End of training
Browse files- README.md +1 -1
- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +1425 -11
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# qwen2.5-32b-openalex
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) on
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 16 |
|
| 17 |
# qwen2.5-32b-openalex
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) on the openalex dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9991031390134529,
|
| 3 |
+
"total_flos": 7841700554735616.0,
|
| 4 |
+
"train_loss": 0.4275342236729456,
|
| 5 |
+
"train_runtime": 63728.3403,
|
| 6 |
+
"train_samples_per_second": 2.239,
|
| 7 |
+
"train_steps_per_second": 0.009
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.9991031390134529,
|
| 3 |
+
"total_flos": 7841700554735616.0,
|
| 4 |
+
"train_loss": 0.4275342236729456,
|
| 5 |
+
"train_runtime": 63728.3403,
|
| 6 |
+
"train_samples_per_second": 2.239,
|
| 7 |
+
"train_steps_per_second": 0.009
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -2494,17 +2494,1431 @@
|
|
| 2494 |
"step": 355
|
| 2495 |
},
|
| 2496 |
{
|
| 2497 |
-
"epoch": 0.
|
| 2498 |
-
"
|
| 2499 |
-
"
|
| 2500 |
-
"
|
| 2501 |
-
"
|
| 2502 |
-
|
| 2503 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2504 |
}
|
| 2505 |
],
|
| 2506 |
"logging_steps": 1,
|
| 2507 |
-
"max_steps":
|
| 2508 |
"num_input_tokens_seen": 0,
|
| 2509 |
"num_train_epochs": 1,
|
| 2510 |
"save_steps": 100,
|
|
@@ -2520,7 +3934,7 @@
|
|
| 2520 |
"attributes": {}
|
| 2521 |
}
|
| 2522 |
},
|
| 2523 |
-
"total_flos":
|
| 2524 |
"train_batch_size": 2,
|
| 2525 |
"trial_name": null,
|
| 2526 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9991031390134529,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 557,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 2494 |
"step": 355
|
| 2495 |
},
|
| 2496 |
{
|
| 2497 |
+
"epoch": 0.6385650224215247,
|
| 2498 |
+
"grad_norm": 0.010235507041215897,
|
| 2499 |
+
"learning_rate": 2e-05,
|
| 2500 |
+
"loss": 1.5054,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.6403587443946188,
|
| 2505 |
+
"grad_norm": 0.010560178197920322,
|
| 2506 |
+
"learning_rate": 2e-05,
|
| 2507 |
+
"loss": 1.5118,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.6421524663677131,
|
| 2512 |
+
"grad_norm": 0.010353959165513515,
|
| 2513 |
+
"learning_rate": 2e-05,
|
| 2514 |
+
"loss": 1.5226,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.6439461883408072,
|
| 2519 |
+
"grad_norm": 0.010382940992712975,
|
| 2520 |
+
"learning_rate": 2e-05,
|
| 2521 |
+
"loss": 1.5078,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.6457399103139013,
|
| 2526 |
+
"grad_norm": 0.009856803342700005,
|
| 2527 |
+
"learning_rate": 2e-05,
|
| 2528 |
+
"loss": 1.5167,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.6475336322869956,
|
| 2533 |
+
"grad_norm": 0.010195410810410976,
|
| 2534 |
+
"learning_rate": 2e-05,
|
| 2535 |
+
"loss": 1.5142,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.6493273542600897,
|
| 2540 |
+
"grad_norm": 0.010302864946424961,
|
| 2541 |
+
"learning_rate": 2e-05,
|
| 2542 |
+
"loss": 1.5136,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.6511210762331838,
|
| 2547 |
+
"grad_norm": 0.010046405717730522,
|
| 2548 |
+
"learning_rate": 2e-05,
|
| 2549 |
+
"loss": 1.5112,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.6529147982062781,
|
| 2554 |
+
"grad_norm": 0.010849208571016788,
|
| 2555 |
+
"learning_rate": 2e-05,
|
| 2556 |
+
"loss": 1.5114,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.6547085201793722,
|
| 2561 |
+
"grad_norm": 0.010421674698591232,
|
| 2562 |
+
"learning_rate": 2e-05,
|
| 2563 |
+
"loss": 1.5173,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.6565022421524663,
|
| 2568 |
+
"grad_norm": 0.00989589188247919,
|
| 2569 |
+
"learning_rate": 2e-05,
|
| 2570 |
+
"loss": 1.5063,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.6582959641255606,
|
| 2575 |
+
"grad_norm": 0.010465629398822784,
|
| 2576 |
+
"learning_rate": 2e-05,
|
| 2577 |
+
"loss": 1.5031,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.6600896860986547,
|
| 2582 |
+
"grad_norm": 0.009964341297745705,
|
| 2583 |
+
"learning_rate": 2e-05,
|
| 2584 |
+
"loss": 1.5207,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.6618834080717488,
|
| 2589 |
+
"grad_norm": 0.01189314667135477,
|
| 2590 |
+
"learning_rate": 2e-05,
|
| 2591 |
+
"loss": 1.5361,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.6636771300448431,
|
| 2596 |
+
"grad_norm": 0.01012677513062954,
|
| 2597 |
+
"learning_rate": 2e-05,
|
| 2598 |
+
"loss": 1.5215,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.6654708520179372,
|
| 2603 |
+
"grad_norm": 0.009877102449536324,
|
| 2604 |
+
"learning_rate": 2e-05,
|
| 2605 |
+
"loss": 1.5262,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.6672645739910313,
|
| 2610 |
+
"grad_norm": 0.01000463031232357,
|
| 2611 |
+
"learning_rate": 2e-05,
|
| 2612 |
+
"loss": 1.5183,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.6690582959641256,
|
| 2617 |
+
"grad_norm": 0.010188892483711243,
|
| 2618 |
+
"learning_rate": 2e-05,
|
| 2619 |
+
"loss": 1.5183,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.6708520179372197,
|
| 2624 |
+
"grad_norm": 0.010129815898835659,
|
| 2625 |
+
"learning_rate": 2e-05,
|
| 2626 |
+
"loss": 1.5245,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.672645739910314,
