Upload multi_student_without_cot/checkpoint-3600/trainer_state.json with huggingface_hub
Browse files
multi_student_without_cot/checkpoint-3600/trainer_state.json
ADDED
|
@@ -0,0 +1,2554 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.7094017094017095,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.004748338081671415,
|
| 14 |
+
"grad_norm": 20.21827706063475,
|
| 15 |
+
"learning_rate": 8.547008547008548e-07,
|
| 16 |
+
"loss": 8.145,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00949667616334283,
|
| 21 |
+
"grad_norm": 18.24167682921799,
|
| 22 |
+
"learning_rate": 1.8043684710351376e-06,
|
| 23 |
+
"loss": 8.0172,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.014245014245014245,
|
| 28 |
+
"grad_norm": 11.789191102542203,
|
| 29 |
+
"learning_rate": 2.7540360873694207e-06,
|
| 30 |
+
"loss": 7.3368,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.01899335232668566,
|
| 35 |
+
"grad_norm": 7.392323142404382,
|
| 36 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 37 |
+
"loss": 6.3843,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.023741690408357077,
|
| 42 |
+
"grad_norm": 5.964800407301951,
|
| 43 |
+
"learning_rate": 4.653371320037987e-06,
|
| 44 |
+
"loss": 5.5837,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02849002849002849,
|
| 49 |
+
"grad_norm": 7.120458677127898,
|
| 50 |
+
"learning_rate": 5.60303893637227e-06,
|
| 51 |
+
"loss": 4.8579,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.03323836657169991,
|
| 56 |
+
"grad_norm": 8.344343007491538,
|
| 57 |
+
"learning_rate": 6.5527065527065525e-06,
|
| 58 |
+
"loss": 3.7314,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03798670465337132,
|
| 63 |
+
"grad_norm": 3.1169502061629655,
|
| 64 |
+
"learning_rate": 7.5023741690408365e-06,
|
| 65 |
+
"loss": 2.5812,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.042735042735042736,
|
| 70 |
+
"grad_norm": 2.157035593083223,
|
| 71 |
+
"learning_rate": 8.452041785375119e-06,
|
| 72 |
+
"loss": 2.0913,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04748338081671415,
|
| 77 |
+
"grad_norm": 1.877682319517242,
|
| 78 |
+
"learning_rate": 9.401709401709402e-06,
|
| 79 |
+
"loss": 1.9562,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.052231718898385564,
|
| 84 |
+
"grad_norm": 2.6870548427576004,
|
| 85 |
+
"learning_rate": 1.0351377018043685e-05,
|
| 86 |
+
"loss": 1.8719,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.05698005698005698,
|
| 91 |
+
"grad_norm": 4.980475392094001,
|
| 92 |
+
"learning_rate": 1.1301044634377968e-05,
|
| 93 |
+
"loss": 1.7976,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.06172839506172839,
|
| 98 |
+
"grad_norm": 8.145965181044149,
|
| 99 |
+
"learning_rate": 1.2250712250712251e-05,
|
| 100 |
+
"loss": 1.7596,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.06647673314339982,
|
| 105 |
+
"grad_norm": 10.041185447423977,
|
| 106 |
+
"learning_rate": 1.3200379867046534e-05,
|
| 107 |
+
"loss": 1.7159,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.07122507122507123,
|
| 112 |
+
"grad_norm": 10.11622125534401,
|
| 113 |
+
"learning_rate": 1.4150047483380819e-05,
|
| 114 |
+
"loss": 1.6898,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.07597340930674264,
|
| 119 |
+
"grad_norm": 8.736601880824086,
|
| 120 |
+
"learning_rate": 1.50997150997151e-05,
|
| 121 |
+
"loss": 1.6485,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.08072174738841406,
|
| 126 |
+
"grad_norm": 13.03281077165562,
|
| 127 |
+
"learning_rate": 1.604938271604938e-05,
|
| 128 |
+
"loss": 1.6175,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.08547008547008547,
|
| 133 |
+
"grad_norm": 9.908577710414223,
|
| 134 |
+
"learning_rate": 1.6999050332383668e-05,
|
| 135 |
+
"loss": 1.5976,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.09021842355175688,
|
| 140 |
+
"grad_norm": 9.237568820174022,
|
| 141 |
+
"learning_rate": 1.794871794871795e-05,
|
| 142 |
+
"loss": 1.5488,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0949667616334283,
|
| 147 |
+
"grad_norm": 4.68415488864487,
|
| 148 |
+
"learning_rate": 1.889838556505223e-05,
|
| 149 |
+
"loss": 1.5014,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.09971509971509972,
|
| 154 |
+
"grad_norm": 4.844935709898889,
|
| 155 |
+
"learning_rate": 1.9848053181386517e-05,
|
| 156 |
+
"loss": 1.4854,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.10446343779677113,
|
| 161 |
+
"grad_norm": 4.323562398013845,
|
| 162 |
+
"learning_rate": 2.07977207977208e-05,
|
| 163 |
+
"loss": 1.4586,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.10921177587844255,
|
| 168 |
+
"grad_norm": 3.4385277344372978,
|
| 169 |
+
"learning_rate": 2.174738841405508e-05,
|
| 170 |
+
"loss": 1.4301,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.11396011396011396,
|
| 175 |
+
"grad_norm": 3.077116780247069,
|
| 176 |
+
"learning_rate": 2.2697056030389367e-05,
|
| 177 |
+
"loss": 1.3928,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.11870845204178537,
|
| 182 |
+
"grad_norm": 4.302072263978756,
|
| 183 |
+
"learning_rate": 2.364672364672365e-05,
|
| 184 |
+
"loss": 1.3771,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.12345679012345678,
|
| 189 |
+
"grad_norm": 3.5795185303541572,
|
| 190 |
+
"learning_rate": 2.459639126305793e-05,
|
| 191 |
+
"loss": 1.3484,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.1282051282051282,
|
| 196 |
+
"grad_norm": 2.071897753499616,
|
| 197 |
+
"learning_rate": 2.5546058879392216e-05,
|
| 198 |
+
"loss": 1.3107,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.13295346628679963,
|
| 203 |
+
"grad_norm": 2.421429428155855,
|
| 204 |
+
"learning_rate": 2.64957264957265e-05,
|
| 205 |
+
"loss": 1.2881,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.13770180436847104,
|
| 210 |
+
"grad_norm": 3.700036964420459,
|
| 211 |
+
"learning_rate": 2.744539411206078e-05,
|
| 212 |
+
"loss": 1.2667,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.14245014245014245,
|
| 217 |
+
"grad_norm": 2.8958286581854944,
|
| 218 |
+
"learning_rate": 2.839506172839506e-05,
|
| 219 |
+
"loss": 1.247,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.14719848053181386,
|
| 224 |
+
"grad_norm": 2.054414468223543,
|
| 225 |
+
"learning_rate": 2.9344729344729345e-05,
|
| 226 |
+
"loss": 1.2217,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.15194681861348527,
|
| 231 |
+
"grad_norm": 1.96235600963249,
|
| 232 |
+
"learning_rate": 3.029439696106363e-05,
|
| 233 |
+
"loss": 1.2,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.15669515669515668,
|
| 238 |
+
"grad_norm": 2.731627929284309,
|
| 239 |
+
"learning_rate": 3.1244064577397914e-05,
|
| 240 |
+
"loss": 1.1817,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.16144349477682812,
|
| 245 |
+
"grad_norm": 2.1653241264851637,
|
| 246 |
+
"learning_rate": 3.2193732193732194e-05,
|
| 247 |
+
"loss": 1.1623,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.16619183285849953,
|
| 252 |
+
"grad_norm": 2.519530114010952,
|
| 253 |
+
"learning_rate": 3.314339981006648e-05,
|
| 254 |
+
"loss": 1.1288,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.17094017094017094,
|
| 259 |
+
"grad_norm": 2.653996907642351,
|
| 260 |
+
"learning_rate": 3.409306742640076e-05,
|
| 261 |
+
"loss": 1.1081,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.17568850902184235,
|
| 266 |
+
"grad_norm": 2.4977140265839126,
|
| 267 |
+
"learning_rate": 3.504273504273504e-05,
|
| 268 |
+
"loss": 1.0854,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.18043684710351376,
|
| 273 |
+
"grad_norm": 2.856768490699816,
|
| 274 |
+
"learning_rate": 3.599240265906933e-05,
|
| 275 |
+
"loss": 1.0636,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.18518518518518517,
|
| 280 |
+
"grad_norm": 2.17649907858301,
|
| 281 |
+
"learning_rate": 3.694207027540361e-05,
|
| 282 |
+
"loss": 1.0478,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.1899335232668566,
|
| 287 |
+
"grad_norm": 3.9461445509883593,
|
| 288 |
+
"learning_rate": 3.789173789173789e-05,
|
| 289 |
+
"loss": 1.0281,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.19468186134852802,
|
| 294 |
+
"grad_norm": 2.871728182164183,
|
| 295 |
+
"learning_rate": 3.884140550807218e-05,
|
| 296 |
+
"loss": 0.9991,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.19943019943019943,
|
| 301 |
+
"grad_norm": 3.013330420912045,
|
| 302 |
+
"learning_rate": 3.979107312440646e-05,
|
| 303 |
+
"loss": 0.9773,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.20417853751187084,
|
| 308 |
+
"grad_norm": 2.420232752543592,
|
| 309 |
+
"learning_rate": 4.074074074074074e-05,
|
| 310 |
+
"loss": 0.9586,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.20892687559354226,
|
| 315 |
+
"grad_norm": 2.5768091814664564,
|
| 316 |
+
"learning_rate": 4.169040835707503e-05,
|
| 317 |
+
"loss": 0.938,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.21367521367521367,
|
| 322 |
+
"grad_norm": 3.113768661249082,
|
| 323 |
+
"learning_rate": 4.264007597340931e-05,
