Upload multi_student_4/checkpoint-3600/trainer_state.json with huggingface_hub
Browse files
multi_student_4/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": 2.3684210526315788,
|
| 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": 0.006578947368421052,
|
| 14 |
+
"grad_norm": 19.201007081270827,
|
| 15 |
+
"learning_rate": 8.458646616541353e-07,
|
| 16 |
+
"loss": 8.2522,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.013157894736842105,
|
| 21 |
+
"grad_norm": 18.114759300569585,
|
| 22 |
+
"learning_rate": 1.7857142857142857e-06,
|
| 23 |
+
"loss": 8.1346,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.019736842105263157,
|
| 28 |
+
"grad_norm": 9.479492840003324,
|
| 29 |
+
"learning_rate": 2.725563909774436e-06,
|
| 30 |
+
"loss": 7.5119,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02631578947368421,
|
| 35 |
+
"grad_norm": 4.976331973381574,
|
| 36 |
+
"learning_rate": 3.6654135338345866e-06,
|
| 37 |
+
"loss": 6.7788,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03289473684210526,
|
| 42 |
+
"grad_norm": 3.2804593438080754,
|
| 43 |
+
"learning_rate": 4.605263157894737e-06,
|
| 44 |
+
"loss": 6.2892,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.039473684210526314,
|
| 49 |
+
"grad_norm": 2.8583967169035027,
|
| 50 |
+
"learning_rate": 5.5451127819548875e-06,
|
| 51 |
+
"loss": 5.9528,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.046052631578947366,
|
| 56 |
+
"grad_norm": 3.5512069733934597,
|
| 57 |
+
"learning_rate": 6.484962406015038e-06,
|
| 58 |
+
"loss": 5.5755,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.05263157894736842,
|
| 63 |
+
"grad_norm": 3.731444843289522,
|
| 64 |
+
"learning_rate": 7.424812030075189e-06,
|
| 65 |
+
"loss": 4.9914,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.05921052631578947,
|
| 70 |
+
"grad_norm": 5.60127901105382,
|
| 71 |
+
"learning_rate": 8.364661654135339e-06,
|
| 72 |
+
"loss": 4.3233,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.06578947368421052,
|
| 77 |
+
"grad_norm": 3.7839180004348996,
|
| 78 |
+
"learning_rate": 9.30451127819549e-06,
|
| 79 |
+
"loss": 3.7305,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07236842105263158,
|
| 84 |
+
"grad_norm": 5.421177691842472,
|
| 85 |
+
"learning_rate": 1.024436090225564e-05,
|
| 86 |
+
"loss": 3.3248,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.07894736842105263,
|
| 91 |
+
"grad_norm": 3.7158079343320694,
|
| 92 |
+
"learning_rate": 1.118421052631579e-05,
|
| 93 |
+
"loss": 3.092,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.08552631578947369,
|
| 98 |
+
"grad_norm": 5.042680618782553,
|
| 99 |
+
"learning_rate": 1.212406015037594e-05,
|
| 100 |
+
"loss": 2.9383,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.09210526315789473,
|
| 105 |
+
"grad_norm": 5.8029975922287615,
|
| 106 |
+
"learning_rate": 1.3063909774436093e-05,
|
| 107 |
+
"loss": 2.8108,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.09868421052631579,
|
| 112 |
+
"grad_norm": 7.443887556478386,
|
| 113 |
+
"learning_rate": 1.4003759398496241e-05,
|
| 114 |
+
"loss": 2.7256,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.10526315789473684,
|
| 119 |
+
"grad_norm": 6.942055828457074,
|
| 120 |
+
"learning_rate": 1.4943609022556393e-05,
|
| 121 |
+
"loss": 2.6465,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1118421052631579,
|
| 126 |
+
"grad_norm": 20.211444588304172,
|
| 127 |
+
"learning_rate": 1.5883458646616542e-05,
|
| 128 |
+
"loss": 2.5756,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.11842105263157894,
|
| 133 |
+
"grad_norm": 23.482614308148037,
|
| 134 |
+
"learning_rate": 1.6823308270676694e-05,
|
| 135 |
+
"loss": 2.5262,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.125,
|
| 140 |
+
"grad_norm": 13.648396628618737,
|
| 141 |
+
"learning_rate": 1.7763157894736842e-05,
|
| 142 |
+
"loss": 2.4699,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.13157894736842105,
|
| 147 |
+
"grad_norm": 15.205800265082752,
|
| 148 |
+
"learning_rate": 1.8703007518796993e-05,
|
| 149 |
+
"loss": 2.4115,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.13815789473684212,
|
| 154 |
+
"grad_norm": 16.093588917694532,
|
| 155 |
+
"learning_rate": 1.9642857142857145e-05,
|
| 156 |
+
"loss": 2.3548,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.14473684210526316,
|
| 161 |
+
"grad_norm": 19.61654573509106,
|
| 162 |
+
"learning_rate": 2.0582706766917297e-05,
|
| 163 |
+
"loss": 2.3042,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.1513157894736842,
|
| 168 |
+
"grad_norm": 12.268305710278183,
|
| 169 |
+
"learning_rate": 2.1522556390977445e-05,
|
| 170 |
+
"loss": 2.2494,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.15789473684210525,
|
| 175 |
+
"grad_norm": 12.399788763360808,
|
| 176 |
+
"learning_rate": 2.2462406015037596e-05,
|
| 177 |
+
"loss": 2.1853,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.16447368421052633,
|
| 182 |
+
"grad_norm": 7.463420275410263,
|
| 183 |
+
"learning_rate": 2.3402255639097744e-05,
|
| 184 |
+
"loss": 2.0998,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.17105263157894737,
|
| 189 |
+
"grad_norm": 12.087535187997679,
|
| 190 |
+