|
| 2631 |
+
"grad_norm": 0.010608335956931114,
|
| 2632 |
+
"learning_rate": 2e-05,
|
| 2633 |
+
"loss": 1.5169,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.6744394618834081,
|
| 2638 |
+
"grad_norm": 0.010223207995295525,
|
| 2639 |
+
"learning_rate": 2e-05,
|
| 2640 |
+
"loss": 1.5185,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.6762331838565022,
|
| 2645 |
+
"grad_norm": 0.010141369886696339,
|
| 2646 |
+
"learning_rate": 2e-05,
|
| 2647 |
+
"loss": 1.5161,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.6780269058295965,
|
| 2652 |
+
"grad_norm": 0.01027351152151823,
|
| 2653 |
+
"learning_rate": 2e-05,
|
| 2654 |
+
"loss": 1.5119,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.6798206278026906,
|
| 2659 |
+
"grad_norm": 0.010362266562879086,
|
| 2660 |
+
"learning_rate": 2e-05,
|
| 2661 |
+
"loss": 1.5126,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.6816143497757847,
|
| 2666 |
+
"grad_norm": 0.010336722247302532,
|
| 2667 |
+
"learning_rate": 2e-05,
|
| 2668 |
+
"loss": 1.5173,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.683408071748879,
|
| 2673 |
+
"grad_norm": 0.01007298193871975,
|
| 2674 |
+
"learning_rate": 2e-05,
|
| 2675 |
+
"loss": 1.5075,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.6852017937219731,
|
| 2680 |
+
"grad_norm": 0.010275410488247871,
|
| 2681 |
+
"learning_rate": 2e-05,
|
| 2682 |
+
"loss": 1.5123,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.6869955156950672,
|
| 2687 |
+
"grad_norm": 0.010203160345554352,
|
| 2688 |
+
"learning_rate": 2e-05,
|
| 2689 |
+
"loss": 1.5151,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.6887892376681615,
|
| 2694 |
+
"grad_norm": 0.010127630084753036,
|
| 2695 |
+
"learning_rate": 2e-05,
|
| 2696 |
+
"loss": 1.5211,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.6905829596412556,
|
| 2701 |
+
"grad_norm": 0.009799284860491753,
|
| 2702 |
+
"learning_rate": 2e-05,
|
| 2703 |
+
"loss": 1.5191,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.6923766816143497,
|
| 2708 |
+
"grad_norm": 0.01014394499361515,
|
| 2709 |
+
"learning_rate": 2e-05,
|
| 2710 |
+
"loss": 1.5261,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.694170403587444,
|
| 2715 |
+
"grad_norm": 0.010567774064838886,
|
| 2716 |
+
"learning_rate": 2e-05,
|
| 2717 |
+
"loss": 1.5232,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.6959641255605381,
|
| 2722 |
+
"grad_norm": 0.010051852092146873,
|
| 2723 |
+
"learning_rate": 2e-05,
|
| 2724 |
+
"loss": 1.5212,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.6977578475336322,
|
| 2729 |
+
"grad_norm": 0.010241293348371983,
|
| 2730 |
+
"learning_rate": 2e-05,
|
| 2731 |
+
"loss": 1.5094,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.6995515695067265,
|
| 2736 |
+
"grad_norm": 0.0095717404037714,
|
| 2737 |
+
"learning_rate": 2e-05,
|
| 2738 |
+
"loss": 1.5115,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.7013452914798206,
|
| 2743 |
+
"grad_norm": 0.00974031537771225,
|
| 2744 |
+
"learning_rate": 2e-05,
|
| 2745 |
+
"loss": 1.5195,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.7031390134529149,
|
| 2750 |
+
"grad_norm": 0.010140657424926758,
|
| 2751 |
+
"learning_rate": 2e-05,
|
| 2752 |
+
"loss": 1.5048,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.704932735426009,
|
| 2757 |
+
"grad_norm": 0.010055477730929852,
|
| 2758 |
+
"learning_rate": 2e-05,
|
| 2759 |
+
"loss": 1.5162,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.7067264573991031,
|
| 2764 |
+
"grad_norm": 0.01005468424409628,
|
| 2765 |
+
"learning_rate": 2e-05,
|
| 2766 |
+
"loss": 1.5258,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.7085201793721974,
|
| 2771 |
+
"grad_norm": 0.010284669697284698,
|
| 2772 |
+
"learning_rate": 2e-05,
|
| 2773 |
+
"loss": 1.5094,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.7103139013452915,
|
| 2778 |
+
"grad_norm": 0.010200968012213707,
|
| 2779 |
+
"learning_rate": 2e-05,
|
| 2780 |
+
"loss": 1.5172,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.7121076233183856,
|
| 2785 |
+
"grad_norm": 0.01015354972332716,
|
| 2786 |
+
"learning_rate": 2e-05,
|
| 2787 |
+
"loss": 1.5117,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.7139013452914799,
|
| 2792 |
+
"grad_norm": 0.009913373738527298,
|
| 2793 |
+
"learning_rate": 2e-05,
|
| 2794 |
+
"loss": 1.5268,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.715695067264574,
|
| 2799 |
+
"grad_norm": 0.010287330485880375,
|
| 2800 |
+
"learning_rate": 2e-05,
|
| 2801 |
+
"loss": 1.5211,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.7174887892376681,
|
| 2806 |
+
"grad_norm": 0.01057345885783434,
|
| 2807 |
+
"learning_rate": 2e-05,
|
| 2808 |
+
"loss": 1.5199,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.7192825112107624,
|
| 2813 |
+
"grad_norm": 0.010113878175616264,
|
| 2814 |
+
"learning_rate": 2e-05,
|
| 2815 |
+
"loss": 1.5168,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.7210762331838565,
|
| 2820 |
+
"grad_norm": 0.009940318763256073,
|
| 2821 |
+
"learning_rate": 2e-05,
|
| 2822 |
+
"loss": 1.5175,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.7228699551569506,
|
| 2827 |
+
"grad_norm": 0.010180394165217876,
|
| 2828 |
+
"learning_rate": 2e-05,
|
| 2829 |
+
"loss": 1.5211,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.7246636771300449,
|