|
| 324 |
+
"loss": 0.9293,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.2184235517568851,
|
| 329 |
+
"grad_norm": 2.294957751769951,
|
| 330 |
+
"learning_rate": 4.358974358974359e-05,
|
| 331 |
+
"loss": 0.9049,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.22317188983855651,
|
| 336 |
+
"grad_norm": 2.4492691119209855,
|
| 337 |
+
"learning_rate": 4.453941120607788e-05,
|
| 338 |
+
"loss": 0.8865,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.22792022792022792,
|
| 343 |
+
"grad_norm": 2.286177237412566,
|
| 344 |
+
"learning_rate": 4.548907882241216e-05,
|
| 345 |
+
"loss": 0.866,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.23266856600189934,
|
| 350 |
+
"grad_norm": 2.589513226874877,
|
| 351 |
+
"learning_rate": 4.643874643874644e-05,
|
| 352 |
+
"loss": 0.8457,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.23741690408357075,
|
| 357 |
+
"grad_norm": 3.0229688171419875,
|
| 358 |
+
"learning_rate": 4.738841405508073e-05,
|
| 359 |
+
"loss": 0.8279,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.24216524216524216,
|
| 364 |
+
"grad_norm": 2.686293470195147,
|
| 365 |
+
"learning_rate": 4.833808167141501e-05,
|
| 366 |
+
"loss": 0.8147,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.24691358024691357,
|
| 371 |
+
"grad_norm": 2.17482173902714,
|
| 372 |
+
"learning_rate": 4.928774928774929e-05,
|
| 373 |
+
"loss": 0.7885,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.251661918328585,
|
| 378 |
+
"grad_norm": 2.8541347160003783,
|
| 379 |
+
"learning_rate": 5.0237416904083576e-05,
|
| 380 |
+
"loss": 0.7784,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.2564102564102564,
|
| 385 |
+
"grad_norm": 2.59370867139089,
|
| 386 |
+
"learning_rate": 5.118708452041786e-05,
|
| 387 |
+
"loss": 0.7673,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.2611585944919278,
|
| 392 |
+
"grad_norm": 3.1513543951371905,
|
| 393 |
+
"learning_rate": 5.213675213675214e-05,
|
| 394 |
+
"loss": 0.7479,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.26590693257359926,
|
| 399 |
+
"grad_norm": 2.536640604487201,
|
| 400 |
+
"learning_rate": 5.308641975308642e-05,
|
| 401 |
+
"loss": 0.7253,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.2706552706552707,
|
| 406 |
+
"grad_norm": 2.8005849730532213,
|
| 407 |
+
"learning_rate": 5.403608736942071e-05,
|
| 408 |
+
"loss": 0.7154,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.2754036087369421,
|
| 413 |
+
"grad_norm": 2.551754718587974,
|
| 414 |
+
"learning_rate": 5.498575498575499e-05,
|
| 415 |
+
"loss": 0.6901,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2801519468186135,
|
| 420 |
+
"grad_norm": 2.6731387859043365,
|
| 421 |
+
"learning_rate": 5.5935422602089274e-05,
|
| 422 |
+
"loss": 0.6784,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.2849002849002849,
|
| 427 |
+
"grad_norm": 2.558888468671125,
|
| 428 |
+
"learning_rate": 5.6885090218423554e-05,
|
| 429 |
+
"loss": 0.6634,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.2896486229819563,
|
| 434 |
+
"grad_norm": 2.7304234847433477,
|
| 435 |
+
"learning_rate": 5.7834757834757834e-05,
|
| 436 |
+
"loss": 0.6528,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.2943969610636277,
|
| 441 |
+
"grad_norm": 2.469120865455278,
|
| 442 |
+
"learning_rate": 5.878442545109212e-05,
|
| 443 |
+
"loss": 0.6357,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.29914529914529914,
|
| 448 |
+
"grad_norm": 2.361617168312385,
|
| 449 |
+
"learning_rate": 5.97340930674264e-05,
|
| 450 |
+
"loss": 0.6184,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.30389363722697055,
|
| 455 |
+
"grad_norm": 2.5062066018611544,
|
| 456 |
+
"learning_rate": 6.068376068376068e-05,
|
| 457 |
+
"loss": 0.6016,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.30864197530864196,
|
| 462 |
+
"grad_norm": 3.35086067384153,
|
| 463 |
+
"learning_rate": 6.163342830009497e-05,
|
| 464 |
+
"loss": 0.5978,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.31339031339031337,
|
| 469 |
+
"grad_norm": 2.177019547847069,
|
| 470 |
+
"learning_rate": 6.258309591642925e-05,
|
| 471 |
+
"loss": 0.5774,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.3181386514719848,
|
| 476 |
+
"grad_norm": 2.6347133971158465,
|
| 477 |
+
"learning_rate": 6.353276353276355e-05,
|
| 478 |
+
"loss": 0.5683,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.32288698955365625,
|
| 483 |
+
"grad_norm": 1.9226802841170278,
|
| 484 |
+
"learning_rate": 6.448243114909783e-05,
|
| 485 |
+
"loss": 0.5504,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.32763532763532766,
|
| 490 |
+
"grad_norm": 1.9462440337757687,
|
| 491 |
+
"learning_rate": 6.54320987654321e-05,
|
| 492 |
+
"loss": 0.5393,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.33238366571699907,
|
| 497 |
+
"grad_norm": 2.64699999142231,
|
| 498 |
+
"learning_rate": 6.638176638176638e-05,
|
| 499 |
+
"loss": 0.5295,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.3371320037986705,
|
| 504 |
+
"grad_norm": 2.428317113805517,
|
| 505 |
+
"learning_rate": 6.733143399810066e-05,
|
| 506 |
+
"loss": 0.5172,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.3418803418803419,
|
| 511 |
+
"grad_norm": 2.2288280010027854,
|
| 512 |
+
"learning_rate": 6.828110161443494e-05,
|
| 513 |
+
"loss": 0.5013,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.3466286799620133,
|
| 518 |
+
"grad_norm": 2.417506087683697,
|
| 519 |
+
"learning_rate": 6.923076923076924e-05,
|
| 520 |
+
"loss": 0.494,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.3513770180436847,
|
| 525 |
+
"grad_norm": 1.9466452806944854,
|
| 526 |
+
"learning_rate": 7.018043684710352e-05,
|
| 527 |
+
"loss": 0.4766,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.3561253561253561,
|
| 532 |
+
"grad_norm": 2.548627924889759,
|
| 533 |
+
"learning_rate": 7.11301044634378e-05,
|
| 534 |
+
"loss": 0.4684,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.36087369420702753,
|
| 539 |
+
"grad_norm": 2.157265574032032,
|
| 540 |
+
"learning_rate": 7.207977207977208e-05,
|
| 541 |
+
"loss": 0.4606,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.36562203228869894,
|
| 546 |
+
"grad_norm": 2.037117411340044,
|
| 547 |
+
"learning_rate": 7.302943969610636e-05,
|
| 548 |
+
"loss": 0.4485,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.37037037037037035,
|
| 553 |
+
"grad_norm": 1.8379017170113745,
|
| 554 |
+
"learning_rate": 7.397910731244065e-05,
|
| 555 |
+
"loss": 0.4338,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.37511870845204176,
|
| 560 |
+
"grad_norm": 2.8763875380181734,
|
| 561 |
+
"learning_rate": 7.492877492877494e-05,
|
| 562 |
+
"loss": 0.4289,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.3798670465337132,
|
| 567 |
+
"grad_norm": 1.8515635424892907,
|
| 568 |
+
"learning_rate": 7.587844254510922e-05,
|
| 569 |
+
"loss": 0.4179,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.38461538461538464,
|
| 574 |
+
"grad_norm": 1.6717226940996468,
|
| 575 |
+
"learning_rate": 7.68281101614435e-05,
|
| 576 |
+
"loss": 0.4092,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.38936372269705605,
|
| 581 |
+
"grad_norm": 2.081127977320627,
|
| 582 |
+
"learning_rate": 7.777777777777778e-05,
|
| 583 |
+
"loss": 0.3973,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.39411206077872746,
|
| 588 |
+
"grad_norm": 1.8193841823557135,
|
| 589 |
+
"learning_rate": 7.872744539411206e-05,
|
| 590 |
+
"loss": 0.3905,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.39886039886039887,
|
| 595 |
+
"grad_norm": 2.029853671008697,
|
| 596 |
+
"learning_rate": 7.967711301044634e-05,
|
| 597 |
+
"loss": 0.3829,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.4036087369420703,
|
| 602 |
+
"grad_norm": 1.7467553830120104,
|
| 603 |
+
"learning_rate": 8.062678062678063e-05,
|
| 604 |
+
"loss": 0.3705,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.4083570750237417,
|
| 609 |
+
"grad_norm": 1.8021994967000676,
|
| 610 |
+
"learning_rate": 8.157644824311491e-05,
|
| 611 |
+
"loss": 0.3639,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.4131054131054131,
|
| 616 |
+
"grad_norm": 1.641013312643542,
|
| 617 |
+
"learning_rate": 8.25261158594492e-05,
|
| 618 |
+
"loss": 0.3608,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.4178537511870845,
|
| 623 |
+
"grad_norm": 1.5079068085978562,
|
| 624 |
+
"learning_rate": 8.347578347578347e-05,
|
| 625 |
+
"loss": 0.3522,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.4226020892687559,
|
| 630 |
+
"grad_norm": 1.4655692445142585,
|
| 631 |
+
"learning_rate": 8.442545109211775e-05,
|
| 632 |
+
"loss": 0.3443,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.42735042735042733,
|
| 637 |
+
"grad_norm": 1.4792862368111255,
|
| 638 |
+
"learning_rate": 8.537511870845205e-05,
|
| 639 |
+
"loss": 0.3373,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.43209876543209874,
|
| 644 |
+
"grad_norm": 1.7589212642036423,
|
| 645 |
+
"learning_rate": 8.632478632478634e-05,