"learning_rate": 2.4342105263157896e-05,
|
| 191 |
+
"loss": 2.0075,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.17763157894736842,
|
| 196 |
+
"grad_norm": 6.62910919540051,
|
| 197 |
+
"learning_rate": 2.5281954887218044e-05,
|
| 198 |
+
"loss": 1.8974,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.18421052631578946,
|
| 203 |
+
"grad_norm": 8.989470792407845,
|
| 204 |
+
"learning_rate": 2.62218045112782e-05,
|
| 205 |
+
"loss": 1.7928,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.19078947368421054,
|
| 210 |
+
"grad_norm": 6.928524369383665,
|
| 211 |
+
"learning_rate": 2.716165413533835e-05,
|
| 212 |
+
"loss": 1.7057,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.19736842105263158,
|
| 217 |
+
"grad_norm": 10.311746680652767,
|
| 218 |
+
"learning_rate": 2.8101503759398495e-05,
|
| 219 |
+
"loss": 1.6251,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.20394736842105263,
|
| 224 |
+
"grad_norm": 7.107313244759928,
|
| 225 |
+
"learning_rate": 2.9041353383458647e-05,
|
| 226 |
+
"loss": 1.5522,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.21052631578947367,
|
| 231 |
+
"grad_norm": 5.8857302431981156,
|
| 232 |
+
"learning_rate": 2.9981203007518798e-05,
|
| 233 |
+
"loss": 1.4974,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.21710526315789475,
|
| 238 |
+
"grad_norm": 6.775917238582609,
|
| 239 |
+
"learning_rate": 3.092105263157895e-05,
|
| 240 |
+
"loss": 1.4355,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2236842105263158,
|
| 245 |
+
"grad_norm": 9.938119729849578,
|
| 246 |
+
"learning_rate": 3.1860902255639094e-05,
|
| 247 |
+
"loss": 1.3895,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.23026315789473684,
|
| 252 |
+
"grad_norm": 4.870796794495703,
|
| 253 |
+
"learning_rate": 3.2800751879699246e-05,
|
| 254 |
+
"loss": 1.3432,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.23684210526315788,
|
| 259 |
+
"grad_norm": 5.353011232164181,
|
| 260 |
+
"learning_rate": 3.37406015037594e-05,
|
| 261 |
+
"loss": 1.3002,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.24342105263157895,
|
| 266 |
+
"grad_norm": 7.078794515741611,
|
| 267 |
+
"learning_rate": 3.468045112781955e-05,
|
| 268 |
+
"loss": 1.2641,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.25,
|
| 273 |
+
"grad_norm": 5.419600483372193,
|
| 274 |
+
"learning_rate": 3.56203007518797e-05,
|
| 275 |
+
"loss": 1.209,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.2565789473684211,
|
| 280 |
+
"grad_norm": 3.5327527407457926,
|
| 281 |
+
"learning_rate": 3.656015037593985e-05,
|
| 282 |
+
"loss": 1.1644,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.2631578947368421,
|
| 287 |
+
"grad_norm": 5.894523313050658,
|
| 288 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 289 |
+
"loss": 1.1319,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.26973684210526316,
|
| 294 |
+
"grad_norm": 3.9863193805342316,
|
| 295 |
+
"learning_rate": 3.8439849624060155e-05,
|
| 296 |
+
"loss": 1.094,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.27631578947368424,
|
| 301 |
+
"grad_norm": 4.225678742387179,
|
| 302 |
+
"learning_rate": 3.9379699248120307e-05,
|
| 303 |
+
"loss": 1.0651,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.28289473684210525,
|
| 308 |
+
"grad_norm": 5.586389531542015,
|
| 309 |
+
"learning_rate": 4.031954887218045e-05,
|
| 310 |
+
"loss": 1.0247,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2894736842105263,
|
| 315 |
+
"grad_norm": 7.0928476432315675,
|
| 316 |
+
"learning_rate": 4.12593984962406e-05,
|
| 317 |
+
"loss": 0.9973,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.29605263157894735,
|
| 322 |
+
"grad_norm": 5.797737179453841,
|
| 323 |
+
"learning_rate": 4.2199248120300754e-05,
|
| 324 |
+
"loss": 0.9738,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3026315789473684,
|
| 329 |
+
"grad_norm": 5.436160406452568,
|
| 330 |
+
"learning_rate": 4.3139097744360906e-05,
|
| 331 |
+
"loss": 0.9539,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3092105263157895,
|
| 336 |
+
"grad_norm": 4.940473111611711,
|
| 337 |
+
"learning_rate": 4.407894736842105e-05,
|
| 338 |
+
"loss": 0.9197,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3157894736842105,
|
| 343 |
+
"grad_norm": 5.844247327503255,
|
| 344 |
+
"learning_rate": 4.50187969924812e-05,
|
| 345 |
+
"loss": 0.8862,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3223684210526316,
|
| 350 |
+
"grad_norm": 4.744124381908569,
|
| 351 |
+
"learning_rate": 4.5958646616541354e-05,
|
| 352 |
+
"loss": 0.8668,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.32894736842105265,
|
| 357 |
+
"grad_norm": 4.689028836800058,
|
| 358 |
+
"learning_rate": 4.6898496240601505e-05,
|
| 359 |
+
"loss": 0.8476,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.3355263157894737,
|
| 364 |
+
"grad_norm": 4.398308995209079,
|
| 365 |
+
"learning_rate": 4.783834586466166e-05,
|
| 366 |
+
"loss": 0.82,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.34210526315789475,
|
| 371 |
+
"grad_norm": 5.554811082819198,
|
| 372 |
+
"learning_rate": 4.877819548872181e-05,
|
| 373 |
+
"loss": 0.7974,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.34868421052631576,
