad_norm": 0.0908203125,
|
| 1136 |
+
"learning_rate": 0.01965272413726258,
|
| 1137 |
+
"loss": 0.1554,
|
| 1138 |
+
"num_input_tokens_seen": 335792,
|
| 1139 |
+
"step": 525,
|
| 1140 |
+
"train_runtime": 135.8854,
|
| 1141 |
+
"train_tokens_per_second": 2471.141
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 9.298245614035087,
|
| 1145 |
+
"grad_norm": 0.039306640625,
|
| 1146 |
+
"learning_rate": 0.01943386569079618,
|
| 1147 |
+
"loss": 0.1411,
|
| 1148 |
+
"num_input_tokens_seen": 338672,
|
| 1149 |
+
"step": 530,
|
| 1150 |
+
"train_runtime": 136.9689,
|
| 1151 |
+
"train_tokens_per_second": 2472.619
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 9.385964912280702,
|
| 1155 |
+
"grad_norm": 0.04638671875,
|
| 1156 |
+
"learning_rate": 0.01921396799660354,
|
| 1157 |
+
"loss": 0.1062,
|
| 1158 |
+
"num_input_tokens_seen": 341968,
|
| 1159 |
+
"step": 535,
|
| 1160 |
+
"train_runtime": 138.0998,
|
| 1161 |
+
"train_tokens_per_second": 2476.238
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 9.473684210526315,
|
| 1165 |
+
"grad_norm": 0.10302734375,
|
| 1166 |
+
"learning_rate": 0.018993082596205,
|
| 1167 |
+
"loss": 0.1947,
|
| 1168 |
+
"num_input_tokens_seen": 345232,
|
| 1169 |
+
"step": 540,
|
| 1170 |
+
"train_runtime": 139.2299,
|
| 1171 |
+
"train_tokens_per_second": 2479.583
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 9.56140350877193,
|
| 1175 |
+
"grad_norm": 0.087890625,
|
| 1176 |
+
"learning_rate": 0.018771261262628014,
|
| 1177 |
+
"loss": 0.1594,
|
| 1178 |
+
"num_input_tokens_seen": 348656,
|
| 1179 |
+
"step": 545,
|
| 1180 |
+
"train_runtime": 140.3927,
|
| 1181 |
+
"train_tokens_per_second": 2483.433
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 9.649122807017545,
|
| 1185 |
+
"grad_norm": 0.06884765625,
|
| 1186 |
+
"learning_rate": 0.018548555988272136,
|
| 1187 |
+
"loss": 0.2227,
|
| 1188 |
+
"num_input_tokens_seen": 351248,
|
| 1189 |
+
"step": 550,
|
| 1190 |
+
"train_runtime": 141.4231,
|
| 1191 |
+
"train_tokens_per_second": 2483.668
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 9.736842105263158,
|
| 1195 |
+
"grad_norm": 0.044189453125,
|
| 1196 |
+
"learning_rate": 0.018325018972722578,
|
| 1197 |
+
"loss": 0.1262,
|
| 1198 |
+
"num_input_tokens_seen": 355024,
|
| 1199 |
+
"step": 555,
|
| 1200 |
+
"train_runtime": 142.6458,
|
| 1201 |
+
"train_tokens_per_second": 2488.85
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 9.824561403508772,
|
| 1205 |
+
"grad_norm": 0.02197265625,
|
| 1206 |
+
"learning_rate": 0.01810070261051526,
|
| 1207 |
+
"loss": 0.1045,
|
| 1208 |
+
"num_input_tokens_seen": 357776,
|
| 1209 |
+
"step": 560,
|
| 1210 |
+
"train_runtime": 143.7013,
|
| 1211 |
+
"train_tokens_per_second": 2489.72
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 9.912280701754385,
|
| 1215 |
+
"grad_norm": 0.09765625,
|
| 1216 |
+
"learning_rate": 0.01787565947885608,
|
| 1217 |
+
"loss": 0.1388,
|
| 1218 |
+
"num_input_tokens_seen": 360784,
|
| 1219 |
+
"step": 565,
|
| 1220 |
+
"train_runtime": 144.7877,
|
| 1221 |
+
"train_tokens_per_second": 2491.815
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 10.0,
|
| 1225 |
+
"grad_norm": 0.036376953125,
|
| 1226 |
+
"learning_rate": 0.01764994232529744,
|
| 1227 |
+
"loss": 0.1643,
|
| 1228 |
+
"num_input_tokens_seen": 363864,
|
| 1229 |
+
"step": 570,
|
| 1230 |
+
"train_runtime": 145.9089,
|
| 1231 |
+
"train_tokens_per_second": 2493.775
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 10.0,
|
| 1235 |
+
"eval_loss": 0.20832745730876923,
|
| 1236 |
+
"eval_runtime": 0.8133,
|
| 1237 |
+
"eval_samples_per_second": 30.737,
|
| 1238 |
+
"eval_steps_per_second": 8.606,
|
| 1239 |
+
"num_input_tokens_seen": 363864,
|
| 1240 |
+
"step": 570
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 10.087719298245615,
|
| 1244 |
+
"grad_norm": 0.039306640625,
|
| 1245 |
+
"learning_rate": 0.01742360405537482,
|
| 1246 |
+
"loss": 0.1147,
|
| 1247 |
+
"num_input_tokens_seen": 368120,
|
| 1248 |
+
"step": 575,
|
| 1249 |
+
"train_runtime": 148.8887,
|
| 1250 |
+
"train_tokens_per_second": 2472.451
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 10.175438596491228,
|
| 1254 |
+
"grad_norm": 0.0986328125,
|
| 1255 |
+
"learning_rate": 0.017196697720206326,
|
| 1256 |
+
"loss": 0.1527,
|
| 1257 |
+
"num_input_tokens_seen": 371416,
|
| 1258 |
+
"step": 580,
|
| 1259 |
+
"train_runtime": 150.0471,
|
| 1260 |
+
"train_tokens_per_second": 2475.33
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 10.263157894736842,
|
| 1264 |
+
"grad_norm": 0.25390625,
|
| 1265 |
+
"learning_rate": 0.01696927650405807,
|
| 1266 |
+
"loss": 0.1329,
|
| 1267 |
+
"num_input_tokens_seen": 374168,
|
| 1268 |
+
"step": 585,
|
| 1269 |
+
"train_runtime": 151.0954,
|
| 1270 |
+
"train_tokens_per_second": 2476.369
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 10.350877192982455,
|
| 1274 |
+
"grad_norm": 0.07080078125,
|
| 1275 |
+
"learning_rate": 0.016741393711878452,
|
| 1276 |
+
"loss": 0.0969,
|
| 1277 |
+
"num_input_tokens_seen": 377496,
|
| 1278 |
+
"step": 590,
|
| 1279 |
+
"train_runtime": 152.2633,
|
| 1280 |
+
"train_tokens_per_second": 2479.232
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 10.43859649122807,
|
| 1284 |
+
"grad_norm": 0.119140625,
|
| 1285 |
+
"learning_rate": 0.016513102756804024,
|
| 1286 |
+
"loss": 0.2399,
|
| 1287 |
+
"num_input_tokens_seen": 381592,
|
| 1288 |
+
"step": 595,
|
| 1289 |
+
"train_runtime": 153.5307,
|
| 1290 |
+
"train_tokens_per_second": 2485.445
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 10.526315789473685,
|
| 1294 |
+
"grad_norm": 0.12255859375,
|
| 1295 |
+
"learning_rate": 0.016284457147640083,
|
| 1296 |
+
"loss": 0.1687,
|
| 1297 |
+
"num_input_tokens_seen": 384760,
|
| 1298 |
+
"step": 600,
|
| 1299 |
+
"train_runtime": 154.6543,
|
| 1300 |
+
"train_tokens_per_second": 2487.872
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 10.614035087719298,
|
| 1304 |
+
"grad_norm": 0.130859375,
|
| 1305 |
+
"learning_rate": 0.016055510476318827,
|
| 1306 |
+
"loss": 0.2093,