|
| 646 |
+
"loss": 0.3349,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.4368471035137702,
|
| 651 |
+
"grad_norm": 1.3662339866574285,
|
| 652 |
+
"learning_rate": 8.727445394112062e-05,
|
| 653 |
+
"loss": 0.3264,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.4415954415954416,
|
| 658 |
+
"grad_norm": 1.6091099458218114,
|
| 659 |
+
"learning_rate": 8.82241215574549e-05,
|
| 660 |
+
"loss": 0.3181,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.44634377967711303,
|
| 665 |
+
"grad_norm": 1.347407542639701,
|
| 666 |
+
"learning_rate": 8.917378917378918e-05,
|
| 667 |
+
"loss": 0.3095,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.45109211775878444,
|
| 672 |
+
"grad_norm": 1.9418517203504135,
|
| 673 |
+
"learning_rate": 9.012345679012346e-05,
|
| 674 |
+
"loss": 0.305,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.45584045584045585,
|
| 679 |
+
"grad_norm": 1.3400978709493576,
|
| 680 |
+
"learning_rate": 9.107312440645774e-05,
|
| 681 |
+
"loss": 0.3008,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.46058879392212726,
|
| 686 |
+
"grad_norm": 1.2579527847439156,
|
| 687 |
+
"learning_rate": 9.202279202279203e-05,
|
| 688 |
+
"loss": 0.2953,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.46533713200379867,
|
| 693 |
+
"grad_norm": 1.2589304750507126,
|
| 694 |
+
"learning_rate": 9.297245963912631e-05,
|
| 695 |
+
"loss": 0.2898,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.4700854700854701,
|
| 700 |
+
"grad_norm": 1.3117037531826747,
|
| 701 |
+
"learning_rate": 9.392212725546059e-05,
|
| 702 |
+
"loss": 0.2843,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4748338081671415,
|
| 707 |
+
"grad_norm": 1.3291209884932123,
|
| 708 |
+
"learning_rate": 9.487179487179487e-05,
|
| 709 |
+
"loss": 0.2815,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.4795821462488129,
|
| 714 |
+
"grad_norm": 1.209968301390431,
|
| 715 |
+
"learning_rate": 9.582146248812915e-05,
|
| 716 |
+
"loss": 0.2799,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4843304843304843,
|
| 721 |
+
"grad_norm": 1.0056322036392946,
|
| 722 |
+
"learning_rate": 9.677113010446344e-05,
|
| 723 |
+
"loss": 0.275,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.4890788224121557,
|
| 728 |
+
"grad_norm": 1.2702143901456204,
|
| 729 |
+
"learning_rate": 9.772079772079774e-05,
|
| 730 |
+
"loss": 0.2691,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.49382716049382713,
|
| 735 |
+
"grad_norm": 1.0226410737222107,
|
| 736 |
+
"learning_rate": 9.867046533713202e-05,
|
| 737 |
+
"loss": 0.2616,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4985754985754986,
|
| 742 |
+
"grad_norm": 0.9723093234963888,
|
| 743 |
+
"learning_rate": 9.96201329534663e-05,
|
| 744 |
+
"loss": 0.2579,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.50332383665717,
|
| 749 |
+
"grad_norm": 1.3912534326159531,
|
| 750 |
+
"learning_rate": 9.999990109907572e-05,
|
| 751 |
+
"loss": 0.2551,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.5080721747388414,
|
| 756 |
+
"grad_norm": 1.1085969512788294,
|
| 757 |
+
"learning_rate": 9.999929670595522e-05,
|
| 758 |
+
"loss": 0.2531,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.5128205128205128,
|
| 763 |
+
"grad_norm": 1.0552909890539075,
|
| 764 |
+
"learning_rate": 9.99981428713058e-05,
|
| 765 |
+
"loss": 0.2483,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.5175688509021842,
|
| 770 |
+
"grad_norm": 1.1040506601466529,
|
| 771 |
+
"learning_rate": 9.999643960780693e-05,
|
| 772 |
+
"loss": 0.2423,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.5223171889838556,
|
| 777 |
+
"grad_norm": 0.9501946984359841,
|
| 778 |
+
"learning_rate": 9.999418693417573e-05,
|
| 779 |
+
"loss": 0.2376,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.5270655270655271,
|
| 784 |
+
"grad_norm": 1.022708548136873,
|
| 785 |
+
"learning_rate": 9.999138487516681e-05,
|
| 786 |
+
"loss": 0.2343,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.5318138651471985,
|
| 791 |
+
"grad_norm": 1.1751853468010314,
|
| 792 |
+
"learning_rate": 9.998803346157195e-05,
|
| 793 |
+
"loss": 0.2362,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.5365622032288699,
|
| 798 |
+
"grad_norm": 0.8927081511146825,
|
| 799 |
+
"learning_rate": 9.99841327302198e-05,
|
| 800 |
+
"loss": 0.2251,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.5413105413105413,
|
| 805 |
+
"grad_norm": 0.9732415935122696,
|
| 806 |
+
"learning_rate": 9.997968272397542e-05,
|
| 807 |
+
"loss": 0.2222,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.5460588793922128,
|
| 812 |
+
"grad_norm": 0.9560857920847607,
|
| 813 |
+
"learning_rate": 9.997468349173985e-05,
|
| 814 |
+
"loss": 0.2189,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.5508072174738842,
|
| 819 |
+
"grad_norm": 0.924098382546985,
|
| 820 |
+
"learning_rate": 9.99691350884496e-05,
|
| 821 |
+
"loss": 0.2179,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.5555555555555556,
|
| 826 |
+
"grad_norm": 0.828396782450117,
|
| 827 |
+
"learning_rate": 9.996303757507602e-05,
|
| 828 |
+
"loss": 0.211,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.560303893637227,
|
| 833 |
+
"grad_norm": 0.9333663244449117,
|
| 834 |
+
"learning_rate": 9.995639101862453e-05,
|
| 835 |
+
"loss": 0.2102,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5650522317188984,
|
| 840 |
+
"grad_norm": 0.9298060626900368,
|
| 841 |
+
"learning_rate": 9.994919549213408e-05,
|
| 842 |
+
"loss": 0.2074,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.5698005698005698,
|
| 847 |
+
"grad_norm": 0.7962626517135125,
|
| 848 |
+
"learning_rate": 9.99414510746762e-05,
|
| 849 |
+
"loss": 0.203,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.5745489078822412,
|
| 854 |
+
"grad_norm": 0.8580673639697189,
|
| 855 |
+
"learning_rate": 9.993315785135416e-05,
|
| 856 |
+
"loss": 0.2013,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.5792972459639126,
|
| 861 |
+
"grad_norm": 0.8048698470169939,
|
| 862 |
+
"learning_rate": 9.99243159133021e-05,
|
| 863 |
+
"loss": 0.1975,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.584045584045584,
|
| 868 |
+
"grad_norm": 0.9257889896229621,
|
| 869 |
+
"learning_rate": 9.991492535768392e-05,
|
| 870 |
+
"loss": 0.1946,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5887939221272555,
|
| 875 |
+
"grad_norm": 0.7558640985069887,
|
| 876 |
+
"learning_rate": 9.99049862876923e-05,
|
| 877 |
+
"loss": 0.1925,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5935422602089269,
|
| 882 |
+
"grad_norm": 0.8056728442959841,
|
| 883 |
+
"learning_rate": 9.989449881254754e-05,
|
| 884 |
+
"loss": 0.1891,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5982905982905983,
|
| 889 |
+
"grad_norm": 0.8438539837910541,
|
| 890 |
+
"learning_rate": 9.988346304749635e-05,
|
| 891 |
+
"loss": 0.1862,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.6030389363722697,
|
| 896 |
+
"grad_norm": 0.7510354369473087,
|
| 897 |
+
"learning_rate": 9.987187911381059e-05,
|
| 898 |
+
"loss": 0.1841,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.6077872744539411,
|
| 903 |
+
"grad_norm": 0.7636415261243265,
|
| 904 |
+
"learning_rate": 9.985974713878592e-05,
|
| 905 |
+
"loss": 0.1808,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.6125356125356125,
|
| 910 |
+
"grad_norm": 0.7974385042371772,
|
| 911 |
+
"learning_rate": 9.984706725574044e-05,
|
| 912 |
+
"loss": 0.1792,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.6172839506172839,
|
| 917 |
+
"grad_norm": 0.7789431577859948,
|
| 918 |
+
"learning_rate": 9.983383960401319e-05,
|
| 919 |
+
"loss": 0.1756,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.6220322886989553,
|
| 924 |
+
"grad_norm": 0.7722426337147819,
|
| 925 |
+
"learning_rate": 9.982006432896266e-05,
|
| 926 |
+
"loss": 0.176,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.6267806267806267,
|
| 931 |
+
"grad_norm": 0.7383737481941087,
|
| 932 |
+
"learning_rate": 9.980574158196512e-05,
|
| 933 |
+
"loss": 0.1723,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.6315289648622981,
|
| 938 |
+
"grad_norm": 0.7442324964032874,
|
| 939 |
+
"learning_rate": 9.979087152041304e-05,
|
| 940 |
+
"loss": 0.1698,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.6362773029439696,
|
| 945 |
+
"grad_norm": 0.7497638974160803,
|
| 946 |
+
"learning_rate": 9.977545430771332e-05,
|
| 947 |
+
"loss": 0.1673,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.6410256410256411,
|
| 952 |
+
"grad_norm": 0.8548555257280125,
|
| 953 |
+
"learning_rate": 9.975949011328542e-05,
|
| 954 |
+
"loss": 0.1636,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.6457739791073125,
|
| 959 |
+
"grad_norm": 0.8279981954343867,
|
| 960 |
+
"learning_rate": 9.97429791125597e-05,
|
| 961 |
+
"loss": 0.1655,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.6505223171889839,
|
| 966 |
+
"grad_norm": 0.6959990480791529,
|
| 967 |
+
"learning_rate": 9.972592148697528e-05,
|
| 968 |
+