|
| 378 |
+
"grad_norm": 3.7243976034524584,
|
| 379 |
+
"learning_rate": 4.971804511278196e-05,
|
| 380 |
+
"loss": 0.7715,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.35526315789473684,
|
| 385 |
+
"grad_norm": 4.879577435611285,
|
| 386 |
+
"learning_rate": 5.0657894736842104e-05,
|
| 387 |
+
"loss": 0.7514,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.3618421052631579,
|
| 392 |
+
"grad_norm": 5.516075231521318,
|
| 393 |
+
"learning_rate": 5.1597744360902256e-05,
|
| 394 |
+
"loss": 0.7291,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.3684210526315789,
|
| 399 |
+
"grad_norm": 5.494655015332771,
|
| 400 |
+
"learning_rate": 5.253759398496241e-05,
|
| 401 |
+
"loss": 0.6974,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.375,
|
| 406 |
+
"grad_norm": 5.0610034411984826,
|
| 407 |
+
"learning_rate": 5.347744360902256e-05,
|
| 408 |
+
"loss": 0.6779,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.3815789473684211,
|
| 413 |
+
"grad_norm": 6.241132985448258,
|
| 414 |
+
"learning_rate": 5.441729323308271e-05,
|
| 415 |
+
"loss": 0.6723,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3881578947368421,
|
| 420 |
+
"grad_norm": 4.582816961569543,
|
| 421 |
+
"learning_rate": 5.535714285714286e-05,
|
| 422 |
+
"loss": 0.652,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.39473684210526316,
|
| 427 |
+
"grad_norm": 3.5845708789821042,
|
| 428 |
+
"learning_rate": 5.6296992481203013e-05,
|
| 429 |
+
"loss": 0.6154,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.40131578947368424,
|
| 434 |
+
"grad_norm": 4.328768044098044,
|
| 435 |
+
"learning_rate": 5.723684210526316e-05,
|
| 436 |
+
"loss": 0.6042,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.40789473684210525,
|
| 441 |
+
"grad_norm": 3.5473602851751793,
|
| 442 |
+
"learning_rate": 5.817669172932331e-05,
|
| 443 |
+
"loss": 0.5803,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.4144736842105263,
|
| 448 |
+
"grad_norm": 4.233181045605189,
|
| 449 |
+
"learning_rate": 5.911654135338346e-05,
|
| 450 |
+
"loss": 0.5688,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.42105263157894735,
|
| 455 |
+
"grad_norm": 4.0831509210740835,
|
| 456 |
+
"learning_rate": 6.005639097744361e-05,
|
| 457 |
+
"loss": 0.5385,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.4276315789473684,
|
| 462 |
+
"grad_norm": 3.8705204303765814,
|
| 463 |
+
"learning_rate": 6.0996240601503764e-05,
|
| 464 |
+
"loss": 0.5201,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.4342105263157895,
|
| 469 |
+
"grad_norm": 3.8269827275761337,
|
| 470 |
+
"learning_rate": 6.193609022556392e-05,
|
| 471 |
+
"loss": 0.504,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.4407894736842105,
|
| 476 |
+
"grad_norm": 3.7042333223437476,
|
| 477 |
+
"learning_rate": 6.287593984962407e-05,
|
| 478 |
+
"loss": 0.485,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.4473684210526316,
|
| 483 |
+
"grad_norm": 3.1454584376141725,
|
| 484 |
+
"learning_rate": 6.381578947368422e-05,
|
| 485 |
+
"loss": 0.4692,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.45394736842105265,
|
| 490 |
+
"grad_norm": 3.6715281367823698,
|
| 491 |
+
"learning_rate": 6.475563909774437e-05,
|
| 492 |
+
"loss": 0.4474,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.4605263157894737,
|
| 497 |
+
"grad_norm": 3.3520165299765203,
|
| 498 |
+
"learning_rate": 6.569548872180451e-05,
|
| 499 |
+
"loss": 0.4335,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.46710526315789475,
|
| 504 |
+
"grad_norm": 3.4830856151143554,
|
| 505 |
+
"learning_rate": 6.663533834586466e-05,
|
| 506 |
+
"loss": 0.4156,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.47368421052631576,
|
| 511 |
+
"grad_norm": 3.4130338264227835,
|
| 512 |
+
"learning_rate": 6.757518796992481e-05,
|
| 513 |
+
"loss": 0.3984,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.48026315789473684,
|
| 518 |
+
"grad_norm": 3.6699720598135834,
|
| 519 |
+
"learning_rate": 6.851503759398496e-05,
|
| 520 |
+
"loss": 0.3884,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.4868421052631579,
|
| 525 |
+
"grad_norm": 3.2598730307422192,
|
| 526 |
+
"learning_rate": 6.945488721804511e-05,
|
| 527 |
+
"loss": 0.3727,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.4934210526315789,
|
| 532 |
+
"grad_norm": 3.0065146921519097,
|
| 533 |
+
"learning_rate": 7.039473684210527e-05,
|
| 534 |
+
"loss": 0.3521,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5,
|
| 539 |
+
"grad_norm": 2.3801132893266956,
|
| 540 |
+
"learning_rate": 7.133458646616542e-05,
|
| 541 |
+
"loss": 0.3415,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.506578947368421,
|
| 546 |
+
"grad_norm": 2.5040976105308466,
|
| 547 |
+
"learning_rate": 7.227443609022557e-05,
|
| 548 |
+
"loss": 0.3303,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.5131578947368421,
|
| 553 |
+
"grad_norm": 2.3095873713136004,
|
| 554 |
+
"learning_rate": 7.321428571428571e-05,
|
| 555 |
+
"loss": 0.3155,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5197368421052632,
|
| 560 |
+
"grad_norm": 2.221909098699267,
|
| 561 |
+
"learning_rate": 7.415413533834586e-05,
|
| 562 |