|
| 1307 |
+
"num_input_tokens_seen": 387512,
|
| 1308 |
+
"step": 605,
|
| 1309 |
+
"train_runtime": 155.6985,
|
| 1310 |
+
"train_tokens_per_second": 2488.861
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 10.701754385964913,
|
| 1314 |
+
"grad_norm": 0.134765625,
|
| 1315 |
+
"learning_rate": 0.015826316405337982,
|
| 1316 |
+
"loss": 0.1559,
|
| 1317 |
+
"num_input_tokens_seen": 390680,
|
| 1318 |
+
"step": 610,
|
| 1319 |
+
"train_runtime": 156.8156,
|
| 1320 |
+
"train_tokens_per_second": 2491.334
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 10.789473684210526,
|
| 1324 |
+
"grad_norm": 0.1796875,
|
| 1325 |
+
"learning_rate": 0.015596928655182963,
|
| 1326 |
+
"loss": 0.1982,
|
| 1327 |
+
"num_input_tokens_seen": 393944,
|
| 1328 |
+
"step": 615,
|
| 1329 |
+
"train_runtime": 157.9554,
|
| 1330 |
+
"train_tokens_per_second": 2494.021
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 10.87719298245614,
|
| 1334 |
+
"grad_norm": 0.0517578125,
|
| 1335 |
+
"learning_rate": 0.015367400991735372,
|
| 1336 |
+
"loss": 0.0931,
|
| 1337 |
+
"num_input_tokens_seen": 396632,
|
| 1338 |
+
"step": 620,
|
| 1339 |
+
"train_runtime": 158.9966,
|
| 1340 |
+
"train_tokens_per_second": 2494.594
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 10.964912280701755,
|
| 1344 |
+
"grad_norm": 0.1064453125,
|
| 1345 |
+
"learning_rate": 0.015137787213670897,
|
| 1346 |
+
"loss": 0.1546,
|
| 1347 |
+
"num_input_tokens_seen": 399288,
|
| 1348 |
+
"step": 625,
|
| 1349 |
+
"train_runtime": 160.0294,
|
| 1350 |
+
"train_tokens_per_second": 2495.091
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 11.0,
|
| 1354 |
+
"eval_loss": 0.1880803257226944,
|
| 1355 |
+
"eval_runtime": 0.8131,
|
| 1356 |
+
"eval_samples_per_second": 30.745,
|
| 1357 |
+
"eval_steps_per_second": 8.609,
|
| 1358 |
+
"num_input_tokens_seen": 400104,
|
| 1359 |
+
"step": 627
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 11.052631578947368,
|
| 1363 |
+
"grad_norm": 0.031982421875,
|
| 1364 |
+
"learning_rate": 0.014908141139849508,
|
| 1365 |
+
"loss": 0.0943,
|
| 1366 |
+
"num_input_tokens_seen": 401928,
|
| 1367 |
+
"step": 630,
|
| 1368 |
+
"train_runtime": 162.7566,
|
| 1369 |
+
"train_tokens_per_second": 2469.503
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 11.140350877192983,
|
| 1373 |
+
"grad_norm": 0.042724609375,
|
| 1374 |
+
"learning_rate": 0.014678516596700955,
|
| 1375 |
+
"loss": 0.1446,
|
| 1376 |
+
"num_input_tokens_seen": 404360,
|
| 1377 |
+
"step": 635,
|
| 1378 |
+
"train_runtime": 163.781,
|
| 1379 |
+
"train_tokens_per_second": 2468.907
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 11.228070175438596,
|
| 1383 |
+
"grad_norm": 0.10205078125,
|
| 1384 |
+
"learning_rate": 0.014448967405608415,
|
| 1385 |
+
"loss": 0.1057,
|
| 1386 |
+
"num_input_tokens_seen": 407496,
|
| 1387 |
+
"step": 640,
|
| 1388 |
+
"train_runtime": 164.8951,
|
| 1389 |
+
"train_tokens_per_second": 2471.243
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 11.31578947368421,
|
| 1393 |
+
"grad_norm": 0.026611328125,
|
| 1394 |
+
"learning_rate": 0.014219547370293413,
|
| 1395 |
+
"loss": 0.0917,
|
| 1396 |
+
"num_input_tokens_seen": 410760,
|
| 1397 |
+
"step": 645,
|
| 1398 |
+
"train_runtime": 166.0422,
|
| 1399 |
+
"train_tokens_per_second": 2473.828
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 11.403508771929825,
|
| 1403 |
+
"grad_norm": 0.056640625,
|
| 1404 |
+
"learning_rate": 0.013990310264204829,
|
| 1405 |
+
"loss": 0.0825,
|
| 1406 |
+
"num_input_tokens_seen": 413832,
|
| 1407 |
+
"step": 650,
|
| 1408 |
+
"train_runtime": 167.1199,
|
| 1409 |
+
"train_tokens_per_second": 2476.259
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 11.491228070175438,
|
| 1413 |
+
"grad_norm": 0.04931640625,
|
| 1414 |
+
"learning_rate": 0.013761309817915016,
|
| 1415 |
+
"loss": 0.131,
|
| 1416 |
+
"num_input_tokens_seen": 417320,
|
| 1417 |
+
"step": 655,
|
| 1418 |
+
"train_runtime": 168.2972,
|
| 1419 |
+
"train_tokens_per_second": 2479.661
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 11.578947368421053,
|
| 1423 |
+
"grad_norm": 0.04296875,
|
| 1424 |
+
"learning_rate": 0.013532599706525941,
|
| 1425 |
+
"loss": 0.1234,
|
| 1426 |
+
"num_input_tokens_seen": 420232,
|
| 1427 |
+
"step": 660,
|
| 1428 |
+
"train_runtime": 169.3786,
|
| 1429 |
+
"train_tokens_per_second": 2481.022
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 11.666666666666666,
|
| 1433 |
+
"grad_norm": 0.01446533203125,
|
| 1434 |
+
"learning_rate": 0.013304233537088392,
|
| 1435 |
+
"loss": 0.0514,
|
| 1436 |
+
"num_input_tokens_seen": 423528,
|
| 1437 |
+
"step": 665,
|
| 1438 |
+
"train_runtime": 170.5074,
|
| 1439 |
+
"train_tokens_per_second": 2483.927
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 11.75438596491228,
|
| 1443 |
+
"grad_norm": 0.1748046875,
|
| 1444 |
+
"learning_rate": 0.013076264836037051,
|
| 1445 |
+
"loss": 0.2042,
|
| 1446 |
+
"num_input_tokens_seen": 426920,
|
| 1447 |
+
"step": 670,
|
| 1448 |
+
"train_runtime": 171.6667,
|
| 1449 |
+
"train_tokens_per_second": 2486.913
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 11.842105263157894,
|
| 1453 |
+
"grad_norm": 0.08447265625,
|
| 1454 |
+
"learning_rate": 0.012848747036644558,
|
| 1455 |
+
"loss": 0.1779,
|
| 1456 |
+
"num_input_tokens_seen": 431016,
|
| 1457 |
+
"step": 675,
|
| 1458 |
+
"train_runtime": 172.9446,
|
| 1459 |
+
"train_tokens_per_second": 2492.22
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 11.929824561403509,
|
| 1463 |
+
"grad_norm": 0.12060546875,
|
| 1464 |
+
"learning_rate": 0.012621733466497287,
|
| 1465 |
+
"loss": 0.3025,
|
| 1466 |
+
"num_input_tokens_seen": 434600,
|
| 1467 |
+
"step": 680,
|
| 1468 |
+
"train_runtime": 174.1288,
|
| 1469 |
+
"train_tokens_per_second": 2495.854
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 12.0,
|
| 1473 |
+
"eval_loss": 0.1755521595478058,
|
| 1474 |
+
"eval_runtime": 0.8133,
|
| 1475 |
+
"eval_samples_per_second": 30.738,
|
| 1476 |
+
"eval_steps_per_second": 8.607,
|
| 1477 |
+