"loss": 0.1612,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.6552706552706553,
|
| 973 |
+
"grad_norm": 0.828078740940128,
|
| 974 |
+
"learning_rate": 9.970831742397815e-05,
|
| 975 |
+
"loss": 0.1612,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.6600189933523267,
|
| 980 |
+
"grad_norm": 0.77793285129046,
|
| 981 |
+
"learning_rate": 9.969016711701908e-05,
|
| 982 |
+
"loss": 0.1611,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.6647673314339981,
|
| 987 |
+
"grad_norm": 0.7194789618958793,
|
| 988 |
+
"learning_rate": 9.967147076555157e-05,
|
| 989 |
+
"loss": 0.1579,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.6695156695156695,
|
| 994 |
+
"grad_norm": 0.7100892012167056,
|
| 995 |
+
"learning_rate": 9.965222857502951e-05,
|
| 996 |
+
"loss": 0.1564,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.674264007597341,
|
| 1001 |
+
"grad_norm": 0.7535295766221981,
|
| 1002 |
+
"learning_rate": 9.963244075690505e-05,
|
| 1003 |
+
"loss": 0.1519,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6790123456790124,
|
| 1008 |
+
"grad_norm": 0.706459243665132,
|
| 1009 |
+
"learning_rate": 9.961210752862626e-05,
|
| 1010 |
+
"loss": 0.1548,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6837606837606838,
|
| 1015 |
+
"grad_norm": 0.6799536660584122,
|
| 1016 |
+
"learning_rate": 9.959122911363463e-05,
|
| 1017 |
+
"loss": 0.1489,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.6885090218423552,
|
| 1022 |
+
"grad_norm": 0.750734170664292,
|
| 1023 |
+
"learning_rate": 9.956980574136281e-05,
|
| 1024 |
+
"loss": 0.1492,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.6932573599240266,
|
| 1029 |
+
"grad_norm": 0.7631275022947855,
|
| 1030 |
+
"learning_rate": 9.954783764723188e-05,
|
| 1031 |
+
"loss": 0.1469,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.698005698005698,
|
| 1036 |
+
"grad_norm": 0.7140096944520468,
|
| 1037 |
+
"learning_rate": 9.952532507264891e-05,
|
| 1038 |
+
"loss": 0.1456,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.7027540360873694,
|
| 1043 |
+
"grad_norm": 0.6820968891017294,
|
| 1044 |
+
"learning_rate": 9.950226826500426e-05,
|
| 1045 |
+
"loss": 0.1459,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.7075023741690408,
|
| 1050 |
+
"grad_norm": 0.6573844442119157,
|
| 1051 |
+
"learning_rate": 9.947866747766882e-05,
|
| 1052 |
+
"loss": 0.1432,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.7122507122507122,
|
| 1057 |
+
"grad_norm": 0.6477315679395246,
|
| 1058 |
+
"learning_rate": 9.945452296999131e-05,
|
| 1059 |
+
"loss": 0.1418,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.7169990503323836,
|
| 1064 |
+
"grad_norm": 0.7854073646179602,
|
| 1065 |
+
"learning_rate": 9.942983500729536e-05,
|
| 1066 |
+
"loss": 0.1415,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.7217473884140551,
|
| 1071 |
+
"grad_norm": 0.6563280291041728,
|
| 1072 |
+
"learning_rate": 9.940460386087667e-05,
|
| 1073 |
+
"loss": 0.1396,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.7264957264957265,
|
| 1078 |
+
"grad_norm": 0.6777516445796997,
|
| 1079 |
+
"learning_rate": 9.937882980799986e-05,
|
| 1080 |
+
"loss": 0.1374,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.7312440645773979,
|
| 1085 |
+
"grad_norm": 0.6482850562773214,
|
| 1086 |
+
"learning_rate": 9.935251313189564e-05,
|
| 1087 |
+
"loss": 0.1341,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.7359924026590693,
|
| 1092 |
+
"grad_norm": 0.6807447861160003,
|
| 1093 |
+
"learning_rate": 9.93256541217576e-05,
|
| 1094 |
+
"loss": 0.1328,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.7407407407407407,
|
| 1099 |
+
"grad_norm": 0.6096762595093985,
|
| 1100 |
+
"learning_rate": 9.929825307273898e-05,
|
| 1101 |
+
"loss": 0.1323,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.7454890788224121,
|
| 1106 |
+
"grad_norm": 0.6758283907647111,
|
| 1107 |
+
"learning_rate": 9.927031028594951e-05,
|
| 1108 |
+
"loss": 0.1323,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.7502374169040835,
|
| 1113 |
+
"grad_norm": 0.6661507717312829,
|
| 1114 |
+
"learning_rate": 9.924182606845204e-05,
|
| 1115 |
+
"loss": 0.1292,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.7549857549857549,
|
| 1120 |
+
"grad_norm": 0.7190496153063343,
|
| 1121 |
+
"learning_rate": 9.921280073325922e-05,
|
| 1122 |
+
"loss": 0.1288,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.7597340930674265,
|
| 1127 |
+
"grad_norm": 0.7792128467633825,
|
| 1128 |
+
"learning_rate": 9.918323459933005e-05,
|
| 1129 |
+
"loss": 0.1315,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.7644824311490979,
|
| 1134 |
+
"grad_norm": 0.7476110366866813,
|
| 1135 |
+
"learning_rate": 9.915312799156632e-05,
|
| 1136 |
+
"loss": 0.127,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.7692307692307693,
|
| 1141 |
+
"grad_norm": 0.723682326262704,
|
| 1142 |
+
"learning_rate": 9.91224812408091e-05,
|
| 1143 |
+
"loss": 0.1252,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.7739791073124407,
|
| 1148 |
+
"grad_norm": 0.6763342918277632,
|
| 1149 |
+
"learning_rate": 9.909129468383506e-05,
|
| 1150 |
+
"loss": 0.1229,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.7787274453941121,
|
| 1155 |
+
"grad_norm": 0.6756385241786282,
|
| 1156 |
+
"learning_rate": 9.905956866335286e-05,
|
| 1157 |
+
"loss": 0.1223,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.7834757834757835,
|
| 1162 |
+
"grad_norm": 0.7031791508138145,
|
| 1163 |
+
"learning_rate": 9.90273035279992e-05,
|
| 1164 |
+
"loss": 0.121,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.7882241215574549,
|
| 1169 |
+
"grad_norm": 0.6901786661268531,
|
| 1170 |
+
"learning_rate": 9.899449963233523e-05,
|
| 1171 |
+
"loss": 0.1218,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.7929724596391263,
|
| 1176 |
+
"grad_norm": 0.655344108091334,
|
| 1177 |
+
"learning_rate": 9.896115733684244e-05,
|
| 1178 |
+
"loss": 0.1186,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7977207977207977,
|
| 1183 |
+
"grad_norm": 0.6429227026044089,
|
| 1184 |
+
"learning_rate": 9.892727700791881e-05,
|
| 1185 |
+
"loss": 0.119,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.8024691358024691,
|
| 1190 |
+
"grad_norm": 0.6445284291633588,
|
| 1191 |
+
"learning_rate": 9.889285901787482e-05,
|
| 1192 |
+
"loss": 0.1171,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.8072174738841406,
|
| 1197 |
+
"grad_norm": 0.6599500074332483,
|
| 1198 |
+
"learning_rate": 9.885790374492921e-05,
|
| 1199 |
+
"loss": 0.114,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.811965811965812,
|
| 1204 |
+
"grad_norm": 0.6458817345788224,
|
| 1205 |
+
"learning_rate": 9.882241157320497e-05,
|
| 1206 |
+
"loss": 0.1161,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.8167141500474834,
|
| 1211 |
+
"grad_norm": 0.6510623099667147,
|
| 1212 |
+
"learning_rate": 9.878638289272507e-05,
|
| 1213 |
+
"loss": 0.1134,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.8214624881291548,
|
| 1218 |
+
"grad_norm": 0.711735747979104,
|
| 1219 |
+
"learning_rate": 9.874981809940808e-05,
|
| 1220 |
+
"loss": 0.1133,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.8262108262108262,
|
| 1225 |
+
"grad_norm": 0.6067344669903363,
|
| 1226 |
+
"learning_rate": 9.871271759506404e-05,
|
| 1227 |
+
"loss": 0.111,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.8309591642924976,
|
| 1232 |
+
"grad_norm": 0.6435729954519224,
|
| 1233 |
+
"learning_rate": 9.86750817873898e-05,
|
| 1234 |
+
"loss": 0.1101,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.835707502374169,
|
| 1239 |
+
"grad_norm": 0.6594866421615497,
|
| 1240 |
+
"learning_rate": 9.863691108996471e-05,
|
| 1241 |
+
"loss": 0.1102,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.8404558404558404,
|
| 1246 |
+
"grad_norm": 0.6463645658702524,
|
| 1247 |
+
"learning_rate": 9.859820592224599e-05,
|
| 1248 |
+
"loss": 0.1086,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.8452041785375118,
|
| 1253 |
+
"grad_norm": 0.6384759072033943,
|
| 1254 |
+
"learning_rate": 9.855896670956415e-05,
|
| 1255 |
+
"loss": 0.107,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.8499525166191833,
|
| 1260 |
+
"grad_norm": 0.5925061730095992,
|
| 1261 |
+
"learning_rate": 9.851919388311832e-05,
|
| 1262 |
+
"loss": 0.1057,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.8547008547008547,
|
| 1267 |
+
"grad_norm": 0.5958838267885576,
|
| 1268 |
+
"learning_rate": 9.84788878799715e-05,
|
| 1269 |
+
"loss": 0.1063,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.8594491927825261,
|
| 1274 |
+
"grad_norm": 0.583253660596266,
|
| 1275 |
+
"learning_rate": 9.843804914304577e-05,
|
| 1276 |
+
"loss": 0.1033,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.8641975308641975,
|
| 1281 |
+
"grad_norm": 0.6509685918136744,
|
| 1282 |
+
"learning_rate": 9.839667812111739e-05,
|
| 1283 |
+
"loss": 0.1039,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.8689458689458689,
|
| 1288 |
+