+
"loss": 0.3,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.5263157894736842,
|
| 567 |
+
"grad_norm": 2.3650008130107008,
|
| 568 |
+
"learning_rate": 7.509398496240601e-05,
|
| 569 |
+
"loss": 0.2881,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.5328947368421053,
|
| 574 |
+
"grad_norm": 2.92228181953594,
|
| 575 |
+
"learning_rate": 7.603383458646618e-05,
|
| 576 |
+
"loss": 0.2759,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.5394736842105263,
|
| 581 |
+
"grad_norm": 2.39576316770683,
|
| 582 |
+
"learning_rate": 7.697368421052633e-05,
|
| 583 |
+
"loss": 0.2611,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.5460526315789473,
|
| 588 |
+
"grad_norm": 1.9345517046210943,
|
| 589 |
+
"learning_rate": 7.791353383458648e-05,
|
| 590 |
+
"loss": 0.2472,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.5526315789473685,
|
| 595 |
+
"grad_norm": 2.321485582474392,
|
| 596 |
+
"learning_rate": 7.885338345864663e-05,
|
| 597 |
+
"loss": 0.2409,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.5592105263157895,
|
| 602 |
+
"grad_norm": 1.8690657258212422,
|
| 603 |
+
"learning_rate": 7.979323308270678e-05,
|
| 604 |
+
"loss": 0.2308,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.5657894736842105,
|
| 609 |
+
"grad_norm": 1.901049780468097,
|
| 610 |
+
"learning_rate": 8.073308270676692e-05,
|
| 611 |
+
"loss": 0.2171,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.5723684210526315,
|
| 616 |
+
"grad_norm": 2.2229311039801596,
|
| 617 |
+
"learning_rate": 8.167293233082707e-05,
|
| 618 |
+
"loss": 0.2072,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.5789473684210527,
|
| 623 |
+
"grad_norm": 2.213221418124993,
|
| 624 |
+
"learning_rate": 8.261278195488722e-05,
|
| 625 |
+
"loss": 0.1965,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.5855263157894737,
|
| 630 |
+
"grad_norm": 1.870540842703205,
|
| 631 |
+
"learning_rate": 8.355263157894737e-05,
|
| 632 |
+
"loss": 0.1877,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.5921052631578947,
|
| 637 |
+
"grad_norm": 1.311680351399902,
|
| 638 |
+
"learning_rate": 8.449248120300753e-05,
|
| 639 |
+
"loss": 0.1792,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.5986842105263158,
|
| 644 |
+
"grad_norm": 1.5798663811679752,
|
| 645 |
+
"learning_rate": 8.543233082706768e-05,
|
| 646 |
+
"loss": 0.1716,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6052631578947368,
|
| 651 |
+
"grad_norm": 7.169134004434136,
|
| 652 |
+
"learning_rate": 8.637218045112783e-05,
|
| 653 |
+
"loss": 0.1721,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.6118421052631579,
|
| 658 |
+
"grad_norm": 1.5044155665130965,
|
| 659 |
+
"learning_rate": 8.731203007518798e-05,
|
| 660 |
+
"loss": 0.193,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.618421052631579,
|
| 665 |
+
"grad_norm": 1.305309283144098,
|
| 666 |
+
"learning_rate": 8.825187969924813e-05,
|
| 667 |
+
"loss": 0.1566,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.625,
|
| 672 |
+
"grad_norm": 1.23686939558907,
|
| 673 |
+
"learning_rate": 8.919172932330827e-05,
|
| 674 |
+
"loss": 0.1445,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.631578947368421,
|
| 679 |
+
"grad_norm": 1.0445490067006857,
|
| 680 |
+
"learning_rate": 9.013157894736842e-05,
|
| 681 |
+
"loss": 0.1356,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.6381578947368421,
|
| 686 |
+
"grad_norm": 1.3136634908844078,
|
| 687 |
+
"learning_rate": 9.107142857142857e-05,
|
| 688 |
+
"loss": 0.1308,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.6447368421052632,
|
| 693 |
+
"grad_norm": 1.7716809252302665,
|
| 694 |
+
"learning_rate": 9.201127819548872e-05,
|
| 695 |
+
"loss": 0.124,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.6513157894736842,
|
| 700 |
+
"grad_norm": 1.0772175167269311,
|
| 701 |
+
"learning_rate": 9.295112781954888e-05,
|
| 702 |
+
"loss": 0.1202,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.6578947368421053,
|
| 707 |
+
"grad_norm": 1.1878390185089365,
|
| 708 |
+
"learning_rate": 9.389097744360903e-05,
|
| 709 |
+
"loss": 0.114,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.6644736842105263,
|
| 714 |
+
"grad_norm": 0.8915150933559234,
|
| 715 |
+
"learning_rate": 9.483082706766918e-05,
|
| 716 |
+
"loss": 0.1054,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.6710526315789473,
|
| 721 |
+
"grad_norm": 0.7818448157917527,
|
| 722 |
+
"learning_rate": 9.577067669172933e-05,
|
| 723 |
+
"loss": 0.0996,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.6776315789473685,
|
| 728 |
+
"grad_norm": 0.990430969229284,
|
| 729 |
+
"learning_rate": 9.671052631578947e-05,
|
| 730 |
+
"loss": 0.0974,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.6842105263157895,
|
| 735 |
+
"grad_norm": 0.9276518666370865,
|
| 736 |
+
"learning_rate": 9.765037593984962e-05,
|
| 737 |
+
"loss": 0.0933,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.6907894736842105,
|
| 742 |
+
"grad_norm": 0.6846215892309722,
|
| 743 |
+
"learning_rate": 9.859022556390977e-05,
|
| 744 |
+
"loss": 0.0874,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.6973684210526315,
|
| 749 |
+
"grad_norm": 0.714553023496577,