| 2834 |
+
"grad_norm": 0.00961736124008894,
|
| 2835 |
+
"learning_rate": 2e-05,
|
| 2836 |
+
"loss": 1.5228,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.726457399103139,
|
| 2841 |
+
"grad_norm": 0.010378845036029816,
|
| 2842 |
+
"learning_rate": 2e-05,
|
| 2843 |
+
"loss": 1.522,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.7282511210762331,
|
| 2848 |
+
"grad_norm": 0.010189516469836235,
|
| 2849 |
+
"learning_rate": 2e-05,
|
| 2850 |
+
"loss": 1.525,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.7300448430493274,
|
| 2855 |
+
"grad_norm": 0.010004358366131783,
|
| 2856 |
+
"learning_rate": 2e-05,
|
| 2857 |
+
"loss": 1.5172,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.7318385650224215,
|
| 2862 |
+
"grad_norm": 0.010387993417680264,
|
| 2863 |
+
"learning_rate": 2e-05,
|
| 2864 |
+
"loss": 1.5246,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.7336322869955157,
|
| 2869 |
+
"grad_norm": 0.010004810988903046,
|
| 2870 |
+
"learning_rate": 2e-05,
|
| 2871 |
+
"loss": 1.5132,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.7354260089686099,
|
| 2876 |
+
"grad_norm": 0.009845850057899952,
|
| 2877 |
+
"learning_rate": 2e-05,
|
| 2878 |
+
"loss": 1.5248,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.737219730941704,
|
| 2883 |
+
"grad_norm": 0.010015097446739674,
|
| 2884 |
+
"learning_rate": 2e-05,
|
| 2885 |
+
"loss": 1.5196,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.7390134529147983,
|
| 2890 |
+
"grad_norm": 0.009975203312933445,
|
| 2891 |
+
"learning_rate": 2e-05,
|
| 2892 |
+
"loss": 1.5096,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.7408071748878924,
|
| 2897 |
+
"grad_norm": 0.010078891180455685,
|
| 2898 |
+
"learning_rate": 2e-05,
|
| 2899 |
+
"loss": 1.5162,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.7426008968609865,
|
| 2904 |
+
"grad_norm": 0.011885426007211208,
|
| 2905 |
+
"learning_rate": 2e-05,
|
| 2906 |
+
"loss": 1.5189,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.7443946188340808,
|
| 2911 |
+
"grad_norm": 0.009693853557109833,
|
| 2912 |
+
"learning_rate": 2e-05,
|
| 2913 |
+
"loss": 1.5194,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.7461883408071749,
|
| 2918 |
+
"grad_norm": 0.010337116196751595,
|
| 2919 |
+
"learning_rate": 2e-05,
|
| 2920 |
+
"loss": 1.5191,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.747982062780269,
|
| 2925 |
+
"grad_norm": 0.00993486400693655,
|
| 2926 |
+
"learning_rate": 2e-05,
|
| 2927 |
+
"loss": 1.5177,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.7497757847533633,
|
| 2932 |
+
"grad_norm": 0.010143253020942211,
|
| 2933 |
+
"learning_rate": 2e-05,
|
| 2934 |
+
"loss": 1.514,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.7515695067264574,
|
| 2939 |
+
"grad_norm": 0.010233073495328426,
|
| 2940 |
+
"learning_rate": 2e-05,
|
| 2941 |
+
"loss": 1.5154,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.7533632286995515,
|
| 2946 |
+
"grad_norm": 0.009982983581721783,
|
| 2947 |
+
"learning_rate": 2e-05,
|
| 2948 |
+
"loss": 1.5223,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.7551569506726458,
|
| 2953 |
+
"grad_norm": 0.010409766808152199,
|
| 2954 |
+
"learning_rate": 2e-05,
|
| 2955 |
+
"loss": 1.5152,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.7569506726457399,
|
| 2960 |
+
"grad_norm": 0.0099264495074749,
|
| 2961 |
+
"learning_rate": 2e-05,
|
| 2962 |
+
"loss": 1.5185,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.758744394618834,
|
| 2967 |
+
"grad_norm": 0.009928545914590359,
|
| 2968 |
+
"learning_rate": 2e-05,
|
| 2969 |
+
"loss": 1.4986,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.7605381165919283,
|
| 2974 |
+
"grad_norm": 0.009940563701093197,
|
| 2975 |
+
"learning_rate": 2e-05,
|
| 2976 |
+
"loss": 1.5071,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.7623318385650224,
|
| 2981 |
+
"grad_norm": 0.010767797008156776,
|
| 2982 |
+
"learning_rate": 2e-05,
|
| 2983 |
+
"loss": 1.5006,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.7641255605381166,
|
| 2988 |
+
"grad_norm": 0.010551121085882187,
|
| 2989 |
+
"learning_rate": 2e-05,
|
| 2990 |
+
"loss": 1.5201,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.7659192825112108,
|
| 2995 |
+
"grad_norm": 0.010118665173649788,
|
| 2996 |
+
"learning_rate": 2e-05,
|
| 2997 |
+
"loss": 1.5213,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.7677130044843049,
|
| 3002 |
+
"grad_norm": 0.010247626341879368,
|
| 3003 |
+
"learning_rate": 2e-05,
|
| 3004 |
+
"loss": 1.5178,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.7695067264573991,
|
| 3009 |
+
"grad_norm": 0.010188435204327106,
|
| 3010 |
+
"learning_rate": 2e-05,
|
| 3011 |
+
"loss": 1.5085,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.7713004484304933,
|
| 3016 |
+
"grad_norm": 0.010428003035485744,
|
| 3017 |
+
"learning_rate": 2e-05,
|
| 3018 |
+
"loss": 1.5124,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.7730941704035874,
|
| 3023 |
+
"grad_norm": 0.01012035645544529,
|
| 3024 |
+
"learning_rate": 2e-05,
|
| 3025 |
+
"loss": 1.5299,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.7748878923766817,
|
| 3030 |
+
"grad_norm": 0.010584665462374687,
|
| 3031 |
+
"learning_rate": 2e-05,
|
| 3032 |
+
"loss": 1.5095,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.7766816143497758,
|
| 3037 |
+
"grad_norm": 0.009979243390262127,