"grad_norm": 0.6654859819781469,
|
| 1289 |
+
"learning_rate": 9.835477526881189e-05,
|
| 1290 |
+
"loss": 0.1026,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.8736942070275404,
|
| 1295 |
+
"grad_norm": 0.6987159643473111,
|
| 1296 |
+
"learning_rate": 9.831234104659913e-05,
|
| 1297 |
+
"loss": 0.1015,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.8784425451092118,
|
| 1302 |
+
"grad_norm": 0.5650671462210827,
|
| 1303 |
+
"learning_rate": 9.826937592078811e-05,
|
| 1304 |
+
"loss": 0.1011,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.8831908831908832,
|
| 1309 |
+
"grad_norm": 0.5727141961302766,
|
| 1310 |
+
"learning_rate": 9.822588036352201e-05,
|
| 1311 |
+
"loss": 0.1012,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.8879392212725546,
|
| 1316 |
+
"grad_norm": 0.6014103938509006,
|
| 1317 |
+
"learning_rate": 9.81818548527728e-05,
|
| 1318 |
+
"loss": 0.099,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.8926875593542261,
|
| 1323 |
+
"grad_norm": 0.5734563478604482,
|
| 1324 |
+
"learning_rate": 9.81372998723362e-05,
|
| 1325 |
+
"loss": 0.0975,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.8974358974358975,
|
| 1330 |
+
"grad_norm": 0.5684437429333495,
|
| 1331 |
+
"learning_rate": 9.809221591182623e-05,
|
| 1332 |
+
"loss": 0.0954,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.9021842355175689,
|
| 1337 |
+
"grad_norm": 0.5550460249948803,
|
| 1338 |
+
"learning_rate": 9.804660346666984e-05,
|
| 1339 |
+
"loss": 0.0948,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.9069325735992403,
|
| 1344 |
+
"grad_norm": 0.6030717817112082,
|
| 1345 |
+
"learning_rate": 9.800046303810154e-05,
|
| 1346 |
+
"loss": 0.095,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.9116809116809117,
|
| 1351 |
+
"grad_norm": 0.5786096312033695,
|
| 1352 |
+
"learning_rate": 9.795379513315779e-05,
|
| 1353 |
+
"loss": 0.0964,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.9164292497625831,
|
| 1358 |
+
"grad_norm": 0.5342317355454276,
|
| 1359 |
+
"learning_rate": 9.790660026467153e-05,
|
| 1360 |
+
"loss": 0.0917,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.9211775878442545,
|
| 1365 |
+
"grad_norm": 0.5729999943038201,
|
| 1366 |
+
"learning_rate": 9.785887895126644e-05,
|
| 1367 |
+
"loss": 0.0934,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.9259259259259259,
|
| 1372 |
+
"grad_norm": 0.5148337352548606,
|
| 1373 |
+
"learning_rate": 9.781063171735134e-05,
|
| 1374 |
+
"loss": 0.0912,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.9306742640075973,
|
| 1379 |
+
"grad_norm": 0.6096023114073594,
|
| 1380 |
+
"learning_rate": 9.776185909311437e-05,
|
| 1381 |
+
"loss": 0.0934,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.9354226020892688,
|
| 1386 |
+
"grad_norm": 0.5953694391053918,
|
| 1387 |
+
"learning_rate": 9.771256161451716e-05,
|
| 1388 |
+
"loss": 0.0915,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.9401709401709402,
|
| 1393 |
+
"grad_norm": 0.5553421636649856,
|
| 1394 |
+
"learning_rate": 9.766273982328899e-05,
|
| 1395 |
+
"loss": 0.0892,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.9449192782526116,
|
| 1400 |
+
"grad_norm": 0.6314566035516022,
|
| 1401 |
+
"learning_rate": 9.761239426692077e-05,
|
| 1402 |
+
"loss": 0.0892,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.949667616334283,
|
| 1407 |
+
"grad_norm": 0.553477341094825,
|
| 1408 |
+
"learning_rate": 9.756152549865907e-05,
|
| 1409 |
+
"loss": 0.0868,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.9544159544159544,
|
| 1414 |
+
"grad_norm": 0.5855428558450079,
|
| 1415 |
+
"learning_rate": 9.751013407750004e-05,
|
| 1416 |
+
"loss": 0.0879,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.9591642924976258,
|
| 1421 |
+
"grad_norm": 0.5564124650704858,
|
| 1422 |
+
"learning_rate": 9.745822056818324e-05,
|
| 1423 |
+
"loss": 0.0878,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.9639126305792972,
|
| 1428 |
+
"grad_norm": 0.5648708660408431,
|
| 1429 |
+
"learning_rate": 9.740578554118545e-05,
|
| 1430 |
+
"loss": 0.0851,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.9686609686609686,
|
| 1435 |
+
"grad_norm": 0.6100865208168226,
|
| 1436 |
+
"learning_rate": 9.735282957271438e-05,
|
| 1437 |
+
"loss": 0.0857,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.97340930674264,
|
| 1442 |
+
"grad_norm": 0.5342516016564061,
|
| 1443 |
+
"learning_rate": 9.729935324470243e-05,
|
| 1444 |
+
"loss": 0.0833,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.9781576448243114,
|
| 1449 |
+
"grad_norm": 0.5617810297891285,
|
| 1450 |
+
"learning_rate": 9.724535714480012e-05,
|
| 1451 |
+
"loss": 0.0823,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.9829059829059829,
|
| 1456 |
+
"grad_norm": 0.4874563065093935,
|
| 1457 |
+
"learning_rate": 9.719084186636984e-05,
|
| 1458 |
+
"loss": 0.0829,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.9876543209876543,
|
| 1463 |
+
"grad_norm": 0.5925800844469832,
|
| 1464 |
+
"learning_rate": 9.713580800847916e-05,
|
| 1465 |
+
"loss": 0.0816,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.9924026590693258,
|
| 1470 |
+
"grad_norm": 0.5192721370514558,
|
| 1471 |
+
"learning_rate": 9.708025617589431e-05,
|
| 1472 |
+
"loss": 0.0801,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.9971509971509972,
|
| 1477 |
+
"grad_norm": 0.5317846561760016,
|
| 1478 |
+
"learning_rate": 9.702418697907357e-05,
|
| 1479 |
+
"loss": 0.08,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.0018993352326686,
|
| 1484 |
+
"grad_norm": 0.5375175588207268,
|
| 1485 |
+
"learning_rate": 9.696760103416049e-05,
|
| 1486 |
+
"loss": 0.0765,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.00664767331434,
|
| 1491 |
+
"grad_norm": 0.4995120344713725,
|
| 1492 |
+
"learning_rate": 9.69104989629772e-05,
|
| 1493 |
+
"loss": 0.0686,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.0113960113960114,
|
| 1498 |
+
"grad_norm": 0.492492367724788,
|
| 1499 |
+
"learning_rate": 9.685288139301749e-05,
|
| 1500 |
+
"loss": 0.0676,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.0161443494776827,
|
| 1505 |
+
"grad_norm": 0.489910902697706,
|
| 1506 |
+
"learning_rate": 9.679474895744002e-05,
|
| 1507 |
+
"loss": 0.0673,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.0208926875593543,
|
| 1512 |
+
"grad_norm": 0.4842805102412133,
|
| 1513 |
+
"learning_rate": 9.673610229506123e-05,
|
| 1514 |
+
"loss": 0.0657,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.0256410256410255,
|
| 1519 |
+
"grad_norm": 0.500590705014481,
|
| 1520 |
+
"learning_rate": 9.667694205034842e-05,
|
| 1521 |
+
"loss": 0.065,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.030389363722697,
|
| 1526 |
+
"grad_norm": 0.5412799666786388,
|
| 1527 |
+
"learning_rate": 9.661726887341264e-05,
|
| 1528 |
+
"loss": 0.0661,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.0351377018043684,
|
| 1533 |
+
"grad_norm": 0.49596368987818673,
|
| 1534 |
+
"learning_rate": 9.655708342000158e-05,
|
| 1535 |
+
"loss": 0.066,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.03988603988604,
|
| 1540 |
+
"grad_norm": 0.5133006065434564,
|
| 1541 |
+
"learning_rate": 9.649638635149225e-05,
|
| 1542 |
+
"loss": 0.0633,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.0446343779677112,
|
| 1547 |
+
"grad_norm": 0.5079148873200521,
|
| 1548 |
+
"learning_rate": 9.643517833488387e-05,
|
| 1549 |
+
"loss": 0.0649,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.0493827160493827,
|
| 1554 |
+
"grad_norm": 0.5222502694032142,
|
| 1555 |
+
"learning_rate": 9.637346004279038e-05,
|
| 1556 |
+
"loss": 0.0635,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.0541310541310542,
|
| 1561 |
+
"grad_norm": 0.5104557333995293,
|
| 1562 |
+
"learning_rate": 9.631123215343323e-05,
|
| 1563 |
+
"loss": 0.0618,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.0588793922127255,
|
| 1568 |
+
"grad_norm": 0.5024303716694327,
|
| 1569 |
+
"learning_rate": 9.624849535063373e-05,
|
| 1570 |
+
"loss": 0.0635,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.063627730294397,
|
| 1575 |
+
"grad_norm": 0.5353909988509568,
|
| 1576 |
+
"learning_rate": 9.61852503238057e-05,
|
| 1577 |
+
"loss": 0.0623,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.0683760683760684,
|
| 1582 |
+
"grad_norm": 0.548569047624678,
|
| 1583 |
+
"learning_rate": 9.612149776794776e-05,
|
| 1584 |
+
"loss": 0.0627,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.0731244064577399,
|
| 1589 |
+
"grad_norm": 0.49061114729721145,
|
| 1590 |
+
"learning_rate": 9.605723838363586e-05,
|
| 1591 |
+
"loss": 0.0617,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.0778727445394112,
|
| 1596 |
+
"grad_norm": 0.5118467274517943,
|
| 1597 |
+
"learning_rate": 9.599247287701539e-05,
|
| 1598 |
+
"loss": 0.0613,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.0826210826210827,
|
| 1603 |
+
"grad_norm": 0.5106993610815542,
|
| 1604 |
+
"learning_rate": 9.592720195979353e-05,