|
| 750 |
+
"learning_rate": 9.953007518796992e-05,
|
| 751 |
+
"loss": 0.0814,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.7039473684210527,
|
| 756 |
+
"grad_norm": 0.7546346945965402,
|
| 757 |
+
"learning_rate": 9.999993273155467e-05,
|
| 758 |
+
"loss": 0.0774,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.7105263157894737,
|
| 763 |
+
"grad_norm": 0.571192300635228,
|
| 764 |
+
"learning_rate": 9.999939458507816e-05,
|
| 765 |
+
"loss": 0.0762,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.7171052631578947,
|
| 770 |
+
"grad_norm": 0.8584193322720889,
|
| 771 |
+
"learning_rate": 9.999831829791713e-05,
|
| 772 |
+
"loss": 0.0756,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.7236842105263158,
|
| 777 |
+
"grad_norm": 1.3952479453359226,
|
| 778 |
+
"learning_rate": 9.999670388165563e-05,
|
| 779 |
+
"loss": 0.075,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.7302631578947368,
|
| 784 |
+
"grad_norm": 0.751465926189691,
|
| 785 |
+
"learning_rate": 9.999455135366951e-05,
|
| 786 |
+
"loss": 0.069,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.7368421052631579,
|
| 791 |
+
"grad_norm": 0.4873110858150168,
|
| 792 |
+
"learning_rate": 9.999186073712635e-05,
|
| 793 |
+
"loss": 0.064,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.743421052631579,
|
| 798 |
+
"grad_norm": 0.4495642582063532,
|
| 799 |
+
"learning_rate": 9.998863206098507e-05,
|
| 800 |
+
"loss": 0.0623,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.75,
|
| 805 |
+
"grad_norm": 0.45684969056385805,
|
| 806 |
+
"learning_rate": 9.998486535999573e-05,
|
| 807 |
+
"loss": 0.059,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.756578947368421,
|
| 812 |
+
"grad_norm": 0.4304228051311271,
|
| 813 |
+
"learning_rate": 9.998056067469915e-05,
|
| 814 |
+
"loss": 0.0574,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.7631578947368421,
|
| 819 |
+
"grad_norm": 0.38692754350368835,
|
| 820 |
+
"learning_rate": 9.997571805142639e-05,
|
| 821 |
+
"loss": 0.0568,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.7697368421052632,
|
| 826 |
+
"grad_norm": 0.44697102978905795,
|
| 827 |
+
"learning_rate": 9.997033754229835e-05,
|
| 828 |
+
"loss": 0.054,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.7763157894736842,
|
| 833 |
+
"grad_norm": 0.3475319362475707,
|
| 834 |
+
"learning_rate": 9.996441920522512e-05,
|
| 835 |
+
"loss": 0.0512,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.7828947368421053,
|
| 840 |
+
"grad_norm": 0.3776964171490181,
|
| 841 |
+
"learning_rate": 9.995796310390547e-05,
|
| 842 |
+
"loss": 0.0496,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.7894736842105263,
|
| 847 |
+
"grad_norm": 0.3599383764053137,
|
| 848 |
+
"learning_rate": 9.995096930782604e-05,
|
| 849 |
+
"loss": 0.049,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.7960526315789473,
|
| 854 |
+
"grad_norm": 0.4004701605242826,
|
| 855 |
+
"learning_rate": 9.994343789226066e-05,
|
| 856 |
+
"loss": 0.0468,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.8026315789473685,
|
| 861 |
+
"grad_norm": 0.3822329385724864,
|
| 862 |
+
"learning_rate": 9.993536893826954e-05,
|
| 863 |
+
"loss": 0.0457,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.8092105263157895,
|
| 868 |
+
"grad_norm": 0.3620248875103048,
|
| 869 |
+
"learning_rate": 9.992676253269837e-05,
|
| 870 |
+
"loss": 0.0446,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.8157894736842105,
|
| 875 |
+
"grad_norm": 0.28358957764617276,
|
| 876 |
+
"learning_rate": 9.991761876817743e-05,
|
| 877 |
+
"loss": 0.0429,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.8223684210526315,
|
| 882 |
+
"grad_norm": 0.32929903528232773,
|
| 883 |
+
"learning_rate": 9.990793774312053e-05,
|
| 884 |
+
"loss": 0.0421,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.8289473684210527,
|
| 889 |
+
"grad_norm": 0.34141322063560897,
|
| 890 |
+
"learning_rate": 9.9897719561724e-05,
|
| 891 |
+
"loss": 0.0418,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.8355263157894737,
|
| 896 |
+
"grad_norm": 0.28296527941759625,
|
| 897 |
+
"learning_rate": 9.988696433396557e-05,
|
| 898 |
+
"loss": 0.0403,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.8421052631578947,
|
| 903 |
+
"grad_norm": 0.3686897659419288,
|
| 904 |
+
"learning_rate": 9.987567217560314e-05,
|
| 905 |
+
"loss": 0.0395,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.8486842105263158,
|
| 910 |
+
"grad_norm": 0.3066999773216274,
|
| 911 |
+
"learning_rate": 9.986384320817359e-05,
|
| 912 |
+
"loss": 0.0395,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.8552631578947368,
|
| 917 |
+
"grad_norm": 0.33225917789321974,
|
| 918 |
+
"learning_rate": 9.985147755899142e-05,
|
| 919 |
+
"loss": 0.0385,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.8618421052631579,
|
| 924 |
+
"grad_norm": 0.3004075266809671,
|
| 925 |
+
"learning_rate": 9.983857536114743e-05,
|
| 926 |
+
"loss": 0.0377,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.868421052631579,
|
| 931 |
+
"grad_norm": 0.2698302147467991,
|
| 932 |
+
"learning_rate": 9.982513675350724e-05,
|
| 933 |
+
"loss": 0.0367,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.875,