|
| 3038 |
+
"learning_rate": 2e-05,
|
| 3039 |
+
"loss": 1.5193,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.7784753363228699,
|
| 3044 |
+
"grad_norm": 0.00958004966378212,
|
| 3045 |
+
"learning_rate": 2e-05,
|
| 3046 |
+
"loss": 1.5214,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.7802690582959642,
|
| 3051 |
+
"grad_norm": 0.00973733700811863,
|
| 3052 |
+
"learning_rate": 2e-05,
|
| 3053 |
+
"loss": 1.5208,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.7820627802690583,
|
| 3058 |
+
"grad_norm": 0.010465665720403194,
|
| 3059 |
+
"learning_rate": 2e-05,
|
| 3060 |
+
"loss": 1.5227,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.7838565022421524,
|
| 3065 |
+
"grad_norm": 0.010098133236169815,
|
| 3066 |
+
"learning_rate": 2e-05,
|
| 3067 |
+
"loss": 1.5248,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.7856502242152467,
|
| 3072 |
+
"grad_norm": 0.10259313136339188,
|
| 3073 |
+
"learning_rate": 2e-05,
|
| 3074 |
+
"loss": 1.5222,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.7874439461883408,
|
| 3079 |
+
"grad_norm": 0.01040815282613039,
|
| 3080 |
+
"learning_rate": 2e-05,
|
| 3081 |
+
"loss": 1.5205,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.7892376681614349,
|
| 3086 |
+
"grad_norm": 0.010325520299375057,
|
| 3087 |
+
"learning_rate": 2e-05,
|
| 3088 |
+
"loss": 1.5189,
|
| 3089 |
+
"step": 440
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.7910313901345292,
|
| 3093 |
+
"grad_norm": 0.010079775005578995,
|
| 3094 |
+
"learning_rate": 2e-05,
|
| 3095 |
+
"loss": 1.5156,
|
| 3096 |
+
"step": 441
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.7928251121076233,
|
| 3100 |
+
"grad_norm": 0.010167201980948448,
|
| 3101 |
+
"learning_rate": 2e-05,
|
| 3102 |
+
"loss": 1.5116,
|
| 3103 |
+
"step": 442
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.7946188340807175,
|
| 3107 |
+
"grad_norm": 0.010806124657392502,
|
| 3108 |
+
"learning_rate": 2e-05,
|
| 3109 |
+
"loss": 1.5153,
|
| 3110 |
+
"step": 443
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.7964125560538117,
|
| 3114 |
+
"grad_norm": 0.010324080474674702,
|
| 3115 |
+
"learning_rate": 2e-05,
|
| 3116 |
+
"loss": 1.5246,
|
| 3117 |
+
"step": 444
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.7982062780269058,
|
| 3121 |
+
"grad_norm": 0.010092305950820446,
|
| 3122 |
+
"learning_rate": 2e-05,
|
| 3123 |
+
"loss": 1.5282,
|
| 3124 |
+
"step": 445
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.8,
|
| 3128 |
+
"grad_norm": 0.01007048413157463,
|
| 3129 |
+
"learning_rate": 2e-05,
|
| 3130 |
+
"loss": 1.5108,
|
| 3131 |
+
"step": 446
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.8017937219730942,
|
| 3135 |
+
"grad_norm": 0.010184276849031448,
|
| 3136 |
+
"learning_rate": 2e-05,
|
| 3137 |
+
"loss": 1.51,
|
| 3138 |
+
"step": 447
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.8035874439461883,
|
| 3142 |
+
"grad_norm": 0.010521662421524525,
|
| 3143 |
+
"learning_rate": 2e-05,
|
| 3144 |
+
"loss": 1.5139,
|
| 3145 |
+
"step": 448
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.8053811659192825,
|
| 3149 |
+
"grad_norm": 0.010600044392049313,
|
| 3150 |
+
"learning_rate": 2e-05,
|
| 3151 |
+
"loss": 1.5091,
|
| 3152 |
+
"step": 449
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.8071748878923767,
|
| 3156 |
+
"grad_norm": 0.009714100509881973,
|
| 3157 |
+
"learning_rate": 2e-05,
|
| 3158 |
+
"loss": 1.5122,
|
| 3159 |
+
"step": 450
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.8089686098654708,
|
| 3163 |
+
"grad_norm": 0.010295005515217781,
|
| 3164 |
+
"learning_rate": 2e-05,
|
| 3165 |
+
"loss": 1.52,
|
| 3166 |
+
"step": 451
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.810762331838565,
|
| 3170 |
+
"grad_norm": 0.010034569539129734,
|
| 3171 |
+
"learning_rate": 2e-05,
|
| 3172 |
+
"loss": 1.5197,
|
| 3173 |
+
"step": 452
|
| 3174 |
+
},
|
| 3175 |
+
{
|
| 3176 |
+
"epoch": 0.8125560538116592,
|
| 3177 |
+
"grad_norm": 0.010086962021887302,
|
| 3178 |
+
"learning_rate": 2e-05,
|
| 3179 |
+
"loss": 1.5117,
|
| 3180 |
+
"step": 453
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"epoch": 0.8143497757847533,
|
| 3184 |
+
"grad_norm": 0.010277335532009602,
|
| 3185 |
+
"learning_rate": 2e-05,
|
| 3186 |
+
"loss": 1.5033,
|
| 3187 |
+
"step": 454
|
| 3188 |
+
},
|
| 3189 |
+
{
|
| 3190 |
+
"epoch": 0.8161434977578476,
|
| 3191 |
+
"grad_norm": 0.010540721006691456,
|
| 3192 |
+
"learning_rate": 2e-05,
|
| 3193 |
+
"loss": 1.5166,
|
| 3194 |
+
"step": 455
|
| 3195 |
+
},
|
| 3196 |
+
{
|
| 3197 |
+
"epoch": 0.8179372197309417,
|
| 3198 |
+
"grad_norm": 0.009755424223840237,
|
| 3199 |
+
"learning_rate": 2e-05,
|
| 3200 |
+
"loss": 1.5149,
|
| 3201 |
+
"step": 456
|
| 3202 |
+
},
|
| 3203 |
+
{
|
| 3204 |
+
"epoch": 0.8197309417040358,
|
| 3205 |
+
"grad_norm": 0.00984253827482462,
|
| 3206 |
+
"learning_rate": 2e-05,
|
| 3207 |
+
"loss": 1.5093,
|
| 3208 |
+
"step": 457
|
| 3209 |
+
},
|
| 3210 |
+
{
|
| 3211 |
+
"epoch": 0.8215246636771301,
|
| 3212 |
+
"grad_norm": 0.009836334735155106,
|
| 3213 |
+
"learning_rate": 2e-05,
|
| 3214 |
+
"loss": 1.5141,
|
| 3215 |
+
"step": 458
|
| 3216 |
+
},
|
| 3217 |
+
{
|
| 3218 |
+
"epoch": 0.8233183856502242,
|
| 3219 |
+
"grad_norm": 0.01032332144677639,
|
| 3220 |
+
"learning_rate": 2e-05,
|
| 3221 |
+
"loss": 1.5241,
|
| 3222 |
+