|
| 1605 |
+
"loss": 0.0616,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.087369420702754,
|
| 1610 |
+
"grad_norm": 0.533554957708412,
|
| 1611 |
+
"learning_rate": 9.586142634923141e-05,
|
| 1612 |
+
"loss": 0.0607,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.0921177587844255,
|
| 1617 |
+
"grad_norm": 0.4961988723784374,
|
| 1618 |
+
"learning_rate": 9.579514676813627e-05,
|
| 1619 |
+
"loss": 0.0603,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.0968660968660968,
|
| 1624 |
+
"grad_norm": 0.5291731061070475,
|
| 1625 |
+
"learning_rate": 9.572836394485345e-05,
|
| 1626 |
+
"loss": 0.0578,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.1016144349477683,
|
| 1631 |
+
"grad_norm": 0.4560676421604984,
|
| 1632 |
+
"learning_rate": 9.566107861325836e-05,
|
| 1633 |
+
"loss": 0.0585,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.1063627730294396,
|
| 1638 |
+
"grad_norm": 0.5078075639209664,
|
| 1639 |
+
"learning_rate": 9.559329151274854e-05,
|
| 1640 |
+
"loss": 0.0578,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.1111111111111112,
|
| 1645 |
+
"grad_norm": 0.4683086272270788,
|
| 1646 |
+
"learning_rate": 9.552500338823543e-05,
|
| 1647 |
+
"loss": 0.0562,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.1158594491927825,
|
| 1652 |
+
"grad_norm": 0.49903313629075563,
|
| 1653 |
+
"learning_rate": 9.54562149901362e-05,
|
| 1654 |
+
"loss": 0.0574,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.120607787274454,
|
| 1659 |
+
"grad_norm": 0.4952267832995423,
|
| 1660 |
+
"learning_rate": 9.538692707436554e-05,
|
| 1661 |
+
"loss": 0.0566,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.1253561253561253,
|
| 1666 |
+
"grad_norm": 0.5143759315257765,
|
| 1667 |
+
"learning_rate": 9.531714040232733e-05,
|
| 1668 |
+
"loss": 0.0568,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.1301044634377968,
|
| 1673 |
+
"grad_norm": 0.5333912877931387,
|
| 1674 |
+
"learning_rate": 9.524685574090627e-05,
|
| 1675 |
+
"loss": 0.0554,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.134852801519468,
|
| 1680 |
+
"grad_norm": 0.506610426425779,
|
| 1681 |
+
"learning_rate": 9.517607386245948e-05,
|
| 1682 |
+
"loss": 0.0557,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.1396011396011396,
|
| 1687 |
+
"grad_norm": 0.4807568687194227,
|
| 1688 |
+
"learning_rate": 9.510479554480793e-05,
|
| 1689 |
+
"loss": 0.057,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.144349477682811,
|
| 1694 |
+
"grad_norm": 0.5059220071837027,
|
| 1695 |
+
"learning_rate": 9.503302157122804e-05,
|
| 1696 |
+
"loss": 0.0564,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.1490978157644824,
|
| 1701 |
+
"grad_norm": 0.5321634332501042,
|
| 1702 |
+
"learning_rate": 9.496075273044293e-05,
|
| 1703 |
+
"loss": 0.0556,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.1538461538461537,
|
| 1708 |
+
"grad_norm": 0.4860319263192551,
|
| 1709 |
+
"learning_rate": 9.488798981661379e-05,
|
| 1710 |
+
"loss": 0.0556,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.1585944919278253,
|
| 1715 |
+
"grad_norm": 0.5450716546784171,
|
| 1716 |
+
"learning_rate": 9.481473362933122e-05,
|
| 1717 |
+
"loss": 0.0549,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.1633428300094968,
|
| 1722 |
+
"grad_norm": 0.48569228023641087,
|
| 1723 |
+
"learning_rate": 9.474098497360637e-05,
|
| 1724 |
+
"loss": 0.0557,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.168091168091168,
|
| 1729 |
+
"grad_norm": 0.5215152833823479,
|
| 1730 |
+
"learning_rate": 9.466674465986211e-05,
|
| 1731 |
+
"loss": 0.0552,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.1728395061728394,
|
| 1736 |
+
"grad_norm": 0.4777704133253495,
|
| 1737 |
+
"learning_rate": 9.459201350392418e-05,
|
| 1738 |
+
"loss": 0.054,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.177587844254511,
|
| 1743 |
+
"grad_norm": 0.4825911992094268,
|
| 1744 |
+
"learning_rate": 9.451679232701212e-05,
|
| 1745 |
+
"loss": 0.0531,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.1823361823361824,
|
| 1750 |
+
"grad_norm": 0.5256696379086168,
|
| 1751 |
+
"learning_rate": 9.444108195573031e-05,
|
| 1752 |
+
"loss": 0.0546,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.1870845204178537,
|
| 1757 |
+
"grad_norm": 0.48041145273942765,
|
| 1758 |
+
"learning_rate": 9.436488322205892e-05,
|
| 1759 |
+
"loss": 0.0547,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.1918328584995252,
|
| 1764 |
+
"grad_norm": 0.49889044556441,
|
| 1765 |
+
"learning_rate": 9.42881969633447e-05,
|
| 1766 |
+
"loss": 0.052,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.1965811965811965,
|
| 1771 |
+
"grad_norm": 0.4691225975029018,
|
| 1772 |
+
"learning_rate": 9.42110240222918e-05,
|
| 1773 |
+
"loss": 0.0526,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.201329534662868,
|
| 1778 |
+
"grad_norm": 0.4699543822621789,
|
| 1779 |
+
"learning_rate": 9.413336524695253e-05,
|
| 1780 |
+
"loss": 0.0529,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.2060778727445394,
|
| 1785 |
+
"grad_norm": 0.5092846176526173,
|
| 1786 |
+
"learning_rate": 9.405522149071801e-05,
|
| 1787 |
+
"loss": 0.0515,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.210826210826211,
|
| 1792 |
+
"grad_norm": 0.48822278879655134,
|
| 1793 |
+
"learning_rate": 9.397659361230883e-05,
|
| 1794 |
+
"loss": 0.0518,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.2155745489078822,
|
| 1799 |
+
"grad_norm": 0.4709051625946915,
|
| 1800 |
+
"learning_rate": 9.389748247576557e-05,
|
| 1801 |
+
"loss": 0.0513,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.2203228869895537,
|
| 1806 |
+
"grad_norm": 0.43081313275948346,
|
| 1807 |
+
"learning_rate": 9.381788895043933e-05,
|
| 1808 |
+
"loss": 0.052,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.225071225071225,
|
| 1813 |
+
"grad_norm": 0.4618949718938529,
|
| 1814 |
+
"learning_rate": 9.373781391098218e-05,
|
| 1815 |
+
"loss": 0.0509,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.2298195631528965,
|
| 1820 |
+
"grad_norm": 0.4414468310558702,
|
| 1821 |
+
"learning_rate": 9.365725823733753e-05,
|
| 1822 |
+
"loss": 0.0505,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.2345679012345678,
|
| 1827 |
+
"grad_norm": 0.4437487857178328,
|
| 1828 |
+
"learning_rate": 9.35762228147305e-05,
|
| 1829 |
+
"loss": 0.0496,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.2393162393162394,
|
| 1834 |
+
"grad_norm": 0.4644815359436171,
|
| 1835 |
+
"learning_rate": 9.349470853365815e-05,
|
| 1836 |
+
"loss": 0.0491,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.2440645773979107,
|
| 1841 |
+
"grad_norm": 0.4604315887784035,
|
| 1842 |
+
"learning_rate": 9.341271628987968e-05,
|
| 1843 |
+
"loss": 0.0501,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.2488129154795822,
|
| 1848 |
+
"grad_norm": 0.45836637793826146,
|
| 1849 |
+
"learning_rate": 9.333024698440667e-05,
|
| 1850 |
+
"loss": 0.0491,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.2535612535612537,
|
| 1855 |
+
"grad_norm": 0.4004595257683082,
|
| 1856 |
+
"learning_rate": 9.324730152349306e-05,
|
| 1857 |
+
"loss": 0.0493,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.258309591642925,
|
| 1862 |
+
"grad_norm": 0.4588694709958647,
|
| 1863 |
+
"learning_rate": 9.31638808186253e-05,
|
| 1864 |
+
"loss": 0.0485,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.2630579297245963,
|
| 1869 |
+
"grad_norm": 0.4289599019381277,
|
| 1870 |
+
"learning_rate": 9.307998578651228e-05,
|
| 1871 |
+
"loss": 0.049,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.2678062678062678,
|
| 1876 |
+
"grad_norm": 0.4727026865591144,
|
| 1877 |
+
"learning_rate": 9.299561734907528e-05,
|
| 1878 |
+
"loss": 0.0459,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.2725546058879393,
|
| 1883 |
+
"grad_norm": 0.4229759230480915,
|
| 1884 |
+
"learning_rate": 9.291077643343776e-05,
|
| 1885 |
+
"loss": 0.0477,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.2773029439696106,
|
| 1890 |
+
"grad_norm": 0.43707322690226125,
|
| 1891 |
+
"learning_rate": 9.282546397191531e-05,
|
| 1892 |
+
"loss": 0.0474,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.282051282051282,
|
| 1897 |
+
"grad_norm": 0.47184366398017824,
|
| 1898 |
+
"learning_rate": 9.273968090200533e-05,
|
| 1899 |
+
"loss": 0.0472,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.2867996201329535,
|
| 1904 |
+
"grad_norm": 0.5027834026428316,
|
| 1905 |
+
"learning_rate": 9.265342816637666e-05,
|
| 1906 |
+
"loss": 0.0465,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.291547958214625,
|
| 1911 |
+
"grad_norm": 0.4454440035871166,
|
| 1912 |
+
"learning_rate": 9.256670671285934e-05,
|
| 1913 |
+
"loss": 0.0462,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.2962962962962963,
|
| 1918 |
+
"grad_norm": 0.43536614903814524,
|
| 1919 |
+
"learning_rate": 9.247951749443413e-05,
|
| 1920 |
+
"loss": 0.0459,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.3010446343779676,