|
| 938 |
+
"grad_norm": 0.26807280624783797,
|
| 939 |
+
"learning_rate": 9.981116188070984e-05,
|
| 940 |
+
"loss": 0.0356,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.881578947368421,
|
| 945 |
+
"grad_norm": 0.3127969801843736,
|
| 946 |
+
"learning_rate": 9.979665089316602e-05,
|
| 947 |
+
"loss": 0.035,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.8881578947368421,
|
| 952 |
+
"grad_norm": 0.28653732054861003,
|
| 953 |
+
"learning_rate": 9.978160394705668e-05,
|
| 954 |
+
"loss": 0.0343,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.8947368421052632,
|
| 959 |
+
"grad_norm": 0.2504478087548026,
|
| 960 |
+
"learning_rate": 9.976602120433133e-05,
|
| 961 |
+
"loss": 0.0341,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.9013157894736842,
|
| 966 |
+
"grad_norm": 0.24532062538161709,
|
| 967 |
+
"learning_rate": 9.97499028327061e-05,
|
| 968 |
+
"loss": 0.0331,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.9078947368421053,
|
| 973 |
+
"grad_norm": 0.5595503828448865,
|
| 974 |
+
"learning_rate": 9.973324900566213e-05,
|
| 975 |
+
"loss": 0.041,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.9144736842105263,
|
| 980 |
+
"grad_norm": 0.32103277633584737,
|
| 981 |
+
"learning_rate": 9.971605990244365e-05,
|
| 982 |
+
"loss": 0.0396,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.9210526315789473,
|
| 987 |
+
"grad_norm": 0.39243105455264066,
|
| 988 |
+
"learning_rate": 9.969833570805599e-05,
|
| 989 |
+
"loss": 0.0345,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.9276315789473685,
|
| 994 |
+
"grad_norm": 0.24537350120004972,
|
| 995 |
+
"learning_rate": 9.968007661326369e-05,
|
| 996 |
+
"loss": 0.0323,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.9342105263157895,
|
| 1001 |
+
"grad_norm": 0.21282115870790244,
|
| 1002 |
+
"learning_rate": 9.966128281458832e-05,
|
| 1003 |
+
"loss": 0.0309,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.9407894736842105,
|
| 1008 |
+
"grad_norm": 0.22792100873990018,
|
| 1009 |
+
"learning_rate": 9.964195451430653e-05,
|
| 1010 |
+
"loss": 0.0302,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.9473684210526315,
|
| 1015 |
+
"grad_norm": 0.2513211642537743,
|
| 1016 |
+
"learning_rate": 9.962209192044772e-05,
|
| 1017 |
+
"loss": 0.0298,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.9539473684210527,
|
| 1022 |
+
"grad_norm": 0.25009559541845094,
|
| 1023 |
+
"learning_rate": 9.960169524679187e-05,
|
| 1024 |
+
"loss": 0.0293,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.9605263157894737,
|
| 1029 |
+
"grad_norm": 0.23293995789061867,
|
| 1030 |
+
"learning_rate": 9.958076471286724e-05,
|
| 1031 |
+
"loss": 0.0289,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.9671052631578947,
|
| 1036 |
+
"grad_norm": 0.22361624312785255,
|
| 1037 |
+
"learning_rate": 9.955930054394798e-05,
|
| 1038 |
+
"loss": 0.0282,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.9736842105263158,
|
| 1043 |
+
"grad_norm": 0.23906361425178743,
|
| 1044 |
+
"learning_rate": 9.953730297105175e-05,
|
| 1045 |
+
"loss": 0.0281,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.9802631578947368,
|
| 1050 |
+
"grad_norm": 0.23143391534241473,
|
| 1051 |
+
"learning_rate": 9.951477223093722e-05,
|
| 1052 |
+
"loss": 0.0275,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.9868421052631579,
|
| 1057 |
+
"grad_norm": 0.2147052689574608,
|
| 1058 |
+
"learning_rate": 9.949170856610144e-05,
|
| 1059 |
+
"loss": 0.0271,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.993421052631579,
|
| 1064 |
+
"grad_norm": 0.20377711124649098,
|
| 1065 |
+
"learning_rate": 9.946811222477737e-05,
|
| 1066 |
+
"loss": 0.0263,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.0,
|
| 1071 |
+
"grad_norm": 0.2135903768894096,
|
| 1072 |
+
"learning_rate": 9.944398346093109e-05,
|
| 1073 |
+
"loss": 0.0262,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.006578947368421,
|
| 1078 |
+
"grad_norm": 0.20254766138767108,
|
| 1079 |
+
"learning_rate": 9.941932253425917e-05,
|
| 1080 |
+
"loss": 0.0248,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.013157894736842,
|
| 1085 |
+
"grad_norm": 0.2197030932035537,
|
| 1086 |
+
"learning_rate": 9.939412971018574e-05,
|
| 1087 |
+
"loss": 0.0246,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.019736842105263,
|
| 1092 |
+
"grad_norm": 0.1933628837351623,
|
| 1093 |
+
"learning_rate": 9.936840525985978e-05,
|
| 1094 |
+
"loss": 0.0242,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.0263157894736843,
|
| 1099 |
+
"grad_norm": 0.19721213696918533,
|
| 1100 |
+
"learning_rate": 9.934214946015213e-05,
|
| 1101 |
+
"loss": 0.0238,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.0328947368421053,
|
| 1106 |
+
"grad_norm": 0.20774618691091581,
|
| 1107 |
+
"learning_rate": 9.931536259365244e-05,
|
| 1108 |
+
"loss": 0.0234,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.0394736842105263,
|
| 1113 |
+
"grad_norm": 0.19785742831408606,
|
| 1114 |
+
"learning_rate": 9.928804494866629e-05,
|
| 1115 |
+
"loss": 0.0228,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.0460526315789473,
|
| 1120 |
+
"grad_norm": 0.2035358859489929,