"step": 459
|
| 3223 |
+
},
|
| 3224 |
+
{
|
| 3225 |
+
"epoch": 0.8251121076233184,
|
| 3226 |
+
"grad_norm": 0.010635129176080227,
|
| 3227 |
+
"learning_rate": 2e-05,
|
| 3228 |
+
"loss": 1.5068,
|
| 3229 |
+
"step": 460
|
| 3230 |
+
},
|
| 3231 |
+
{
|
| 3232 |
+
"epoch": 0.8269058295964126,
|
| 3233 |
+
"grad_norm": 0.009664127603173256,
|
| 3234 |
+
"learning_rate": 2e-05,
|
| 3235 |
+
"loss": 1.5052,
|
| 3236 |
+
"step": 461
|
| 3237 |
+
},
|
| 3238 |
+
{
|
| 3239 |
+
"epoch": 0.8286995515695067,
|
| 3240 |
+
"grad_norm": 0.010554889217019081,
|
| 3241 |
+
"learning_rate": 2e-05,
|
| 3242 |
+
"loss": 1.5071,
|
| 3243 |
+
"step": 462
|
| 3244 |
+
},
|
| 3245 |
+
{
|
| 3246 |
+
"epoch": 0.8304932735426009,
|
| 3247 |
+
"grad_norm": 0.009871057234704494,
|
| 3248 |
+
"learning_rate": 2e-05,
|
| 3249 |
+
"loss": 1.5189,
|
| 3250 |
+
"step": 463
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"epoch": 0.8322869955156951,
|
| 3254 |
+
"grad_norm": 0.010431516915559769,
|
| 3255 |
+
"learning_rate": 2e-05,
|
| 3256 |
+
"loss": 1.5183,
|
| 3257 |
+
"step": 464
|
| 3258 |
+
},
|
| 3259 |
+
{
|
| 3260 |
+
"epoch": 0.8340807174887892,
|
| 3261 |
+
"grad_norm": 0.009860005229711533,
|
| 3262 |
+
"learning_rate": 2e-05,
|
| 3263 |
+
"loss": 1.5213,
|
| 3264 |
+
"step": 465
|
| 3265 |
+
},
|
| 3266 |
+
{
|
| 3267 |
+
"epoch": 0.8358744394618834,
|
| 3268 |
+
"grad_norm": 0.010233579203486443,
|
| 3269 |
+
"learning_rate": 2e-05,
|
| 3270 |
+
"loss": 1.5182,
|
| 3271 |
+
"step": 466
|
| 3272 |
+
},
|
| 3273 |
+
{
|
| 3274 |
+
"epoch": 0.8376681614349776,
|
| 3275 |
+
"grad_norm": 0.010311591438949108,
|
| 3276 |
+
"learning_rate": 2e-05,
|
| 3277 |
+
"loss": 1.5092,
|
| 3278 |
+
"step": 467
|
| 3279 |
+
},
|
| 3280 |
+
{
|
| 3281 |
+
"epoch": 0.8394618834080717,
|
| 3282 |
+
"grad_norm": 0.010733729228377342,
|
| 3283 |
+
"learning_rate": 2e-05,
|
| 3284 |
+
"loss": 1.5186,
|
| 3285 |
+
"step": 468
|
| 3286 |
+
},
|
| 3287 |
+
{
|
| 3288 |
+
"epoch": 0.841255605381166,
|
| 3289 |
+
"grad_norm": 0.009951340965926647,
|
| 3290 |
+
"learning_rate": 2e-05,
|
| 3291 |
+
"loss": 1.5097,
|
| 3292 |
+
"step": 469
|
| 3293 |
+
},
|
| 3294 |
+
{
|
| 3295 |
+
"epoch": 0.8430493273542601,
|
| 3296 |
+
"grad_norm": 0.01003777701407671,
|
| 3297 |
+
"learning_rate": 2e-05,
|
| 3298 |
+
"loss": 1.5173,
|
| 3299 |
+
"step": 470
|
| 3300 |
+
},
|
| 3301 |
+
{
|
| 3302 |
+
"epoch": 0.8448430493273542,
|
| 3303 |
+
"grad_norm": 0.009939250536262989,
|
| 3304 |
+
"learning_rate": 2e-05,
|
| 3305 |
+
"loss": 1.5108,
|
| 3306 |
+
"step": 471
|
| 3307 |
+
},
|
| 3308 |
+
{
|
| 3309 |
+
"epoch": 0.8466367713004485,
|
| 3310 |
+
"grad_norm": 0.009835812263190746,
|
| 3311 |
+
"learning_rate": 2e-05,
|
| 3312 |
+
"loss": 1.5272,
|
| 3313 |
+
"step": 472
|
| 3314 |
+
},
|
| 3315 |
+
{
|
| 3316 |
+
"epoch": 0.8484304932735426,
|
| 3317 |
+
"grad_norm": 0.010321546345949173,
|
| 3318 |
+
"learning_rate": 2e-05,
|
| 3319 |
+
"loss": 1.5193,
|
| 3320 |
+
"step": 473
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"epoch": 0.8502242152466367,
|
| 3324 |
+
"grad_norm": 0.01006554439663887,
|
| 3325 |
+
"learning_rate": 2e-05,
|
| 3326 |
+
"loss": 1.5165,
|
| 3327 |
+
"step": 474
|
| 3328 |
+
},
|
| 3329 |
+
{
|
| 3330 |
+
"epoch": 0.852017937219731,
|
| 3331 |
+
"grad_norm": 0.009972809813916683,
|
| 3332 |
+
"learning_rate": 2e-05,
|
| 3333 |
+
"loss": 1.5228,
|
| 3334 |
+
"step": 475
|
| 3335 |
+
},
|
| 3336 |
+
{
|
| 3337 |
+
"epoch": 0.8538116591928251,
|
| 3338 |
+
"grad_norm": 0.010388972237706184,
|
| 3339 |
+
"learning_rate": 2e-05,
|
| 3340 |
+
"loss": 1.5188,
|
| 3341 |
+
"step": 476
|
| 3342 |
+
},
|
| 3343 |
+
{
|
| 3344 |
+
"epoch": 0.8556053811659193,
|
| 3345 |
+
"grad_norm": 0.010111154057085514,
|
| 3346 |
+
"learning_rate": 2e-05,
|
| 3347 |
+
"loss": 1.5199,
|
| 3348 |
+
"step": 477
|
| 3349 |
+
},
|
| 3350 |
+
{
|
| 3351 |
+
"epoch": 0.8573991031390135,
|
| 3352 |
+
"grad_norm": 0.01029327604919672,
|
| 3353 |
+
"learning_rate": 2e-05,
|
| 3354 |
+
"loss": 1.516,
|
| 3355 |
+
"step": 478
|
| 3356 |
+
},
|
| 3357 |
+
{
|
| 3358 |
+
"epoch": 0.8591928251121076,
|
| 3359 |
+
"grad_norm": 0.010400544852018356,
|
| 3360 |
+
"learning_rate": 2e-05,
|
| 3361 |
+
"loss": 1.5218,
|
| 3362 |
+
"step": 479
|
| 3363 |
+
},
|
| 3364 |
+
{
|
| 3365 |
+
"epoch": 0.8609865470852018,
|
| 3366 |
+
"grad_norm": 0.0099885743111372,
|
| 3367 |
+
"learning_rate": 2e-05,
|
| 3368 |
+
"loss": 1.5155,
|
| 3369 |
+
"step": 480
|
| 3370 |
+
},
|
| 3371 |
+
{
|
| 3372 |
+
"epoch": 0.862780269058296,
|
| 3373 |
+
"grad_norm": 0.010007279925048351,
|
| 3374 |
+
"learning_rate": 2e-05,
|
| 3375 |
+
"loss": 1.5205,
|
| 3376 |
+
"step": 481
|
| 3377 |
+
},
|
| 3378 |
+
{
|
| 3379 |
+
"epoch": 0.8645739910313901,
|
| 3380 |
+
"grad_norm": 0.01053563691675663,
|
| 3381 |
+
"learning_rate": 2e-05,
|
| 3382 |
+
"loss": 1.5019,
|
| 3383 |
+
"step": 482
|
| 3384 |
+
},
|
| 3385 |
+
{
|
| 3386 |
+
"epoch": 0.8663677130044843,
|
| 3387 |
+
"grad_norm": 0.01031608134508133,
|
| 3388 |
+
"learning_rate": 2e-05,
|
| 3389 |
+
"loss": 1.5217,
|
| 3390 |
+
"step": 483
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"epoch": 0.8681614349775785,
|
| 3394 |
+
"grad_norm": 0.010082092136144638,
|
| 3395 |
+
"learning_rate": 2e-05,
|
| 3396 |
+
"loss": 1.5073,
|
| 3397 |
+
"step": 484
|
| 3398 |
+
},
|
| 3399 |
+
{
|
| 3400 |
+
"epoch": 0.8699551569506726,