"num_input_tokens_seen": 436440,
|
| 1478 |
+
"step": 684
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 12.017543859649123,
|
| 1482 |
+
"grad_norm": 0.1044921875,
|
| 1483 |
+
"learning_rate": 0.012395277334996044,
|
| 1484 |
+
"loss": 0.1346,
|
| 1485 |
+
"num_input_tokens_seen": 437016,
|
| 1486 |
+
"step": 685,
|
| 1487 |
+
"train_runtime": 176.8026,
|
| 1488 |
+
"train_tokens_per_second": 2471.774
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 12.105263157894736,
|
| 1492 |
+
"grad_norm": 0.0289306640625,
|
| 1493 |
+
"learning_rate": 0.012169431720884335,
|
| 1494 |
+
"loss": 0.0914,
|
| 1495 |
+
"num_input_tokens_seen": 440280,
|
| 1496 |
+
"step": 690,
|
| 1497 |
+
"train_runtime": 177.9478,
|
| 1498 |
+
"train_tokens_per_second": 2474.209
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 12.192982456140351,
|
| 1502 |
+
"grad_norm": 0.1357421875,
|
| 1503 |
+
"learning_rate": 0.01194424955980734,
|
| 1504 |
+
"loss": 0.1231,
|
| 1505 |
+
"num_input_tokens_seen": 443800,
|
| 1506 |
+
"step": 695,
|
| 1507 |
+
"train_runtime": 179.1313,
|
| 1508 |
+
"train_tokens_per_second": 2477.513
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 12.280701754385966,
|
| 1512 |
+
"grad_norm": 0.0294189453125,
|
| 1513 |
+
"learning_rate": 0.011719783631904363,
|
| 1514 |
+
"loss": 0.0562,
|
| 1515 |
+
"num_input_tokens_seen": 446168,
|
| 1516 |
+
"step": 700,
|
| 1517 |
+
"train_runtime": 180.1222,
|
| 1518 |
+
"train_tokens_per_second": 2477.03
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 12.368421052631579,
|
| 1522 |
+
"grad_norm": 0.1396484375,
|
| 1523 |
+
"learning_rate": 0.01149608654943782,
|
| 1524 |
+
"loss": 0.164,
|
| 1525 |
+
"num_input_tokens_seen": 449560,
|
| 1526 |
+
"step": 705,
|
| 1527 |
+
"train_runtime": 181.2784,
|
| 1528 |
+
"train_tokens_per_second": 2479.943
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 12.456140350877194,
|
| 1532 |
+
"grad_norm": 0.048095703125,
|
| 1533 |
+
"learning_rate": 0.011273210744461479,
|
| 1534 |
+
"loss": 0.0786,
|
| 1535 |
+
"num_input_tokens_seen": 452824,
|
| 1536 |
+
"step": 710,
|
| 1537 |
+
"train_runtime": 182.4118,
|
| 1538 |
+
"train_tokens_per_second": 2482.428
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 12.543859649122806,
|
| 1542 |
+
"grad_norm": 0.07568359375,
|
| 1543 |
+
"learning_rate": 0.011051208456531014,
|
| 1544 |
+
"loss": 0.0965,
|
| 1545 |
+
"num_input_tokens_seen": 456120,
|
| 1546 |
+
"step": 715,
|
| 1547 |
+
"train_runtime": 183.5585,
|
| 1548 |
+
"train_tokens_per_second": 2484.876
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 12.631578947368421,
|
| 1552 |
+
"grad_norm": 0.26171875,
|
| 1553 |
+
"learning_rate": 0.010830131720459601,
|
| 1554 |
+
"loss": 0.2232,
|
| 1555 |
+
"num_input_tokens_seen": 459736,
|
| 1556 |
+
"step": 720,
|
| 1557 |
+
"train_runtime": 184.7433,
|
| 1558 |
+
"train_tokens_per_second": 2488.513
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 12.719298245614034,
|
| 1562 |
+
"grad_norm": 0.2333984375,
|
| 1563 |
+
"learning_rate": 0.010610032354121612,
|
| 1564 |
+
"loss": 0.1335,
|
| 1565 |
+
"num_input_tokens_seen": 462584,
|
| 1566 |
+
"step": 725,
|
| 1567 |
+
"train_runtime": 185.8146,
|
| 1568 |
+
"train_tokens_per_second": 2489.493
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 12.807017543859649,
|
| 1572 |
+
"grad_norm": 0.0361328125,
|
| 1573 |
+
"learning_rate": 0.01039096194630704,
|
| 1574 |
+
"loss": 0.081,
|
| 1575 |
+
"num_input_tokens_seen": 465688,
|
| 1576 |
+
"step": 730,
|
| 1577 |
+
"train_runtime": 186.9248,
|
| 1578 |
+
"train_tokens_per_second": 2491.312
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 12.894736842105264,
|
| 1582 |
+
"grad_norm": 0.21484375,
|
| 1583 |
+
"learning_rate": 0.010172971844629716,
|
| 1584 |
+
"loss": 0.1283,
|
| 1585 |
+
"num_input_tokens_seen": 468856,
|
| 1586 |
+
"step": 735,
|
| 1587 |
+
"train_runtime": 188.0635,
|
| 1588 |
+
"train_tokens_per_second": 2493.072
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 12.982456140350877,
|
| 1592 |
+
"grad_norm": 0.166015625,
|
| 1593 |
+
"learning_rate": 0.009956113143491957,
|
| 1594 |
+
"loss": 0.1343,
|
| 1595 |
+
"num_input_tokens_seen": 471768,
|
| 1596 |
+
"step": 740,
|
| 1597 |
+
"train_runtime": 189.1334,
|
| 1598 |
+
"train_tokens_per_second": 2494.366
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 13.0,
|
| 1602 |
+
"eval_loss": 0.14562781155109406,
|
| 1603 |
+
"eval_runtime": 0.8148,
|
| 1604 |
+
"eval_samples_per_second": 30.684,
|
| 1605 |
+
"eval_steps_per_second": 8.592,
|
| 1606 |
+
"num_input_tokens_seen": 471944,
|
| 1607 |
+
"step": 741
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 13.070175438596491,
|
| 1611 |
+
"grad_norm": 0.1376953125,
|
| 1612 |
+
"learning_rate": 0.009740436672108685,
|
| 1613 |
+
"loss": 0.3144,
|
| 1614 |
+
"num_input_tokens_seen": 474344,
|
| 1615 |
+
"step": 745,
|
| 1616 |
+
"train_runtime": 191.8522,
|
| 1617 |
+
"train_tokens_per_second": 2472.444
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 13.157894736842104,
|
| 1621 |
+
"grad_norm": 0.11669921875,
|
| 1622 |
+
"learning_rate": 0.009525992982593583,
|
| 1623 |
+
"loss": 0.1393,
|
| 1624 |
+
"num_input_tokens_seen": 478120,
|
| 1625 |
+
"step": 750,
|
| 1626 |
+
"train_runtime": 193.0855,
|
| 1627 |
+
"train_tokens_per_second": 2476.208
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 13.24561403508772,
|
| 1631 |
+
"grad_norm": 0.166015625,
|
| 1632 |
+
"learning_rate": 0.009312832338110292,
|
| 1633 |
+
"loss": 0.148,
|
| 1634 |
+
"num_input_tokens_seen": 480840,
|
| 1635 |
+
"step": 755,
|
| 1636 |
+
"train_runtime": 194.1509,
|
| 1637 |
+
"train_tokens_per_second": 2476.63
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 13.333333333333334,
|
| 1641 |
+
"grad_norm": 0.146484375,
|
| 1642 |
+
"learning_rate": 0.009101004701091252,
|
| 1643 |
+
"loss": 0.2076,
|
| 1644 |
+
"num_input_tokens_seen": 484552,
|
| 1645 |
+
"step": 760,
|
| 1646 |
+
"train_runtime": 195.3575,
|
| 1647 |
+
"train_tokens_per_second": 2480.334