|
| 1121 |
+
"learning_rate": 9.926019681921196e-05,
|
| 1122 |
+
"loss": 0.0227,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.0526315789473684,
|
| 1127 |
+
"grad_norm": 0.20629560858059492,
|
| 1128 |
+
"learning_rate": 9.92318185050173e-05,
|
| 1129 |
+
"loss": 0.0227,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.0592105263157894,
|
| 1134 |
+
"grad_norm": 0.17943582514083126,
|
| 1135 |
+
"learning_rate": 9.920291031151652e-05,
|
| 1136 |
+
"loss": 0.0223,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.0657894736842106,
|
| 1141 |
+
"grad_norm": 0.2069047937737935,
|
| 1142 |
+
"learning_rate": 9.917347254984688e-05,
|
| 1143 |
+
"loss": 0.0219,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.0723684210526316,
|
| 1148 |
+
"grad_norm": 0.18149990279040348,
|
| 1149 |
+
"learning_rate": 9.91435055368454e-05,
|
| 1150 |
+
"loss": 0.0215,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.0789473684210527,
|
| 1155 |
+
"grad_norm": 0.21588259501861692,
|
| 1156 |
+
"learning_rate": 9.91130095950453e-05,
|
| 1157 |
+
"loss": 0.0216,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.0855263157894737,
|
| 1162 |
+
"grad_norm": 0.19248955273518795,
|
| 1163 |
+
"learning_rate": 9.908198505267273e-05,
|
| 1164 |
+
"loss": 0.0211,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.0921052631578947,
|
| 1169 |
+
"grad_norm": 0.21323425974286558,
|
| 1170 |
+
"learning_rate": 9.90504322436431e-05,
|
| 1171 |
+
"loss": 0.0212,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.0986842105263157,
|
| 1176 |
+
"grad_norm": 0.19659430552879165,
|
| 1177 |
+
"learning_rate": 9.901835150755754e-05,
|
| 1178 |
+
"loss": 0.0204,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.1052631578947367,
|
| 1183 |
+
"grad_norm": 0.18723287056746327,
|
| 1184 |
+
"learning_rate": 9.898574318969918e-05,
|
| 1185 |
+
"loss": 0.0202,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.111842105263158,
|
| 1190 |
+
"grad_norm": 0.18512625335755417,
|
| 1191 |
+
"learning_rate": 9.895260764102956e-05,
|
| 1192 |
+
"loss": 0.0198,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.118421052631579,
|
| 1197 |
+
"grad_norm": 0.19124114649058113,
|
| 1198 |
+
"learning_rate": 9.89189452181847e-05,
|
| 1199 |
+
"loss": 0.0201,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.125,
|
| 1204 |
+
"grad_norm": 0.18895302686535878,
|
| 1205 |
+
"learning_rate": 9.888475628347142e-05,
|
| 1206 |
+
"loss": 0.0195,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.131578947368421,
|
| 1211 |
+
"grad_norm": 0.1772267382799342,
|
| 1212 |
+
"learning_rate": 9.885004120486326e-05,
|
| 1213 |
+
"loss": 0.0191,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.138157894736842,
|
| 1218 |
+
"grad_norm": 0.17908235775562148,
|
| 1219 |
+
"learning_rate": 9.881480035599667e-05,
|
| 1220 |
+
"loss": 0.0189,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.1447368421052633,
|
| 1225 |
+
"grad_norm": 0.17580586627136524,
|
| 1226 |
+
"learning_rate": 9.877903411616697e-05,
|
| 1227 |
+
"loss": 0.0189,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.1513157894736843,
|
| 1232 |
+
"grad_norm": 0.18518601084678493,
|
| 1233 |
+
"learning_rate": 9.874274287032417e-05,
|
| 1234 |
+
"loss": 0.0186,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.1578947368421053,
|
| 1239 |
+
"grad_norm": 0.18048608002452143,
|
| 1240 |
+
"learning_rate": 9.870592700906892e-05,
|
| 1241 |
+
"loss": 0.0183,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.1644736842105263,
|
| 1246 |
+
"grad_norm": 0.15777383982627308,
|
| 1247 |
+
"learning_rate": 9.866858692864829e-05,
|
| 1248 |
+
"loss": 0.018,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.1710526315789473,
|
| 1253 |
+
"grad_norm": 0.20607208990685497,
|
| 1254 |
+
"learning_rate": 9.863072303095144e-05,
|
| 1255 |
+
"loss": 0.0175,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.1776315789473684,
|
| 1260 |
+
"grad_norm": 0.1611318675087161,
|
| 1261 |
+
"learning_rate": 9.859233572350541e-05,
|
| 1262 |
+
"loss": 0.0171,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.1842105263157894,
|
| 1267 |
+
"grad_norm": 0.17589801094486104,
|
| 1268 |
+
"learning_rate": 9.855342541947065e-05,
|
| 1269 |
+
"loss": 0.0171,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.1907894736842106,
|
| 1274 |
+
"grad_norm": 0.17121384019730954,
|
| 1275 |
+
"learning_rate": 9.851399253763656e-05,
|
| 1276 |
+
"loss": 0.017,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.1973684210526316,
|
| 1281 |
+
"grad_norm": 0.16253563174923477,
|
| 1282 |
+
"learning_rate": 9.847403750241704e-05,
|
| 1283 |
+
"loss": 0.0166,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.2039473684210527,
|
| 1288 |
+
"grad_norm": 0.17851024747091337,
|
| 1289 |
+
"learning_rate": 9.84335607438459e-05,
|
| 1290 |
+
"loss": 0.0166,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.2105263157894737,
|
| 1295 |
+
"grad_norm": 0.16919035493728743,
|
| 1296 |
+
"learning_rate": 9.839256269757223e-05,
|
| 1297 |
+
"loss": 0.0161,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.2171052631578947,
|
| 1302 |
+
"grad_norm": 0.1939807706315173,
|
| 1303 |
+