|
| 3401 |
+
"grad_norm": 0.01012254785746336,
|
| 3402 |
+
"learning_rate": 2e-05,
|
| 3403 |
+
"loss": 1.5101,
|
| 3404 |
+
"step": 485
|
| 3405 |
+
},
|
| 3406 |
+
{
|
| 3407 |
+
"epoch": 0.8717488789237668,
|
| 3408 |
+
"grad_norm": 0.010539901442825794,
|
| 3409 |
+
"learning_rate": 2e-05,
|
| 3410 |
+
"loss": 1.5209,
|
| 3411 |
+
"step": 486
|
| 3412 |
+
},
|
| 3413 |
+
{
|
| 3414 |
+
"epoch": 0.873542600896861,
|
| 3415 |
+
"grad_norm": 0.009883386082947254,
|
| 3416 |
+
"learning_rate": 2e-05,
|
| 3417 |
+
"loss": 1.5275,
|
| 3418 |
+
"step": 487
|
| 3419 |
+
},
|
| 3420 |
+
{
|
| 3421 |
+
"epoch": 0.8753363228699551,
|
| 3422 |
+
"grad_norm": 0.010055874474346638,
|
| 3423 |
+
"learning_rate": 2e-05,
|
| 3424 |
+
"loss": 1.521,
|
| 3425 |
+
"step": 488
|
| 3426 |
+
},
|
| 3427 |
+
{
|
| 3428 |
+
"epoch": 0.8771300448430494,
|
| 3429 |
+
"grad_norm": 0.010441599413752556,
|
| 3430 |
+
"learning_rate": 2e-05,
|
| 3431 |
+
"loss": 1.5253,
|
| 3432 |
+
"step": 489
|
| 3433 |
+
},
|
| 3434 |
+
{
|
| 3435 |
+
"epoch": 0.8789237668161435,
|
| 3436 |
+
"grad_norm": 0.010321282781660557,
|
| 3437 |
+
"learning_rate": 2e-05,
|
| 3438 |
+
"loss": 1.5128,
|
| 3439 |
+
"step": 490
|
| 3440 |
+
},
|
| 3441 |
+
{
|
| 3442 |
+
"epoch": 0.8807174887892377,
|
| 3443 |
+
"grad_norm": 0.010404079221189022,
|
| 3444 |
+
"learning_rate": 2e-05,
|
| 3445 |
+
"loss": 1.5216,
|
| 3446 |
+
"step": 491
|
| 3447 |
+
},
|
| 3448 |
+
{
|
| 3449 |
+
"epoch": 0.8825112107623319,
|
| 3450 |
+
"grad_norm": 0.010680857114493847,
|
| 3451 |
+
"learning_rate": 2e-05,
|
| 3452 |
+
"loss": 1.5102,
|
| 3453 |
+
"step": 492
|
| 3454 |
+
},
|
| 3455 |
+
{
|
| 3456 |
+
"epoch": 0.884304932735426,
|
| 3457 |
+
"grad_norm": 0.009785238653421402,
|
| 3458 |
+
"learning_rate": 2e-05,
|
| 3459 |
+
"loss": 1.5152,
|
| 3460 |
+
"step": 493
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"epoch": 0.8860986547085202,
|
| 3464 |
+
"grad_norm": 0.010622934438288212,
|
| 3465 |
+
"learning_rate": 2e-05,
|
| 3466 |
+
"loss": 1.5134,
|
| 3467 |
+
"step": 494
|
| 3468 |
+
},
|
| 3469 |
+
{
|
| 3470 |
+
"epoch": 0.8878923766816144,
|
| 3471 |
+
"grad_norm": 0.009563595987856388,
|
| 3472 |
+
"learning_rate": 2e-05,
|
| 3473 |
+
"loss": 1.5213,
|
| 3474 |
+
"step": 495
|
| 3475 |
+
},
|
| 3476 |
+
{
|
| 3477 |
+
"epoch": 0.8896860986547085,
|
| 3478 |
+
"grad_norm": 0.009900403209030628,
|
| 3479 |
+
"learning_rate": 2e-05,
|
| 3480 |
+
"loss": 1.5254,
|
| 3481 |
+
"step": 496
|
| 3482 |
+
},
|
| 3483 |
+
{
|
| 3484 |
+
"epoch": 0.8914798206278027,
|
| 3485 |
+
"grad_norm": 0.010441206395626068,
|
| 3486 |
+
"learning_rate": 2e-05,
|
| 3487 |
+
"loss": 1.5042,
|
| 3488 |
+
"step": 497
|
| 3489 |
+
},
|
| 3490 |
+
{
|
| 3491 |
+
"epoch": 0.8932735426008969,
|
| 3492 |
+
"grad_norm": 0.010110273025929928,
|
| 3493 |
+
"learning_rate": 2e-05,
|
| 3494 |
+
"loss": 1.5141,
|
| 3495 |
+
"step": 498
|
| 3496 |
+
},
|
| 3497 |
+
{
|
| 3498 |
+
"epoch": 0.895067264573991,
|
| 3499 |
+
"grad_norm": 0.00976527575403452,
|
| 3500 |
+
"learning_rate": 2e-05,
|
| 3501 |
+
"loss": 1.5189,
|
| 3502 |
+
"step": 499
|
| 3503 |
+
},
|
| 3504 |
+
{
|
| 3505 |
+
"epoch": 0.8968609865470852,
|
| 3506 |
+
"grad_norm": 0.010270185768604279,
|
| 3507 |
+
"learning_rate": 2e-05,
|
| 3508 |
+
"loss": 1.5128,
|
| 3509 |
+
"step": 500
|
| 3510 |
+
},
|
| 3511 |
+
{
|
| 3512 |
+
"epoch": 0.8986547085201794,
|
| 3513 |
+
"grad_norm": 0.010477078147232533,
|
| 3514 |
+
"learning_rate": 2e-05,
|
| 3515 |
+
"loss": 1.5331,
|
| 3516 |
+
"step": 501
|
| 3517 |
+
},
|
| 3518 |
+
{
|
| 3519 |
+
"epoch": 0.9004484304932735,
|
| 3520 |
+
"grad_norm": 0.009786723181605339,
|
| 3521 |
+
"learning_rate": 2e-05,
|
| 3522 |
+
"loss": 1.5143,
|
| 3523 |
+
"step": 502
|
| 3524 |
+
},
|
| 3525 |
+
{
|
| 3526 |
+
"epoch": 0.9022421524663677,
|
| 3527 |
+
"grad_norm": 0.009838691912591457,
|
| 3528 |
+
"learning_rate": 2e-05,
|
| 3529 |
+
"loss": 1.5237,
|
| 3530 |
+
"step": 503
|
| 3531 |
+
},
|
| 3532 |
+
{
|
| 3533 |
+
"epoch": 0.9040358744394619,
|
| 3534 |
+
"grad_norm": 0.010305250994861126,
|
| 3535 |
+
"learning_rate": 2e-05,
|
| 3536 |
+
"loss": 1.5236,
|
| 3537 |
+
"step": 504
|
| 3538 |
+
},
|
| 3539 |
+
{
|
| 3540 |
+
"epoch": 0.905829596412556,
|
| 3541 |
+
"grad_norm": 0.010098317638039589,
|
| 3542 |
+
"learning_rate": 2e-05,
|
| 3543 |
+
"loss": 1.5189,
|
| 3544 |
+
"step": 505
|
| 3545 |
+
},
|
| 3546 |
+
{
|
| 3547 |
+
"epoch": 0.9076233183856502,
|
| 3548 |
+
"grad_norm": 0.010335841216146946,
|
| 3549 |
+
"learning_rate": 2e-05,
|
| 3550 |
+
"loss": 1.519,
|
| 3551 |
+
"step": 506
|
| 3552 |
+
},
|
| 3553 |
+
{
|
| 3554 |
+
"epoch": 0.9094170403587444,
|
| 3555 |
+
"grad_norm": 0.009809168055653572,
|
| 3556 |
+
"learning_rate": 2e-05,
|
| 3557 |
+
"loss": 1.5176,
|
| 3558 |
+
"step": 507
|
| 3559 |
+
},
|
| 3560 |
+
{
|
| 3561 |
+
"epoch": 0.9112107623318386,
|
| 3562 |
+
"grad_norm": 0.01069081760942936,
|
| 3563 |
+
"learning_rate": 2e-05,
|
| 3564 |
+
"loss": 1.5055,
|
| 3565 |
+
"step": 508
|
| 3566 |
+
},
|
| 3567 |
+
{
|
| 3568 |
+
"epoch": 0.9130044843049328,
|
| 3569 |
+
"grad_norm": 0.009927291423082352,
|
| 3570 |
+
"learning_rate": 2e-05,
|
| 3571 |
+
"loss": 1.5224,
|
| 3572 |
+
"step": 509
|
| 3573 |
+
},
|
| 3574 |
+
{
|
| 3575 |
+
"epoch": 0.9147982062780269,
|
| 3576 |
+
"grad_norm": 0.010560589842498302,
|
| 3577 |
+