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 13.421052631578947,
|
| 1651 |
+
"grad_norm": 0.0208740234375,
|
| 1652 |
+
"learning_rate": 0.008890559721527138,
|
| 1653 |
+
"loss": 0.0925,
|
| 1654 |
+
"num_input_tokens_seen": 487304,
|
| 1655 |
+
"step": 765,
|
| 1656 |
+
"train_runtime": 196.4146,
|
| 1657 |
+
"train_tokens_per_second": 2480.997
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 13.508771929824562,
|
| 1661 |
+
"grad_norm": 0.275390625,
|
| 1662 |
+
"learning_rate": 0.008681546725329408,
|
| 1663 |
+
"loss": 0.1007,
|
| 1664 |
+
"num_input_tokens_seen": 490536,
|
| 1665 |
+
"step": 770,
|
| 1666 |
+
"train_runtime": 197.5428,
|
| 1667 |
+
"train_tokens_per_second": 2483.189
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 13.596491228070175,
|
| 1671 |
+
"grad_norm": 0.146484375,
|
| 1672 |
+
"learning_rate": 0.008474014702768904,
|
| 1673 |
+
"loss": 0.124,
|
| 1674 |
+
"num_input_tokens_seen": 493128,
|
| 1675 |
+
"step": 775,
|
| 1676 |
+
"train_runtime": 198.5672,
|
| 1677 |
+
"train_tokens_per_second": 2483.431
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 13.68421052631579,
|
| 1681 |
+
"grad_norm": 0.08544921875,
|
| 1682 |
+
"learning_rate": 0.008268012296993067,
|
| 1683 |
+
"loss": 0.1165,
|
| 1684 |
+
"num_input_tokens_seen": 496136,
|
| 1685 |
+
"step": 780,
|
| 1686 |
+
"train_runtime": 199.6562,
|
| 1687 |
+
"train_tokens_per_second": 2484.951
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 13.771929824561404,
|
| 1691 |
+
"grad_norm": 0.1337890625,
|
| 1692 |
+
"learning_rate": 0.008063587792624567,
|
| 1693 |
+
"loss": 0.0905,
|
| 1694 |
+
"num_input_tokens_seen": 499432,
|
| 1695 |
+
"step": 785,
|
| 1696 |
+
"train_runtime": 200.7987,
|
| 1697 |
+
"train_tokens_per_second": 2487.227
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 13.859649122807017,
|
| 1701 |
+
"grad_norm": 0.076171875,
|
| 1702 |
+
"learning_rate": 0.007860789104443896,
|
| 1703 |
+
"loss": 0.1697,
|
| 1704 |
+
"num_input_tokens_seen": 502984,
|
| 1705 |
+
"step": 790,
|
| 1706 |
+
"train_runtime": 201.9799,
|
| 1707 |
+
"train_tokens_per_second": 2490.268
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 13.947368421052632,
|
| 1711 |
+
"grad_norm": 0.169921875,
|
| 1712 |
+
"learning_rate": 0.0076596637661587325,
|
| 1713 |
+
"loss": 0.1078,
|
| 1714 |
+
"num_input_tokens_seen": 506376,
|
| 1715 |
+
"step": 795,
|
| 1716 |
+
"train_runtime": 203.1371,
|
| 1717 |
+
"train_tokens_per_second": 2492.779
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 14.0,
|
| 1721 |
+
"eval_loss": 0.20142777264118195,
|
| 1722 |
+
"eval_runtime": 0.8139,
|
| 1723 |
+
"eval_samples_per_second": 30.716,
|
| 1724 |
+
"eval_steps_per_second": 8.601,
|
| 1725 |
+
"num_input_tokens_seen": 508424,
|
| 1726 |
+
"step": 798
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 14.035087719298245,
|
| 1730 |
+
"grad_norm": 0.150390625,
|
| 1731 |
+
"learning_rate": 0.007460258919262529,
|
| 1732 |
+
"loss": 0.1162,
|
| 1733 |
+
"num_input_tokens_seen": 509672,
|
| 1734 |
+
"step": 800,
|
| 1735 |
+
"train_runtime": 205.9194,
|
| 1736 |
+
"train_tokens_per_second": 2475.104
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 14.12280701754386,
|
| 1740 |
+
"grad_norm": 0.1572265625,
|
| 1741 |
+
"learning_rate": 0.007262621301985144,
|
| 1742 |
+
"loss": 0.1316,
|
| 1743 |
+
"num_input_tokens_seen": 513320,
|
| 1744 |
+
"step": 805,
|
| 1745 |
+
"train_runtime": 207.1437,
|
| 1746 |
+
"train_tokens_per_second": 2478.087
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 14.210526315789474,
|
| 1750 |
+
"grad_norm": 0.119140625,
|
| 1751 |
+
"learning_rate": 0.007066797238337862,
|
| 1752 |
+
"loss": 0.1569,
|
| 1753 |
+
"num_input_tokens_seen": 516296,
|
| 1754 |
+
"step": 810,
|
| 1755 |
+
"train_runtime": 208.232,
|
| 1756 |
+
"train_tokens_per_second": 2479.426
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 14.298245614035087,
|
| 1760 |
+
"grad_norm": 0.01104736328125,
|
| 1761 |
+
"learning_rate": 0.006872832627255643,
|
| 1762 |
+
"loss": 0.0317,
|
| 1763 |
+
"num_input_tokens_seen": 520040,
|
| 1764 |
+
"step": 815,
|
| 1765 |
+
"train_runtime": 209.4322,
|
| 1766 |
+
"train_tokens_per_second": 2483.095
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 14.385964912280702,
|
| 1770 |
+
"grad_norm": 0.181640625,
|
| 1771 |
+
"learning_rate": 0.006680772931838868,
|
| 1772 |
+
"loss": 0.0949,
|
| 1773 |
+
"num_input_tokens_seen": 523272,
|
| 1774 |
+
"step": 820,
|
| 1775 |
+
"train_runtime": 210.5613,
|
| 1776 |
+
"train_tokens_per_second": 2485.129
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 14.473684210526315,
|
| 1780 |
+
"grad_norm": 0.0556640625,
|
| 1781 |
+
"learning_rate": 0.00649066316869736,
|
| 1782 |
+
"loss": 0.1057,
|
| 1783 |
+
"num_input_tokens_seen": 527112,
|
| 1784 |
+
"step": 825,
|
| 1785 |
+
"train_runtime": 211.7745,
|
| 1786 |
+
"train_tokens_per_second": 2489.024
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 14.56140350877193,
|
| 1790 |
+
"grad_norm": 0.0654296875,
|
| 1791 |
+
"learning_rate": 0.0063025478973989585,
|
| 1792 |
+
"loss": 0.1103,
|
| 1793 |
+
"num_input_tokens_seen": 530568,
|
| 1794 |
+
"step": 830,
|
| 1795 |
+
"train_runtime": 212.9448,
|
| 1796 |
+
"train_tokens_per_second": 2491.575
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 14.649122807017545,
|
| 1800 |
+
"grad_norm": 0.2216796875,
|
| 1801 |
+
"learning_rate": 0.006116471210025301,
|
| 1802 |
+
"loss": 0.0813,
|
| 1803 |
+
"num_input_tokens_seen": 533192,
|
| 1804 |
+
"step": 835,
|
| 1805 |
+
"train_runtime": 213.9825,
|
| 1806 |
+
"train_tokens_per_second": 2491.755
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 14.736842105263158,
|
| 1810 |
+
"grad_norm": 0.10107421875,
|
| 1811 |
+
"learning_rate": 0.005932476720837105,
|
| 1812 |
+
"loss": 0.1104,
|
| 1813 |
+
"num_input_tokens_seen": 536232,
|
| 1814 |
+
"step": 840,
|
| 1815 |
+
"train_runtime": 215.073,
|
| 1816 |
+
"train_tokens_per_second": 2493.256
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 14.824561403508772,
|
| 1820 |
+