|
| 1925 |
+
"grad_norm": 0.49496070699402434,
|
| 1926 |
+
"learning_rate": 9.239186146922208e-05,
|
| 1927 |
+
"loss": 0.0451,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.305792972459639,
|
| 1932 |
+
"grad_norm": 0.43375607265420135,
|
| 1933 |
+
"learning_rate": 9.230373960047394e-05,
|
| 1934 |
+
"loss": 0.0454,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.3105413105413106,
|
| 1939 |
+
"grad_norm": 0.42267249461132017,
|
| 1940 |
+
"learning_rate": 9.22151528565596e-05,
|
| 1941 |
+
"loss": 0.045,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.315289648622982,
|
| 1946 |
+
"grad_norm": 0.47137884223204835,
|
| 1947 |
+
"learning_rate": 9.212610221095748e-05,
|
| 1948 |
+
"loss": 0.0453,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.3200379867046534,
|
| 1953 |
+
"grad_norm": 0.5199454437668323,
|
| 1954 |
+
"learning_rate": 9.203658864224382e-05,
|
| 1955 |
+
"loss": 0.0439,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.3247863247863247,
|
| 1960 |
+
"grad_norm": 0.4570822874899115,
|
| 1961 |
+
"learning_rate": 9.194661313408187e-05,
|
| 1962 |
+
"loss": 0.044,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.3295346628679963,
|
| 1967 |
+
"grad_norm": 0.43113616092372226,
|
| 1968 |
+
"learning_rate": 9.185617667521117e-05,
|
| 1969 |
+
"loss": 0.0444,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.3342830009496676,
|
| 1974 |
+
"grad_norm": 0.4222760906355151,
|
| 1975 |
+
"learning_rate": 9.176528025943663e-05,
|
| 1976 |
+
"loss": 0.0444,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.339031339031339,
|
| 1981 |
+
"grad_norm": 0.4952066697298851,
|
| 1982 |
+
"learning_rate": 9.167392488561758e-05,
|
| 1983 |
+
"loss": 0.0433,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.3437796771130104,
|
| 1988 |
+
"grad_norm": 0.46359163249702656,
|
| 1989 |
+
"learning_rate": 9.158211155765692e-05,
|
| 1990 |
+
"loss": 0.0432,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.348528015194682,
|
| 1995 |
+
"grad_norm": 0.47893564069863104,
|
| 1996 |
+
"learning_rate": 9.14898412844899e-05,
|
| 1997 |
+
"loss": 0.0437,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.3532763532763532,
|
| 2002 |
+
"grad_norm": 0.4011060385937812,
|
| 2003 |
+
"learning_rate": 9.13971150800732e-05,
|
| 2004 |
+
"loss": 0.0414,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.3580246913580247,
|
| 2009 |
+
"grad_norm": 0.46085185858271926,
|
| 2010 |
+
"learning_rate": 9.130393396337368e-05,
|
| 2011 |
+
"loss": 0.0418,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.362773029439696,
|
| 2016 |
+
"grad_norm": 0.46535476597347625,
|
| 2017 |
+
"learning_rate": 9.121029895835731e-05,
|
| 2018 |
+
"loss": 0.042,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.3675213675213675,
|
| 2023 |
+
"grad_norm": 0.39410170446969295,
|
| 2024 |
+
"learning_rate": 9.111621109397769e-05,
|
| 2025 |
+
"loss": 0.0418,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.3722697056030388,
|
| 2030 |
+
"grad_norm": 0.4243876470935551,
|
| 2031 |
+
"learning_rate": 9.102167140416503e-05,
|
| 2032 |
+
"loss": 0.041,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.3770180436847104,
|
| 2037 |
+
"grad_norm": 0.4585829009513368,
|
| 2038 |
+
"learning_rate": 9.092668092781454e-05,
|
| 2039 |
+
"loss": 0.0414,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.381766381766382,
|
| 2044 |
+
"grad_norm": 0.4262377269345021,
|
| 2045 |
+
"learning_rate": 9.083124070877516e-05,
|
| 2046 |
+
"loss": 0.042,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.3865147198480532,
|
| 2051 |
+
"grad_norm": 0.40844548158991334,
|
| 2052 |
+
"learning_rate": 9.073535179583806e-05,
|
| 2053 |
+
"loss": 0.0419,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.3912630579297245,
|
| 2058 |
+
"grad_norm": 0.3935442259634633,
|
| 2059 |
+
"learning_rate": 9.063901524272505e-05,
|
| 2060 |
+
"loss": 0.0412,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.396011396011396,
|
| 2065 |
+
"grad_norm": 0.3693483555639858,
|
| 2066 |
+
"learning_rate": 9.054223210807709e-05,
|
| 2067 |
+
"loss": 0.0402,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.4007597340930675,
|
| 2072 |
+
"grad_norm": 0.40078308269859275,
|
| 2073 |
+
"learning_rate": 9.044500345544262e-05,
|
| 2074 |
+
"loss": 0.04,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.4055080721747388,
|
| 2079 |
+
"grad_norm": 0.414448129205789,
|
| 2080 |
+
"learning_rate": 9.034733035326584e-05,
|
| 2081 |
+
"loss": 0.0419,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.4102564102564101,
|
| 2086 |
+
"grad_norm": 0.4314293044106671,
|
| 2087 |
+
"learning_rate": 9.024921387487504e-05,
|
| 2088 |
+
"loss": 0.0395,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.4150047483380817,
|
| 2093 |
+
"grad_norm": 0.3830262611826239,
|
| 2094 |
+
"learning_rate": 9.015065509847075e-05,
|
| 2095 |
+
"loss": 0.0385,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.4197530864197532,
|
| 2100 |
+
"grad_norm": 0.40774106075840866,
|
| 2101 |
+
"learning_rate": 9.005165510711387e-05,
|
| 2102 |
+
"loss": 0.0393,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.4245014245014245,
|
| 2107 |
+
"grad_norm": 0.4072063711912796,
|
| 2108 |
+
"learning_rate": 8.995221498871387e-05,
|
| 2109 |
+
"loss": 0.0376,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.429249762583096,
|
| 2114 |
+
"grad_norm": 0.418318685699483,
|
| 2115 |
+
"learning_rate": 8.98523358360167e-05,
|
| 2116 |
+
"loss": 0.0388,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.4339981006647673,
|
| 2121 |
+
"grad_norm": 0.38188198729370343,
|
| 2122 |
+
"learning_rate": 8.975201874659292e-05,
|
| 2123 |
+
"loss": 0.0377,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.4387464387464388,
|
| 2128 |
+
"grad_norm": 0.4240963163227532,
|
| 2129 |
+
"learning_rate": 8.965126482282549e-05,
|
| 2130 |
+
"loss": 0.0394,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.4434947768281101,
|
| 2135 |
+
"grad_norm": 0.40891409921975486,
|
| 2136 |
+
"learning_rate": 8.955007517189784e-05,
|
| 2137 |
+
"loss": 0.0389,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.4482431149097816,
|
| 2142 |
+
"grad_norm": 0.4343466782916782,
|
| 2143 |
+
"learning_rate": 8.944845090578149e-05,
|
| 2144 |
+
"loss": 0.038,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.452991452991453,
|
| 2149 |
+
"grad_norm": 0.47914926317681983,
|
| 2150 |
+
"learning_rate": 8.934639314122401e-05,
|
| 2151 |
+
"loss": 0.0383,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.4577397910731245,
|
| 2156 |
+
"grad_norm": 0.4726822552336102,
|
| 2157 |
+
"learning_rate": 8.924390299973666e-05,
|
| 2158 |
+
"loss": 0.037,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.4624881291547958,
|
| 2163 |
+
"grad_norm": 0.40639896910669704,
|
| 2164 |
+
"learning_rate": 8.91409816075821e-05,
|
| 2165 |
+
"loss": 0.0379,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.4672364672364673,
|
| 2170 |
+
"grad_norm": 0.4297631622330081,
|
| 2171 |
+
"learning_rate": 8.903763009576195e-05,
|
| 2172 |
+
"loss": 0.0378,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.4719848053181386,
|
| 2177 |
+
"grad_norm": 0.397553702278778,
|
| 2178 |
+
"learning_rate": 8.893384960000445e-05,
|
| 2179 |
+
"loss": 0.0366,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.47673314339981,
|
| 2184 |
+
"grad_norm": 0.3842838408196586,
|
| 2185 |
+
"learning_rate": 8.882964126075192e-05,
|
| 2186 |
+
"loss": 0.0355,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.4814814814814814,
|
| 2191 |
+
"grad_norm": 0.4087672118416129,
|
| 2192 |
+
"learning_rate": 8.872500622314829e-05,
|
| 2193 |
+
"loss": 0.0363,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.486229819563153,
|
| 2198 |
+
"grad_norm": 0.4098031203055733,
|
| 2199 |
+
"learning_rate": 8.861994563702638e-05,
|
| 2200 |
+
"loss": 0.035,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.4909781576448244,
|
| 2205 |
+
"grad_norm": 0.3876997066775908,
|
| 2206 |
+
"learning_rate": 8.851446065689544e-05,
|
| 2207 |
+
"loss": 0.0364,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.4957264957264957,
|
| 2212 |
+
"grad_norm": 0.4134151182799379,
|
| 2213 |
+
"learning_rate": 8.840855244192836e-05,
|
| 2214 |
+
"loss": 0.0349,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.500474833808167,
|
| 2219 |
+
"grad_norm": 0.3927794557722139,
|
| 2220 |
+
"learning_rate": 8.83022221559489e-05,
|
| 2221 |
+
"loss": 0.0357,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.5052231718898386,
|
| 2226 |
+
"grad_norm": 0.41635322736234687,
|
| 2227 |
+
"learning_rate": 8.819547096741904e-05,
|
| 2228 |
+
"loss": 0.0346,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.50997150997151,
|
| 2233 |
+
"grad_norm": 0.3773849224872641,
|
| 2234 |
+
"learning_rate": 8.8088300049426e-05,
|
| 2235 |
+
"loss": 0.0342,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.5147198480531814,
|
| 2240 |
+
"grad_norm": 0.34735786246540556,
|
| 2241 |