"learning_rate": 2e-05,
|
| 3578 |
+
"loss": 1.5129,
|
| 3579 |
+
"step": 510
|
| 3580 |
+
},
|
| 3581 |
+
{
|
| 3582 |
+
"epoch": 0.9165919282511211,
|
| 3583 |
+
"grad_norm": 0.010154438205063343,
|
| 3584 |
+
"learning_rate": 2e-05,
|
| 3585 |
+
"loss": 1.52,
|
| 3586 |
+
"step": 511
|
| 3587 |
+
},
|
| 3588 |
+
{
|
| 3589 |
+
"epoch": 0.9183856502242153,
|
| 3590 |
+
"grad_norm": 0.010346156544983387,
|
| 3591 |
+
"learning_rate": 2e-05,
|
| 3592 |
+
"loss": 1.5194,
|
| 3593 |
+
"step": 512
|
| 3594 |
+
},
|
| 3595 |
+
{
|
| 3596 |
+
"epoch": 0.9201793721973094,
|
| 3597 |
+
"grad_norm": 0.010523281060159206,
|
| 3598 |
+
"learning_rate": 2e-05,
|
| 3599 |
+
"loss": 1.5187,
|
| 3600 |
+
"step": 513
|
| 3601 |
+
},
|
| 3602 |
+
{
|
| 3603 |
+
"epoch": 0.9219730941704036,
|
| 3604 |
+
"grad_norm": 0.010443002916872501,
|
| 3605 |
+
"learning_rate": 2e-05,
|
| 3606 |
+
"loss": 1.5059,
|
| 3607 |
+
"step": 514
|
| 3608 |
+
},
|
| 3609 |
+
{
|
| 3610 |
+
"epoch": 0.9237668161434978,
|
| 3611 |
+
"grad_norm": 0.010005362331867218,
|
| 3612 |
+
"learning_rate": 2e-05,
|
| 3613 |
+
"loss": 1.5102,
|
| 3614 |
+
"step": 515
|
| 3615 |
+
},
|
| 3616 |
+
{
|
| 3617 |
+
"epoch": 0.9255605381165919,
|
| 3618 |
+
"grad_norm": 0.010285025462508202,
|
| 3619 |
+
"learning_rate": 2e-05,
|
| 3620 |
+
"loss": 1.5217,
|
| 3621 |
+
"step": 516
|
| 3622 |
+
},
|
| 3623 |
+
{
|
| 3624 |
+
"epoch": 0.9273542600896861,
|
| 3625 |
+
"grad_norm": 0.010401098988950253,
|
| 3626 |
+
"learning_rate": 2e-05,
|
| 3627 |
+
"loss": 1.5243,
|
| 3628 |
+
"step": 517
|
| 3629 |
+
},
|
| 3630 |
+
{
|
| 3631 |
+
"epoch": 0.9291479820627803,
|
| 3632 |
+
"grad_norm": 0.010455128736793995,
|
| 3633 |
+
"learning_rate": 2e-05,
|
| 3634 |
+
"loss": 1.5054,
|
| 3635 |
+
"step": 518
|
| 3636 |
+
},
|
| 3637 |
+
{
|
| 3638 |
+
"epoch": 0.9309417040358744,
|
| 3639 |
+
"grad_norm": 0.00987928081303835,
|
| 3640 |
+
"learning_rate": 2e-05,
|
| 3641 |
+
"loss": 1.5053,
|
| 3642 |
+
"step": 519
|
| 3643 |
+
},
|
| 3644 |
+
{
|
| 3645 |
+
"epoch": 0.9327354260089686,
|
| 3646 |
+
"grad_norm": 0.010212692432105541,
|
| 3647 |
+
"learning_rate": 2e-05,
|
| 3648 |
+
"loss": 1.5151,
|
| 3649 |
+
"step": 520
|
| 3650 |
+
},
|
| 3651 |
+
{
|
| 3652 |
+
"epoch": 0.9345291479820628,
|
| 3653 |
+
"grad_norm": 0.010937588289380074,
|
| 3654 |
+
"learning_rate": 2e-05,
|
| 3655 |
+
"loss": 1.5099,
|
| 3656 |
+
"step": 521
|
| 3657 |
+
},
|
| 3658 |
+
{
|
| 3659 |
+
"epoch": 0.9363228699551569,
|
| 3660 |
+
"grad_norm": 0.010248001664876938,
|
| 3661 |
+
"learning_rate": 2e-05,
|
| 3662 |
+
"loss": 1.5256,
|
| 3663 |
+
"step": 522
|
| 3664 |
+
},
|
| 3665 |
+
{
|
| 3666 |
+
"epoch": 0.9381165919282511,
|
| 3667 |
+
"grad_norm": 0.010430903173983097,
|
| 3668 |
+
"learning_rate": 2e-05,
|
| 3669 |
+
"loss": 1.5056,
|
| 3670 |
+
"step": 523
|
| 3671 |
+
},
|
| 3672 |
+
{
|
| 3673 |
+
"epoch": 0.9399103139013453,
|
| 3674 |
+
"grad_norm": 0.0102499695494771,
|
| 3675 |
+
"learning_rate": 2e-05,
|
| 3676 |
+
"loss": 1.5285,
|
| 3677 |
+
"step": 524
|
| 3678 |
+
},
|
| 3679 |
+
{
|
| 3680 |
+
"epoch": 0.9417040358744395,
|
| 3681 |
+
"grad_norm": 0.010674213990569115,
|
| 3682 |
+
"learning_rate": 2e-05,
|
| 3683 |
+
"loss": 1.5136,
|
| 3684 |
+
"step": 525
|
| 3685 |
+
},
|
| 3686 |
+
{
|
| 3687 |
+
"epoch": 0.9434977578475336,
|
| 3688 |
+
"grad_norm": 0.010732615366578102,
|
| 3689 |
+
"learning_rate": 2e-05,
|
| 3690 |
+
"loss": 1.5119,
|
| 3691 |
+
"step": 526
|
| 3692 |
+
},
|
| 3693 |
+
{
|
| 3694 |
+
"epoch": 0.9452914798206278,
|
| 3695 |
+
"grad_norm": 0.009994648396968842,
|
| 3696 |
+
"learning_rate": 2e-05,
|
| 3697 |
+
"loss": 1.5228,
|
| 3698 |
+
"step": 527
|
| 3699 |
+
},
|
| 3700 |
+
{
|
| 3701 |
+
"epoch": 0.947085201793722,
|
| 3702 |
+
"grad_norm": 0.010234368033707142,
|
| 3703 |
+
"learning_rate": 2e-05,
|
| 3704 |
+
"loss": 1.5258,
|
| 3705 |
+
"step": 528
|
| 3706 |
+
},
|
| 3707 |
+
{
|
| 3708 |
+
"epoch": 0.9488789237668162,
|
| 3709 |
+
"grad_norm": 0.010327205993235111,
|
| 3710 |
+
"learning_rate": 2e-05,
|
| 3711 |
+
"loss": 1.5156,
|
| 3712 |
+
"step": 529
|
| 3713 |
+
},
|
| 3714 |
+
{
|
| 3715 |
+
"epoch": 0.9506726457399103,
|
| 3716 |
+
"grad_norm": 0.009836922399699688,
|
| 3717 |
+
"learning_rate": 2e-05,
|
| 3718 |
+
"loss": 1.5171,
|
| 3719 |
+
"step": 530
|
| 3720 |
+
},
|
| 3721 |
+
{
|
| 3722 |
+
"epoch": 0.9524663677130045,
|
| 3723 |
+
"grad_norm": 0.009962068870663643,
|
| 3724 |
+
"learning_rate": 2e-05,
|
| 3725 |
+
"loss": 1.5125,
|
| 3726 |
+
"step": 531
|
| 3727 |
+
},
|
| 3728 |
+
{
|
| 3729 |
+
"epoch": 0.9542600896860987,
|
| 3730 |
+
"grad_norm": 0.010127882473170757,
|
| 3731 |
+
"learning_rate": 2e-05,
|
| 3732 |
+
"loss": 1.5182,
|
| 3733 |
+
"step": 532
|
| 3734 |
+
},
|
| 3735 |
+
{
|
| 3736 |
+
"epoch": 0.9560538116591928,
|
| 3737 |
+
"grad_norm": 0.010251611471176147,
|
| 3738 |
+
"learning_rate": 2e-05,
|
| 3739 |
+
"loss": 1.5139,
|
| 3740 |
+
"step": 533
|
| 3741 |
+
},
|
| 3742 |
+
{
|
| 3743 |
+
"epoch": 0.957847533632287,
|
| 3744 |
+
"grad_norm": 0.010081682354211807,
|
| 3745 |
+
"learning_rate": 2e-05,
|
| 3746 |
+
"loss": 1.5239,
|
| 3747 |
+
"step": 534
|
| 3748 |
+
},
|
| 3749 |
+
{
|
| 3750 |
+
"epoch": 0.9596412556053812,
|
| 3751 |
+
"grad_norm": 0.010235367342829704,
|
| 3752 |
+
"learning_rate": 2e-05,
|
| 3753 |
+
"loss": 1.5159,
|
| 3754 |
+
"step": 535
|
| 3755 |
+
},
|
| 3756 |