"grad_norm": 0.059814453125,
|
| 1821 |
+
"learning_rate": 0.005750607556051514,
|
| 1822 |
+
"loss": 0.0651,
|
| 1823 |
+
"num_input_tokens_seen": 539624,
|
| 1824 |
+
"step": 845,
|
| 1825 |
+
"train_runtime": 216.2313,
|
| 1826 |
+
"train_tokens_per_second": 2495.587
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 14.912280701754385,
|
| 1830 |
+
"grad_norm": 0.2109375,
|
| 1831 |
+
"learning_rate": 0.0055709063437337685,
|
| 1832 |
+
"loss": 0.0877,
|
| 1833 |
+
"num_input_tokens_seen": 542664,
|
| 1834 |
+
"step": 850,
|
| 1835 |
+
"train_runtime": 217.3342,
|
| 1836 |
+
"train_tokens_per_second": 2496.91
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 15.0,
|
| 1840 |
+
"grad_norm": 0.58984375,
|
| 1841 |
+
"learning_rate": 0.005393415203805707,
|
| 1842 |
+
"loss": 0.2187,
|
| 1843 |
+
"num_input_tokens_seen": 545352,
|
| 1844 |
+
"step": 855,
|
| 1845 |
+
"train_runtime": 218.3882,
|
| 1846 |
+
"train_tokens_per_second": 2497.168
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 15.0,
|
| 1850 |
+
"eval_loss": 0.13247625529766083,
|
| 1851 |
+
"eval_runtime": 0.8159,
|
| 1852 |
+
"eval_samples_per_second": 30.642,
|
| 1853 |
+
"eval_steps_per_second": 8.58,
|
| 1854 |
+
"num_input_tokens_seen": 545352,
|
| 1855 |
+
"step": 855
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 15.087719298245615,
|
| 1859 |
+
"grad_norm": 0.134765625,
|
| 1860 |
+
"learning_rate": 0.005218175738173303,
|
| 1861 |
+
"loss": 0.0707,
|
| 1862 |
+
"num_input_tokens_seen": 548488,
|
| 1863 |
+
"step": 860,
|
| 1864 |
+
"train_runtime": 221.1757,
|
| 1865 |
+
"train_tokens_per_second": 2479.875
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 15.175438596491228,
|
| 1869 |
+
"grad_norm": 0.091796875,
|
| 1870 |
+
"learning_rate": 0.005045229020975681,
|
| 1871 |
+
"loss": 0.0327,
|
| 1872 |
+
"num_input_tokens_seen": 552008,
|
| 1873 |
+
"step": 865,
|
| 1874 |
+
"train_runtime": 222.3601,
|
| 1875 |
+
"train_tokens_per_second": 2482.496
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 15.263157894736842,
|
| 1879 |
+
"grad_norm": 0.1767578125,
|
| 1880 |
+
"learning_rate": 0.004874615588957773,
|
| 1881 |
+
"loss": 0.1188,
|
| 1882 |
+
"num_input_tokens_seen": 555944,
|
| 1883 |
+
"step": 870,
|
| 1884 |
+
"train_runtime": 223.6053,
|
| 1885 |
+
"train_tokens_per_second": 2486.273
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 15.350877192982455,
|
| 1889 |
+
"grad_norm": 0.5234375,
|
| 1890 |
+
"learning_rate": 0.0047063754319689975,
|
| 1891 |
+
"loss": 0.0771,
|
| 1892 |
+
"num_input_tokens_seen": 558888,
|
| 1893 |
+
"step": 875,
|
| 1894 |
+
"train_runtime": 224.6953,
|
| 1895 |
+
"train_tokens_per_second": 2487.315
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 15.43859649122807,
|
| 1899 |
+
"grad_norm": 0.080078125,
|
| 1900 |
+
"learning_rate": 0.004540547983590069,
|
| 1901 |
+
"loss": 0.0614,
|
| 1902 |
+
"num_input_tokens_seen": 562536,
|
| 1903 |
+
"step": 880,
|
| 1904 |
+
"train_runtime": 225.8954,
|
| 1905 |
+
"train_tokens_per_second": 2490.25
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 15.526315789473685,
|
| 1909 |
+
"grad_norm": 0.11328125,
|
| 1910 |
+
"learning_rate": 0.0043771721118902335,
|
| 1911 |
+
"loss": 0.0345,
|
| 1912 |
+
"num_input_tokens_seen": 565160,
|
| 1913 |
+
"step": 885,
|
| 1914 |
+
"train_runtime": 226.923,
|
| 1915 |
+
"train_tokens_per_second": 2490.536
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 15.614035087719298,
|
| 1919 |
+
"grad_norm": 0.09912109375,
|
| 1920 |
+
"learning_rate": 0.004216286110317013,
|
| 1921 |
+
"loss": 0.0391,
|
| 1922 |
+
"num_input_tokens_seen": 568232,
|
| 1923 |
+
"step": 890,
|
| 1924 |
+
"train_runtime": 228.0426,
|
| 1925 |
+
"train_tokens_per_second": 2491.78
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 15.701754385964913,
|
| 1929 |
+
"grad_norm": 0.10595703125,
|
| 1930 |
+
"learning_rate": 0.00405792768872069,
|
| 1931 |
+
"loss": 0.115,
|
| 1932 |
+
"num_input_tokens_seen": 571048,
|
| 1933 |
+
"step": 895,
|
| 1934 |
+
"train_runtime": 229.1035,
|
| 1935 |
+
"train_tokens_per_second": 2492.533
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 15.789473684210526,
|
| 1939 |
+
"grad_norm": 0.310546875,
|
| 1940 |
+
"learning_rate": 0.003902133964515502,
|
| 1941 |
+
"loss": 0.0447,
|
| 1942 |
+
"num_input_tokens_seen": 574664,
|
| 1943 |
+
"step": 900,
|
| 1944 |
+
"train_runtime": 230.2946,
|
| 1945 |
+
"train_tokens_per_second": 2495.343
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 15.87719298245614,
|
| 1949 |
+
"grad_norm": 0.2001953125,
|
| 1950 |
+
"learning_rate": 0.00374894145397979,
|
| 1951 |
+
"loss": 0.0932,
|
| 1952 |
+
"num_input_tokens_seen": 577608,
|
| 1953 |
+
"step": 905,
|
| 1954 |
+
"train_runtime": 231.3778,
|
| 1955 |
+
"train_tokens_per_second": 2496.384
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 15.964912280701755,
|
| 1959 |
+
"grad_norm": 0.154296875,
|
| 1960 |
+
"learning_rate": 0.0035983860636969525,
|
| 1961 |
+
"loss": 0.0431,
|
| 1962 |
+
"num_input_tokens_seen": 580680,
|
| 1963 |
+
"step": 910,
|
| 1964 |
+
"train_runtime": 232.5024,
|
| 1965 |
+
"train_tokens_per_second": 2497.523
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 16.0,
|
| 1969 |
+
"eval_loss": 0.15890370309352875,
|
| 1970 |
+
"eval_runtime": 0.8161,
|
| 1971 |
+
"eval_samples_per_second": 30.634,
|
| 1972 |
+
"eval_steps_per_second": 8.577,
|
| 1973 |
+
"num_input_tokens_seen": 581368,
|
| 1974 |
+
"step": 912
|
| 1975 |
+
},
|
| 1976 |
+
{
|
| 1977 |
+
"epoch": 16.05263157894737,
|
| 1978 |
+
"grad_norm": 0.0142822265625,
|
| 1979 |
+
"learning_rate": 0.003450503082139393,
|
| 1980 |
+
"loss": 0.0722,
|
| 1981 |
+
"num_input_tokens_seen": 583544,
|
| 1982 |
+
"step": 915,
|
| 1983 |
+
"train_runtime": 235.2462,
|
| 1984 |
+
"train_tokens_per_second": 2480.568
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 16.140350877192983,
|
| 1988 |
+
"grad_norm": 0.11962890625,
|
| 1989 |
+
"learning_rate": 0.003305327171397263,
|
| 1990 |
+
"loss": 0.0529,
|
| 1991 |
+
"num_input_tokens_seen": 586424,
|