+
"learning_rate": 8.798071057966935e-05,
|
| 2242 |
+
"loss": 0.0335,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.5194681861348527,
|
| 2247 |
+
"grad_norm": 0.39208660749913393,
|
| 2248 |
+
"learning_rate": 8.787270374044824e-05,
|
| 2249 |
+
"loss": 0.0351,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.5242165242165242,
|
| 2254 |
+
"grad_norm": 0.4048707970102346,
|
| 2255 |
+
"learning_rate": 8.776428071864819e-05,
|
| 2256 |
+
"loss": 0.0336,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.5289648622981957,
|
| 2261 |
+
"grad_norm": 0.3843214639707809,
|
| 2262 |
+
"learning_rate": 8.76554427057282e-05,
|
| 2263 |
+
"loss": 0.0338,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.533713200379867,
|
| 2268 |
+
"grad_norm": 0.4132367494817212,
|
| 2269 |
+
"learning_rate": 8.754619089770755e-05,
|
| 2270 |
+
"loss": 0.0336,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.5384615384615383,
|
| 2275 |
+
"grad_norm": 0.38075123816436185,
|
| 2276 |
+
"learning_rate": 8.743652649515275e-05,
|
| 2277 |
+
"loss": 0.0329,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.5432098765432098,
|
| 2282 |
+
"grad_norm": 0.39999216549632843,
|
| 2283 |
+
"learning_rate": 8.73264507031643e-05,
|
| 2284 |
+
"loss": 0.0345,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.5479582146248814,
|
| 2289 |
+
"grad_norm": 0.36778246346411,
|
| 2290 |
+
"learning_rate": 8.721596473136342e-05,
|
| 2291 |
+
"loss": 0.0323,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.5527065527065527,
|
| 2296 |
+
"grad_norm": 0.41572199384610614,
|
| 2297 |
+
"learning_rate": 8.710506979387882e-05,
|
| 2298 |
+
"loss": 0.0321,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.557454890788224,
|
| 2303 |
+
"grad_norm": 0.3608557161549764,
|
| 2304 |
+
"learning_rate": 8.699376710933336e-05,
|
| 2305 |
+
"loss": 0.0325,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.5622032288698955,
|
| 2310 |
+
"grad_norm": 0.33558725315518206,
|
| 2311 |
+
"learning_rate": 8.688205790083053e-05,
|
| 2312 |
+
"loss": 0.0313,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.566951566951567,
|
| 2317 |
+
"grad_norm": 0.40918172973994715,
|
| 2318 |
+
"learning_rate": 8.676994339594118e-05,
|
| 2319 |
+
"loss": 0.0319,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.5716999050332383,
|
| 2324 |
+
"grad_norm": 0.43337497124362845,
|
| 2325 |
+
"learning_rate": 8.665742482668996e-05,
|
| 2326 |
+
"loss": 0.0317,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.5764482431149098,
|
| 2331 |
+
"grad_norm": 0.3812998243057314,
|
| 2332 |
+
"learning_rate": 8.654450342954175e-05,
|
| 2333 |
+
"loss": 0.0323,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.5811965811965814,
|
| 2338 |
+
"grad_norm": 0.35509189264018565,
|
| 2339 |
+
"learning_rate": 8.643118044538808e-05,
|
| 2340 |
+
"loss": 0.0326,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.5859449192782527,
|
| 2345 |
+
"grad_norm": 0.38772936620972853,
|
| 2346 |
+
"learning_rate": 8.631745711953357e-05,
|
| 2347 |
+
"loss": 0.0317,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.590693257359924,
|
| 2352 |
+
"grad_norm": 0.4020905021794808,
|
| 2353 |
+
"learning_rate": 8.620333470168214e-05,
|
| 2354 |
+
"loss": 0.0322,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.5954415954415955,
|
| 2359 |
+
"grad_norm": 0.42014276075969054,
|
| 2360 |
+
"learning_rate": 8.608881444592337e-05,
|
| 2361 |
+
"loss": 0.0319,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.600189933523267,
|
| 2366 |
+
"grad_norm": 0.3486263773747954,
|
| 2367 |
+
"learning_rate": 8.597389761071862e-05,
|
| 2368 |
+
"loss": 0.0309,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.6049382716049383,
|
| 2373 |
+
"grad_norm": 0.3635863122166198,
|
| 2374 |
+
"learning_rate": 8.58585854588873e-05,
|
| 2375 |
+
"loss": 0.0292,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.6096866096866096,
|
| 2380 |
+
"grad_norm": 0.35660749991777047,
|
| 2381 |
+
"learning_rate": 8.574287925759294e-05,
|
| 2382 |
+
"loss": 0.0305,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.6144349477682811,
|
| 2387 |
+
"grad_norm": 0.36832555363654856,
|
| 2388 |
+
"learning_rate": 8.562678027832925e-05,
|
| 2389 |
+
"loss": 0.0302,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.6191832858499526,
|
| 2394 |
+
"grad_norm": 0.41904584869446504,
|
| 2395 |
+
"learning_rate": 8.551028979690621e-05,
|
| 2396 |
+
"loss": 0.0286,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.623931623931624,
|
| 2401 |
+
"grad_norm": 0.389795337040796,
|
| 2402 |
+
"learning_rate": 8.539340909343597e-05,
|
| 2403 |
+
"loss": 0.0298,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.6286799620132952,
|
| 2408 |
+
"grad_norm": 0.36335077372557817,
|
| 2409 |
+
"learning_rate": 8.527613945231885e-05,
|
| 2410 |
+
"loss": 0.0298,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.6334283000949668,
|
| 2415 |
+
"grad_norm": 0.34642329798347377,
|
| 2416 |
+
"learning_rate": 8.51584821622292e-05,
|
| 2417 |
+
"loss": 0.0302,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.6381766381766383,
|
| 2422 |
+
"grad_norm": 0.35397287574395203,
|
| 2423 |
+
"learning_rate": 8.504043851610122e-05,
|
| 2424 |
+
"loss": 0.0285,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.6429249762583096,
|
| 2429 |
+
"grad_norm": 0.35181123197530567,
|
| 2430 |
+
"learning_rate": 8.492200981111482e-05,
|
| 2431 |
+
"loss": 0.0276,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.6476733143399809,
|
| 2436 |
+
"grad_norm": 0.39700888323674166,
|
| 2437 |
+
"learning_rate": 8.480319734868124e-05,
|
| 2438 |
+
"loss": 0.029,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.6524216524216524,
|
| 2443 |
+
"grad_norm": 0.36508298744053175,
|
| 2444 |
+
"learning_rate": 8.468400243442888e-05,
|
| 2445 |
+
"loss": 0.0293,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.657169990503324,
|
| 2450 |
+
"grad_norm": 0.36796649088087074,
|
| 2451 |
+
"learning_rate": 8.456442637818889e-05,
|
| 2452 |
+
"loss": 0.0279,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.6619183285849952,
|
| 2457 |
+
"grad_norm": 0.40423201885327376,
|
| 2458 |
+
"learning_rate": 8.444447049398076e-05,
|
| 2459 |
+
"loss": 0.0282,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.6666666666666665,
|
| 2464 |
+
"grad_norm": 0.4012837410158799,
|
| 2465 |
+
"learning_rate": 8.432413609999794e-05,
|
| 2466 |
+
"loss": 0.0282,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.671415004748338,
|
| 2471 |
+
"grad_norm": 0.3591977885417729,
|
| 2472 |
+
"learning_rate": 8.420342451859329e-05,
|
| 2473 |
+
"loss": 0.0289,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.6761633428300096,
|
| 2478 |
+
"grad_norm": 0.38049716791304095,
|
| 2479 |
+
"learning_rate": 8.408233707626458e-05,
|
| 2480 |
+
"loss": 0.0273,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.6809116809116809,
|
| 2485 |
+
"grad_norm": 0.38381110184655687,
|
| 2486 |
+
"learning_rate": 8.39608751036399e-05,
|
| 2487 |
+
"loss": 0.0279,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.6856600189933522,
|
| 2492 |
+
"grad_norm": 0.3758606793118219,
|
| 2493 |
+
"learning_rate": 8.383903993546311e-05,
|
| 2494 |
+
"loss": 0.0279,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.690408357075024,
|
| 2499 |
+
"grad_norm": 0.3652014999666907,
|
| 2500 |
+
"learning_rate": 8.371683291057902e-05,
|
| 2501 |
+
"loss": 0.0276,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.6951566951566952,
|
| 2506 |
+
"grad_norm": 0.35062250770884223,
|
| 2507 |
+
"learning_rate": 8.359425537191883e-05,
|
| 2508 |
+
"loss": 0.0279,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.6999050332383665,
|
| 2513 |
+
"grad_norm": 0.35429427902173394,
|
| 2514 |
+
"learning_rate": 8.347130866648528e-05,
|
| 2515 |
+
"loss": 0.027,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.704653371320038,
|
| 2520 |
+
"grad_norm": 0.39021507202575545,
|
| 2521 |
+
"learning_rate": 8.334799414533791e-05,
|
| 2522 |
+
"loss": 0.0275,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.7094017094017095,
|
| 2527 |
+
"grad_norm": 0.3337642290497236,
|
| 2528 |
+
"learning_rate": 8.322431316357815e-05,
|
| 2529 |
+
"loss": 0.0274,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
}
|
| 2532 |
+
],
|
| 2533 |
+
"logging_steps": 10,
|
| 2534 |
+
"max_steps": 10530,
|
| 2535 |
+
"num_input_tokens_seen": 0,
|
| 2536 |
+
"num_train_epochs": 5,
|
| 2537 |
+
"save_steps": 1800,
|
| 2538 |
+
"stateful_callbacks": {
|
| 2539 |
+
"TrainerControl": {
|
| 2540 |
+
"args": {
|
| 2541 |
+
"should_epoch_stop": false,
|
| 2542 |
+
"should_evaluate": false,
|
| 2543 |
+
"should_log": false,
|
| 2544 |
+
"should_save": true,
|
| 2545 |
+
"should_training_stop": false
|
| 2546 |
+
},
|
| 2547 |
+
"attributes": {}
|
| 2548 |
+
}
|
| 2549 |
+
},
|
| 2550 |
+
"total_flos": 226861792100352.0,
|
| 2551 |
+
"train_batch_size": 128,
|
| 2552 |
+
"trial_name": null,
|
| 2553 |
+
"trial_params": null
|
| 2554 |
+
}
|