+
{
|
| 3757 |
+
"epoch": 0.9614349775784753,
|
| 3758 |
+
"grad_norm": 0.009694702923297882,
|
| 3759 |
+
"learning_rate": 2e-05,
|
| 3760 |
+
"loss": 1.5174,
|
| 3761 |
+
"step": 536
|
| 3762 |
+
},
|
| 3763 |
+
{
|
| 3764 |
+
"epoch": 0.9632286995515695,
|
| 3765 |
+
"grad_norm": 0.010224996134638786,
|
| 3766 |
+
"learning_rate": 2e-05,
|
| 3767 |
+
"loss": 1.5171,
|
| 3768 |
+
"step": 537
|
| 3769 |
+
},
|
| 3770 |
+
{
|
| 3771 |
+
"epoch": 0.9650224215246637,
|
| 3772 |
+
"grad_norm": 0.010206632316112518,
|
| 3773 |
+
"learning_rate": 2e-05,
|
| 3774 |
+
"loss": 1.5223,
|
| 3775 |
+
"step": 538
|
| 3776 |
+
},
|
| 3777 |
+
{
|
| 3778 |
+
"epoch": 0.9668161434977578,
|
| 3779 |
+
"grad_norm": 0.010011864826083183,
|
| 3780 |
+
"learning_rate": 2e-05,
|
| 3781 |
+
"loss": 1.5282,
|
| 3782 |
+
"step": 539
|
| 3783 |
+
},
|
| 3784 |
+
{
|
| 3785 |
+
"epoch": 0.968609865470852,
|
| 3786 |
+
"grad_norm": 0.010364921763539314,
|
| 3787 |
+
"learning_rate": 2e-05,
|
| 3788 |
+
"loss": 1.5092,
|
| 3789 |
+
"step": 540
|
| 3790 |
+
},
|
| 3791 |
+
{
|
| 3792 |
+
"epoch": 0.9704035874439462,
|
| 3793 |
+
"grad_norm": 0.010109508410096169,
|
| 3794 |
+
"learning_rate": 2e-05,
|
| 3795 |
+
"loss": 1.5068,
|
| 3796 |
+
"step": 541
|
| 3797 |
+
},
|
| 3798 |
+
{
|
| 3799 |
+
"epoch": 0.9721973094170404,
|
| 3800 |
+
"grad_norm": 0.00964987464249134,
|
| 3801 |
+
"learning_rate": 2e-05,
|
| 3802 |
+
"loss": 1.5089,
|
| 3803 |
+
"step": 542
|
| 3804 |
+
},
|
| 3805 |
+
{
|
| 3806 |
+
"epoch": 0.9739910313901345,
|
| 3807 |
+
"grad_norm": 0.010244207456707954,
|
| 3808 |
+
"learning_rate": 2e-05,
|
| 3809 |
+
"loss": 1.5217,
|
| 3810 |
+
"step": 543
|
| 3811 |
+
},
|
| 3812 |
+
{
|
| 3813 |
+
"epoch": 0.9757847533632287,
|
| 3814 |
+
"grad_norm": 0.009797874838113785,
|
| 3815 |
+
"learning_rate": 2e-05,
|
| 3816 |
+
"loss": 1.5143,
|
| 3817 |
+
"step": 544
|
| 3818 |
+
},
|
| 3819 |
+
{
|
| 3820 |
+
"epoch": 0.9775784753363229,
|
| 3821 |
+
"grad_norm": 0.010056640952825546,
|
| 3822 |
+
"learning_rate": 2e-05,
|
| 3823 |
+
"loss": 1.5276,
|
| 3824 |
+
"step": 545
|
| 3825 |
+
},
|
| 3826 |
+
{
|
| 3827 |
+
"epoch": 0.979372197309417,
|
| 3828 |
+
"grad_norm": 0.009898710064589977,
|
| 3829 |
+
"learning_rate": 2e-05,
|
| 3830 |
+
"loss": 1.5222,
|
| 3831 |
+
"step": 546
|
| 3832 |
+
},
|
| 3833 |
+
{
|
| 3834 |
+
"epoch": 0.9811659192825112,
|
| 3835 |
+
"grad_norm": 0.0099082225933671,
|
| 3836 |
+
"learning_rate": 2e-05,
|
| 3837 |
+
"loss": 1.5276,
|
| 3838 |
+
"step": 547
|
| 3839 |
+
},
|
| 3840 |
+
{
|
| 3841 |
+
"epoch": 0.9829596412556054,
|
| 3842 |
+
"grad_norm": 0.01018478162586689,
|
| 3843 |
+
"learning_rate": 2e-05,
|
| 3844 |
+
"loss": 1.5217,
|
| 3845 |
+
"step": 548
|
| 3846 |
+
},
|
| 3847 |
+
{
|
| 3848 |
+
"epoch": 0.9847533632286996,
|
| 3849 |
+
"grad_norm": 0.009828625246882439,
|
| 3850 |
+
"learning_rate": 2e-05,
|
| 3851 |
+
"loss": 1.5194,
|
| 3852 |
+
"step": 549
|
| 3853 |
+
},
|
| 3854 |
+
{
|
| 3855 |
+
"epoch": 0.9865470852017937,
|
| 3856 |
+
"grad_norm": 0.010311014950275421,
|
| 3857 |
+
"learning_rate": 2e-05,
|
| 3858 |
+
"loss": 1.5138,
|
| 3859 |
+
"step": 550
|
| 3860 |
+
},
|
| 3861 |
+
{
|
| 3862 |
+
"epoch": 0.9883408071748879,
|
| 3863 |
+
"grad_norm": 0.010840130038559437,
|
| 3864 |
+
"learning_rate": 2e-05,
|
| 3865 |
+
"loss": 1.5044,
|
| 3866 |
+
"step": 551
|
| 3867 |
+
},
|
| 3868 |
+
{
|
| 3869 |
+
"epoch": 0.9901345291479821,
|
| 3870 |
+
"grad_norm": 0.009595104493200779,
|
| 3871 |
+
"learning_rate": 2e-05,
|
| 3872 |
+
"loss": 1.5165,
|
| 3873 |
+
"step": 552
|
| 3874 |
+
},
|
| 3875 |
+
{
|
| 3876 |
+
"epoch": 0.9919282511210762,
|
| 3877 |
+
"grad_norm": 0.01027593482285738,
|
| 3878 |
+
"learning_rate": 2e-05,
|
| 3879 |
+
"loss": 1.5291,
|
| 3880 |
+
"step": 553
|
| 3881 |
+
},
|
| 3882 |
+
{
|
| 3883 |
+
"epoch": 0.9937219730941704,
|
| 3884 |
+
"grad_norm": 0.010394555516541004,
|
| 3885 |
+
"learning_rate": 2e-05,
|
| 3886 |
+
"loss": 1.5109,
|
| 3887 |
+
"step": 554
|
| 3888 |
+
},
|
| 3889 |
+
{
|
| 3890 |
+
"epoch": 0.9955156950672646,
|
| 3891 |
+
"grad_norm": 0.00996735692024231,
|
| 3892 |
+
"learning_rate": 2e-05,
|
| 3893 |
+
"loss": 1.5212,
|
| 3894 |
+
"step": 555
|
| 3895 |
+
},
|
| 3896 |
+
{
|
| 3897 |
+
"epoch": 0.9973094170403587,
|
| 3898 |
+
"grad_norm": 0.010095257312059402,
|
| 3899 |
+
"learning_rate": 2e-05,
|
| 3900 |
+
"loss": 1.5106,
|
| 3901 |
+
"step": 556
|
| 3902 |
+
},
|
| 3903 |
+
{
|
| 3904 |
+
"epoch": 0.9991031390134529,
|
| 3905 |
+
"grad_norm": 0.01082176435738802,
|
| 3906 |
+
"learning_rate": 2e-05,
|
| 3907 |
+
"loss": 1.5099,
|
| 3908 |
+
"step": 557
|
| 3909 |
+
},
|
| 3910 |
+
{
|
| 3911 |
+
"epoch": 0.9991031390134529,
|
| 3912 |
+
"step": 557,
|
| 3913 |
+
"total_flos": 7841700554735616.0,
|
| 3914 |
+
"train_loss": 0.4275342236729456,
|
| 3915 |
+
"train_runtime": 63728.3403,
|
| 3916 |
+
"train_samples_per_second": 2.239,
|
| 3917 |
+
"train_steps_per_second": 0.009
|
| 3918 |
}
|
| 3919 |
],
|
| 3920 |
"logging_steps": 1,
|
| 3921 |
+
"max_steps": 557,
|
| 3922 |
"num_input_tokens_seen": 0,
|
| 3923 |
"num_train_epochs": 1,
|
| 3924 |
"save_steps": 100,
|
|
|
|
| 3934 |
"attributes": {}
|
| 3935 |
}
|
| 3936 |
},
|
| 3937 |
+
"total_flos": 7841700554735616.0,
|
| 3938 |
"train_batch_size": 2,
|
| 3939 |
"trial_name": null,
|
| 3940 |
"trial_params": null
|
training_loss.png
CHANGED
|
|