| 1992 |
+
"step": 920,
|
| 1993 |
+
"train_runtime": 236.3539,
|
| 1994 |
+
"train_tokens_per_second": 2481.127
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 16.228070175438596,
|
| 1998 |
+
"grad_norm": 0.044189453125,
|
| 1999 |
+
"learning_rate": 0.003162892359054098,
|
| 2000 |
+
"loss": 0.0497,
|
| 2001 |
+
"num_input_tokens_seen": 589976,
|
| 2002 |
+
"step": 925,
|
| 2003 |
+
"train_runtime": 237.5448,
|
| 2004 |
+
"train_tokens_per_second": 2483.641
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 16.31578947368421,
|
| 2008 |
+
"grad_norm": 0.03076171875,
|
| 2009 |
+
"learning_rate": 0.0030232320302111047,
|
| 2010 |
+
"loss": 0.0336,
|
| 2011 |
+
"num_input_tokens_seen": 593048,
|
| 2012 |
+
"step": 930,
|
| 2013 |
+
"train_runtime": 238.6634,
|
| 2014 |
+
"train_tokens_per_second": 2484.872
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 16.403508771929825,
|
| 2018 |
+
"grad_norm": 0.06201171875,
|
| 2019 |
+
"learning_rate": 0.0028863789196621093,
|
| 2020 |
+
"loss": 0.0242,
|
| 2021 |
+
"num_input_tokens_seen": 596472,
|
| 2022 |
+
"step": 935,
|
| 2023 |
+
"train_runtime": 239.8271,
|
| 2024 |
+
"train_tokens_per_second": 2487.092
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 16.49122807017544,
|
| 2028 |
+
"grad_norm": 0.10595703125,
|
| 2029 |
+
"learning_rate": 0.0027523651042208564,
|
| 2030 |
+
"loss": 0.0565,
|
| 2031 |
+
"num_input_tokens_seen": 599544,
|
| 2032 |
+
"step": 940,
|
| 2033 |
+
"train_runtime": 240.9279,
|
| 2034 |
+
"train_tokens_per_second": 2488.479
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 16.57894736842105,
|
| 2038 |
+
"grad_norm": 0.07421875,
|
| 2039 |
+
"learning_rate": 0.0026212219952026,
|
| 2040 |
+
"loss": 0.014,
|
| 2041 |
+
"num_input_tokens_seen": 603032,
|
| 2042 |
+
"step": 945,
|
| 2043 |
+
"train_runtime": 242.1006,
|
| 2044 |
+
"train_tokens_per_second": 2490.832
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 16.666666666666668,
|
| 2048 |
+
"grad_norm": 0.0400390625,
|
| 2049 |
+
"learning_rate": 0.0024929803310616222,
|
| 2050 |
+
"loss": 0.0114,
|
| 2051 |
+
"num_input_tokens_seen": 605560,
|
| 2052 |
+
"step": 950,
|
| 2053 |
+
"train_runtime": 243.1261,
|
| 2054 |
+
"train_tokens_per_second": 2490.724
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 16.75438596491228,
|
| 2058 |
+
"grad_norm": 0.010498046875,
|
| 2059 |
+
"learning_rate": 0.002367670170186516,
|
| 2060 |
+
"loss": 0.0465,
|
| 2061 |
+
"num_input_tokens_seen": 609176,
|
| 2062 |
+
"step": 955,
|
| 2063 |
+
"train_runtime": 244.3106,
|
| 2064 |
+
"train_tokens_per_second": 2493.449
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 16.842105263157894,
|
| 2068 |
+
"grad_norm": 0.07568359375,
|
| 2069 |
+
"learning_rate": 0.00224532088385481,
|
| 2070 |
+
"loss": 0.033,
|
| 2071 |
+
"num_input_tokens_seen": 612024,
|
| 2072 |
+
"step": 960,
|
| 2073 |
+
"train_runtime": 245.3849,
|
| 2074 |
+
"train_tokens_per_second": 2494.138
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 16.92982456140351,
|
| 2078 |
+
"grad_norm": 0.1826171875,
|
| 2079 |
+
"learning_rate": 0.002125961149348706,
|
| 2080 |
+
"loss": 0.0308,
|
| 2081 |
+
"num_input_tokens_seen": 615064,
|
| 2082 |
+
"step": 965,
|
| 2083 |
+
"train_runtime": 246.4914,
|
| 2084 |
+
"train_tokens_per_second": 2495.275
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 17.0,
|
| 2088 |
+
"eval_loss": 0.1602199375629425,
|
| 2089 |
+
"eval_runtime": 0.8142,
|
| 2090 |
+
"eval_samples_per_second": 30.706,
|
| 2091 |
+
"eval_steps_per_second": 8.598,
|
| 2092 |
+
"num_input_tokens_seen": 616776,
|
| 2093 |
+
"step": 969
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 17.017543859649123,
|
| 2097 |
+
"grad_norm": 0.095703125,
|
| 2098 |
+
"learning_rate": 0.0020096189432334192,
|
| 2099 |
+
"loss": 0.0364,
|
| 2100 |
+
"num_input_tokens_seen": 617640,
|
| 2101 |
+
"step": 970,
|
| 2102 |
+
"train_runtime": 249.6714,
|
| 2103 |
+
"train_tokens_per_second": 2473.812
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 17.105263157894736,
|
| 2107 |
+
"grad_norm": 0.042236328125,
|
| 2108 |
+
"learning_rate": 0.001896321534799823,
|
| 2109 |
+
"loss": 0.0222,
|
| 2110 |
+
"num_input_tokens_seen": 620424,
|
| 2111 |
+
"step": 975,
|
| 2112 |
+
"train_runtime": 250.7368,
|
| 2113 |
+
"train_tokens_per_second": 2474.403
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 17.19298245614035,
|
| 2117 |
+
"grad_norm": 0.06298828125,
|
| 2118 |
+
"learning_rate": 0.0017860954796727994,
|
| 2119 |
+
"loss": 0.0159,
|
| 2120 |
+
"num_input_tokens_seen": 623176,
|
| 2121 |
+
"step": 980,
|
| 2122 |
+
"train_runtime": 251.7883,
|
| 2123 |
+
"train_tokens_per_second": 2475.0
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 17.280701754385966,
|
| 2127 |
+
"grad_norm": 0.0283203125,
|
| 2128 |
+
"learning_rate": 0.0016789666135869374,
|
| 2129 |
+
"loss": 0.0073,
|
| 2130 |
+
"num_input_tokens_seen": 626600,
|
| 2131 |
+
"step": 985,
|
| 2132 |
+
"train_runtime": 252.9495,
|
| 2133 |
+
"train_tokens_per_second": 2477.174
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 17.36842105263158,
|
| 2137 |
+
"grad_norm": 0.119140625,
|
| 2138 |
+
"learning_rate": 0.0015749600463309049,
|
| 2139 |
+
"loss": 0.0164,
|
| 2140 |
+
"num_input_tokens_seen": 629992,
|
| 2141 |
+
"step": 990,
|
| 2142 |
+
"train_runtime": 254.1075,
|
| 2143 |
+
"train_tokens_per_second": 2479.234
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 17.45614035087719,
|
| 2147 |
+
"grad_norm": 0.0654296875,
|
| 2148 |
+
"learning_rate": 0.0014741001558620163,
|
| 2149 |
+
"loss": 0.0192,
|
| 2150 |
+
"num_input_tokens_seen": 633128,
|
| 2151 |
+
"step": 995,
|
| 2152 |
+
"train_runtime": 255.2377,
|
| 2153 |
+
"train_tokens_per_second": 2480.542
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 17.54385964912281,
|
| 2157 |
+
"grad_norm": 0.043701171875,
|
| 2158 |
+
"learning_rate": 0.0013764105825923066,
|
| 2159 |
+
"loss": 0.0282,
|
| 2160 |
+
"num_input_tokens_seen": 635880,
|
| 2161 |
+
"step": 1000,
|
| 2162 |
+
"train_runtime": 256.2989,
|
| 2163 |
+
"train_tokens_per_second": 2481.009
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 17.63157894736842,
|
| 2167 |
+
"grad_norm": 0.043701171875,
|
| 2168 |
+
"learning_rate": 0.0012819142238474862,
|
| 2169 |
+
"loss": 0.0314,
|
| 2170 |
+
"num_input_tokens_seen": 639208,
|
| 2171 |
+
"step": 1005,
|
| 2172 |
+
"train_runtime": 257.44,
|
| 2173 |
+
"train_tokens_per_second": 2482.939
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 17.719298245614034,
|
| 2177 |
+
"grad_norm": 0.0036163330078125,
|
| 2178 |
+
"learning_rate": 0.0011906332285000793,
|
| 2179 |
+
"loss": 0.0184,
|
| 2180 |
+
"num_input_tokens_seen": 642152,
|
| 2181 |
+
"step": 1010,
|
| 2182 |
+
"train_runtime": 258.5369,
|
| 2183 |
+
"train_tokens_per_second": 2483.793
|
| 2184 |
+
},
|
| 2185 |
+
{
|
| 2186 |
+
"epoch": 17.80701754385965,
|
| 2187 |
+
"grad_norm": 0.053955078125,
|
| 2188 |
+
"learning_rate": 0.0011025889917779736,
|
| 2189 |
+
"loss": 0.0421,
|
| 2190 |
+
"num_input_tokens_seen": 645800,
|
| 2191 |
+
"step": 1015,
|
| 2192 |
+
"train_runtime": 259.7348,
|
| 2193 |
+
"train_tokens_per_second": 2486.382
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 17.894736842105264,
|
| 2197 |
+
"grad_norm": 0.203125,
|
| 2198 |
+
"learning_rate": 0.0010178021502496165,
|
| 2199 |
+
"loss": 0.0375,
|
| 2200 |
+
"num_input_tokens_seen": 649736,
|
| 2201 |
+
"step": 1020,
|
| 2202 |
+
"train_runtime": 260.9682,
|
| 2203 |
+
"train_tokens_per_second": 2489.713
|
| 2204 |
+
},
|
| 2205 |
+
{
|
| 2206 |
+
"epoch": 17.982456140350877,
|
| 2207 |
+
"grad_norm": 0.18359375,
|
| 2208 |
+
"learning_rate": 0.0009362925769870394,
|
| 2209 |
+
"loss": 0.0234,
|
| 2210 |
+
"num_input_tokens_seen": 653064,
|
| 2211 |
+
"step": 1025,
|
| 2212 |
+
"train_runtime": 262.1226,
|
| 2213 |
+
"train_tokens_per_second": 2491.444
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 18.0,
|
| 2217 |
+
"eval_loss": 0.167332723736763,
|
| 2218 |
+
"eval_runtime": 0.8132,
|
| 2219 |
+
"eval_samples_per_second": 30.744,
|
| 2220 |
+
"eval_steps_per_second": 8.608,
|
| 2221 |
+
"num_input_tokens_seen": 653152,
|
| 2222 |
+
"step": 1026
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 18.07017543859649,
|
| 2226 |
+
"grad_norm": 0.1083984375,
|
| 2227 |
+
"learning_rate": 0.0008580793769078487,
|
| 2228 |
+
"loss": 0.0297,
|
| 2229 |
+
"num_input_tokens_seen": 655424,
|
| 2230 |
+
"step": 1030,
|
| 2231 |
+
"train_runtime": 264.7587,
|
| 2232 |
+
"train_tokens_per_second": 2475.553
|
| 2233 |
+
},
|
| 2234 |
+
{
|
| 2235 |
+
"epoch": 18.157894736842106,
|
| 2236 |
+
"grad_norm": 0.037841796875,
|
| 2237 |
+
"learning_rate": 0.0007831808822972391,
|
| 2238 |
+
"loss": 0.0349,
|
| 2239 |
+
"num_input_tokens_seen": 658848,
|
| 2240 |
+
"step": 1035,
|
| 2241 |
+
"train_runtime": 265.9884,
|
| 2242 |
+
"train_tokens_per_second": 2476.98
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 18.24561403508772,
|
| 2246 |
+
"grad_norm": 0.05517578125,
|
| 2247 |
+
"learning_rate": 0.0007116146485111063,
|
| 2248 |
+
"loss": 0.0087,
|
| 2249 |
+
"num_input_tokens_seen": 661888,
|
| 2250 |
+
"step": 1040,
|
| 2251 |
+
"train_runtime": 267.0809,
|
| 2252 |
+
"train_tokens_per_second": 2478.23
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 18.333333333333332,
|
| 2256 |
+
"grad_norm": 0.0576171875,
|
| 2257 |
+
"learning_rate": 0.0006433974498612882,
|
| 2258 |
+
"loss": 0.0125,
|
| 2259 |
+
"num_input_tokens_seen": 665664,
|
| 2260 |
+
"step": 1045,
|
| 2261 |
+
"train_runtime": 268.3008,
|
| 2262 |
+
"train_tokens_per_second": 2481.037
|
| 2263 |
+
},
|
| 2264 |
+
{
|
| 2265 |
+
"epoch": 18.42105263157895,
|
| 2266 |
+
"grad_norm": 0.087890625,
|
| 2267 |
+
"learning_rate": 0.0005785452756838482,
|
| 2268 |
+
"loss": 0.0184,
|
| 2269 |
+
"num_input_tokens_seen": 668896,
|
| 2270 |
+
"step": 1050,
|
| 2271 |
+
"train_runtime": 269.4151,
|
| 2272 |
+
"train_tokens_per_second": 2482.771
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 18.50877192982456,
|
| 2276 |
+
"grad_norm": 0.004119873046875,
|
| 2277 |
+
"learning_rate": 0.0005170733265913585,
|
| 2278 |
+
"loss": 0.0122,
|
| 2279 |
+
"num_input_tokens_seen": 672352,
|
| 2280 |
+
"step": 1055,
|
| 2281 |
+
"train_runtime": 270.5796,
|
| 2282 |
+
"train_tokens_per_second": 2484.858
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 18.596491228070175,
|
| 2286 |
+
"grad_norm": 0.146484375,
|
| 2287 |
+
"learning_rate": 0.0004589960109100444,
|
| 2288 |
+
"loss": 0.0324,
|
| 2289 |
+
"num_input_tokens_seen": 675392,
|
| 2290 |
+
"step": 1060,
|
| 2291 |
+
"train_runtime": 271.6711,
|
| 2292 |
+
"train_tokens_per_second": 2486.065
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 18.68421052631579,
|
| 2296 |
+
"grad_norm": 0.04931640625,
|
| 2297 |
+
"learning_rate": 0.00040432694130264286,
|
| 2298 |
+
"loss": 0.0231,
|
| 2299 |
+
"num_input_tokens_seen": 678080,
|
| 2300 |
+
"step": 1065,
|
| 2301 |
+
"train_runtime": 272.7179,
|
| 2302 |
+
"train_tokens_per_second": 2486.379
|
| 2303 |
+
},
|
| 2304 |
+
{
|
| 2305 |
+
"epoch": 18.771929824561404,
|
| 2306 |
+
"grad_norm": 0.04248046875,
|
| 2307 |
+
"learning_rate": 0.0003530789315777466,
|
| 2308 |
+
"loss": 0.0089,
|
| 2309 |
+
"num_input_tokens_seen": 681440,
|
| 2310 |
+
"step": 1070,
|
| 2311 |
+
"train_runtime": 273.8742,
|
| 2312 |
+
"train_tokens_per_second": 2488.149
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 18.859649122807017,
|
| 2316 |
+
"grad_norm": 0.07763671875,
|
| 2317 |
+
"learning_rate": 0.000305263993686391,
|
| 2318 |
+
"loss": 0.0199,
|
| 2319 |
+
"num_input_tokens_seen": 684608,
|
| 2320 |
+
"step": 1075,
|
| 2321 |
+
"train_runtime": 274.9942,
|
| 2322 |
+
"train_tokens_per_second": 2489.536
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 18.94736842105263,
|
| 2326 |
+
"grad_norm": 0.0245361328125,
|
| 2327 |
+
"learning_rate": 0.000260893334906595,
|
| 2328 |
+
"loss": 0.0289,
|
| 2329 |
+
"num_input_tokens_seen": 688480,
|
| 2330 |
+
"step": 1080,
|
| 2331 |
+
"train_runtime": 276.2315,
|
| 2332 |
+
"train_tokens_per_second": 2492.402
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 19.0,
|
| 2336 |
+
"eval_loss": 0.17038847506046295,
|
| 2337 |
+
"eval_runtime": 0.8147,
|
| 2338 |
+
"eval_samples_per_second": 30.687,
|
| 2339 |
+
"eval_steps_per_second": 8.592,
|
| 2340 |
+
"num_input_tokens_seen": 689856,
|
| 2341 |
+
"step": 1083
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 19.035087719298247,
|
| 2345 |
+
"grad_norm": 0.1513671875,
|
| 2346 |
+
"learning_rate": 0.00021997735521649408,
|
| 2347 |
+
"loss": 0.0351,
|
| 2348 |
+
"num_input_tokens_seen": 691200,
|
| 2349 |
+
"step": 1085,
|
| 2350 |
+
"train_runtime": 278.924,
|
| 2351 |
+
"train_tokens_per_second": 2478.094
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"epoch": 19.12280701754386,
|
| 2355 |
+
"grad_norm": 0.039306640625,
|
| 2356 |
+
"learning_rate": 0.00018252564485670973,
|
| 2357 |
+
"loss": 0.0203,
|
| 2358 |
+
"num_input_tokens_seen": 694240,
|
| 2359 |
+
"step": 1090,
|
| 2360 |
+
"train_runtime": 280.0551,
|
| 2361 |
+
"train_tokens_per_second": 2478.941
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 19.210526315789473,
|
| 2365 |
+
"grad_norm": 0.052734375,
|
| 2366 |
+
"learning_rate": 0.00014854698208250638,
|
| 2367 |
+
"loss": 0.0152,
|
| 2368 |
+
"num_input_tokens_seen": 697472,
|
| 2369 |
+
"step": 1095,
|
| 2370 |
+
"train_runtime": 281.1954,
|
| 2371 |
+
"train_tokens_per_second": 2480.382
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 19.29824561403509,
|
| 2375 |
+
"grad_norm": 0.01318359375,
|
| 2376 |
+
"learning_rate": 0.00011804933110626359,
|
| 2377 |
+
"loss": 0.0162,
|
| 2378 |
+
"num_input_tokens_seen": 699872,
|
| 2379 |
+
"step": 1100,
|
| 2380 |
+
"train_runtime": 282.2078,
|
| 2381 |
+
"train_tokens_per_second": 2479.988
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 19.385964912280702,
|
| 2385 |
+
"grad_norm": 0.1689453125,
|
| 2386 |
+
"learning_rate": 9.103984023075773e-05,
|
| 2387 |
+
"loss": 0.0398,
|
| 2388 |
+
"num_input_tokens_seen": 704736,
|
| 2389 |
+
"step": 1105,
|
| 2390 |
+
"train_runtime": 283.6238,
|
| 2391 |
+
"train_tokens_per_second": 2484.756
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 19.473684210526315,
|
| 2395 |
+
"grad_norm": 0.017822265625,
|
| 2396 |
+
"learning_rate": 6.752484017368553e-05,
|
| 2397 |
+
"loss": 0.0059,
|
| 2398 |
+
"num_input_tokens_seen": 707968,
|
| 2399 |
+
"step": 1110,
|
| 2400 |
+
"train_runtime": 284.7645,
|
| 2401 |
+
"train_tokens_per_second": 2486.153
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 19.56140350877193,
|
| 2405 |
+
"grad_norm": 0.12353515625,
|
| 2406 |
+
"learning_rate": 4.750984258380608e-05,
|
| 2407 |
+
"loss": 0.0162,
|
| 2408 |
+
"num_input_tokens_seen": 711136,
|
| 2409 |
+
"step": 1115,
|
| 2410 |
+
"train_runtime": 285.8868,
|
| 2411 |
+
"train_tokens_per_second": 2487.474
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 19.649122807017545,
|
| 2415 |
+
"grad_norm": 0.038330078125,
|
| 2416 |
+
"learning_rate": 3.099953874908079e-05,
|
| 2417 |
+
"loss": 0.0065,
|
| 2418 |
+
"num_input_tokens_seen": 714432,
|
| 2419 |
+
"step": 1120,
|
| 2420 |
+
"train_runtime": 287.029,
|
| 2421 |
+
"train_tokens_per_second": 2489.059
|
| 2422 |
+
},
|
| 2423 |
+
{
|
| 2424 |
+
"epoch": 19.736842105263158,
|
| 2425 |
+
"grad_norm": 0.0380859375,
|
| 2426 |
+
"learning_rate": 1.7997798497084714e-05,
|
| 2427 |
+
"loss": 0.0208,
|
| 2428 |
+
"num_input_tokens_seen": 717440,
|
| 2429 |
+
"step": 1125,
|
| 2430 |
+
"train_runtime": 288.131,
|
| 2431 |
+
"train_tokens_per_second": 2489.978
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 19.82456140350877,
|
| 2435 |
+
"grad_norm": 0.10791015625,
|
| 2436 |
+
"learning_rate": 8.50766928796709e-06,
|
| 2437 |
+
"loss": 0.035,
|
| 2438 |
+
"num_input_tokens_seen": 720480,
|
| 2439 |
+
"step": 1130,
|
| 2440 |
+
"train_runtime": 289.2249,
|
| 2441 |
+
"train_tokens_per_second": 2491.072
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 19.912280701754387,
|
| 2445 |
+
"grad_norm": 0.00958251953125,
|
| 2446 |
+
"learning_rate": 2.5313755001593604e-06,
|
| 2447 |
+
"loss": 0.0293,
|
| 2448 |
+
"num_input_tokens_seen": 723552,
|
| 2449 |
+
"step": 1135,
|
| 2450 |
+
"train_runtime": 290.3197,
|
| 2451 |
+
"train_tokens_per_second": 2492.259
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 20.0,
|
| 2455 |
+
"grad_norm": 0.052978515625,
|
| 2456 |
+
"learning_rate": 7.03179089989181e-08,
|
| 2457 |
+
"loss": 0.0121,
|
| 2458 |
+
"num_input_tokens_seen": 725992,
|
| 2459 |
+
"step": 1140,
|
| 2460 |
+
"train_runtime": 291.314,
|
| 2461 |
+
"train_tokens_per_second": 2492.129
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 20.0,
|
| 2465 |
+
"eval_loss": 0.16927888989448547,
|
| 2466 |
+
"eval_runtime": 0.811,
|
| 2467 |
+
"eval_samples_per_second": 30.825,
|
| 2468 |
+
"eval_steps_per_second": 8.631,
|
| 2469 |
+
"num_input_tokens_seen": 725992,
|
| 2470 |
+
"step": 1140
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 20.0,
|
| 2474 |
+
"num_input_tokens_seen": 725992,
|
| 2475 |
+
"step": 1140,
|
| 2476 |
+
"total_flos": 3.2691091634847744e+16,
|
| 2477 |
+
"train_loss": 0.2356332610214227,
|
| 2478 |
+
"train_runtime": 292.8644,
|
| 2479 |
+
"train_samples_per_second": 15.365,
|
| 2480 |
+
"train_steps_per_second": 3.893
|
| 2481 |
+
}
|
| 2482 |
+
],
|
| 2483 |
+
"logging_steps": 5,
|
| 2484 |
+
"max_steps": 1140,
|
| 2485 |
+
"num_input_tokens_seen": 725992,
|
| 2486 |
+
"num_train_epochs": 20,
|
| 2487 |
+
"save_steps": 57,
|
| 2488 |
+
"stateful_callbacks": {
|
| 2489 |
+
"TrainerControl": {
|
| 2490 |
+
"args": {
|
| 2491 |
+
"should_epoch_stop": false,
|
| 2492 |
+
"should_evaluate": false,
|
| 2493 |
+
"should_log": false,
|
| 2494 |
+
"should_save": true,
|
| 2495 |
+
"should_training_stop": true
|
| 2496 |
+
},
|
| 2497 |
+
"attributes": {}
|
| 2498 |
+
}
|
| 2499 |
+
},
|
| 2500 |
+
"total_flos": 3.2691091634847744e+16,
|
| 2501 |
+
"train_batch_size": 4,
|
| 2502 |
+
"trial_name": null,
|
| 2503 |
+
"trial_params": null
|
| 2504 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|