"learning_rate": 9.835104380485572e-05,
|
| 1304 |
+
"loss": 0.0163,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.2236842105263157,
|
| 1309 |
+
"grad_norm": 0.17147592181772123,
|
| 1310 |
+
"learning_rate": 9.830900451256187e-05,
|
| 1311 |
+
"loss": 0.0157,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.2302631578947367,
|
| 1316 |
+
"grad_norm": 0.17218945397214414,
|
| 1317 |
+
"learning_rate": 9.826644527315722e-05,
|
| 1318 |
+
"loss": 0.0156,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.236842105263158,
|
| 1323 |
+
"grad_norm": 0.16873620481355664,
|
| 1324 |
+
"learning_rate": 9.82233665447045e-05,
|
| 1325 |
+
"loss": 0.0151,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.243421052631579,
|
| 1330 |
+
"grad_norm": 0.16469742855514352,
|
| 1331 |
+
"learning_rate": 9.817976879085764e-05,
|
| 1332 |
+
"loss": 0.0153,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.25,
|
| 1337 |
+
"grad_norm": 0.1692877712959756,
|
| 1338 |
+
"learning_rate": 9.813565248085686e-05,
|
| 1339 |
+
"loss": 0.0148,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.256578947368421,
|
| 1344 |
+
"grad_norm": 0.16625631046909073,
|
| 1345 |
+
"learning_rate": 9.809101808952348e-05,
|
| 1346 |
+
"loss": 0.0148,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.263157894736842,
|
| 1351 |
+
"grad_norm": 0.18262802890291954,
|
| 1352 |
+
"learning_rate": 9.804586609725499e-05,
|
| 1353 |
+
"loss": 0.0147,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.2697368421052633,
|
| 1358 |
+
"grad_norm": 0.18607861180613644,
|
| 1359 |
+
"learning_rate": 9.800019699001974e-05,
|
| 1360 |
+
"loss": 0.0144,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.2763157894736843,
|
| 1365 |
+
"grad_norm": 0.16285676137508193,
|
| 1366 |
+
"learning_rate": 9.795401125935177e-05,
|
| 1367 |
+
"loss": 0.0143,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.2828947368421053,
|
| 1372 |
+
"grad_norm": 0.1707670608457863,
|
| 1373 |
+
"learning_rate": 9.790730940234552e-05,
|
| 1374 |
+
"loss": 0.0143,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.2894736842105263,
|
| 1379 |
+
"grad_norm": 0.15525731103278195,
|
| 1380 |
+
"learning_rate": 9.786009192165048e-05,
|
| 1381 |
+
"loss": 0.0137,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.2960526315789473,
|
| 1386 |
+
"grad_norm": 0.16347793214903164,
|
| 1387 |
+
"learning_rate": 9.781235932546574e-05,
|
| 1388 |
+
"loss": 0.0133,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.3026315789473684,
|
| 1393 |
+
"grad_norm": 0.15933544137667022,
|
| 1394 |
+
"learning_rate": 9.776411212753453e-05,
|
| 1395 |
+
"loss": 0.0134,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.3092105263157894,
|
| 1400 |
+
"grad_norm": 0.13814532510818728,
|
| 1401 |
+
"learning_rate": 9.771535084713879e-05,
|
| 1402 |
+
"loss": 0.0129,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.3157894736842106,
|
| 1407 |
+
"grad_norm": 0.16523793061268452,
|
| 1408 |
+
"learning_rate": 9.766607600909341e-05,
|
| 1409 |
+
"loss": 0.0128,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.3223684210526316,
|
| 1414 |
+
"grad_norm": 0.14022908820622706,
|
| 1415 |
+
"learning_rate": 9.761628814374073e-05,
|
| 1416 |
+
"loss": 0.0126,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.3289473684210527,
|
| 1421 |
+
"grad_norm": 0.2069333200433914,
|
| 1422 |
+
"learning_rate": 9.756598778694477e-05,
|
| 1423 |
+
"loss": 0.0125,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.3355263157894737,
|
| 1428 |
+
"grad_norm": 0.17334974425963282,
|
| 1429 |
+
"learning_rate": 9.751517548008542e-05,
|
| 1430 |
+
"loss": 0.0123,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.3421052631578947,
|
| 1435 |
+
"grad_norm": 0.1505657722626437,
|
| 1436 |
+
"learning_rate": 9.746385177005272e-05,
|
| 1437 |
+
"loss": 0.0121,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.3486842105263157,
|
| 1442 |
+
"grad_norm": 0.168192073561971,
|
| 1443 |
+
"learning_rate": 9.741201720924089e-05,
|
| 1444 |
+
"loss": 0.0117,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.3552631578947367,
|
| 1449 |
+
"grad_norm": 0.15074266703733896,
|
| 1450 |
+
"learning_rate": 9.73596723555424e-05,
|
| 1451 |
+
"loss": 0.0119,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.361842105263158,
|
| 1456 |
+
"grad_norm": 0.17348324277964447,
|
| 1457 |
+
"learning_rate": 9.730681777234196e-05,
|
| 1458 |
+
"loss": 0.0115,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.368421052631579,
|
| 1463 |
+
"grad_norm": 0.14638399471630661,
|
| 1464 |
+
"learning_rate": 9.725345402851053e-05,
|
| 1465 |
+
"loss": 0.0118,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.375,
|
| 1470 |
+
"grad_norm": 0.15352208983536622,
|
| 1471 |
+
"learning_rate": 9.719958169839906e-05,
|
| 1472 |
+
"loss": 0.0113,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.381578947368421,
|
| 1477 |
+
"grad_norm": 0.16084962439952788,
|
| 1478 |
+
"learning_rate": 9.714520136183243e-05,
|
| 1479 |
+
"loss": 0.0115,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.388157894736842,
|
| 1484 |
+
"grad_norm": 0.1344941180659645,
|
| 1485 |
+
"learning_rate": 9.709031360410318e-05,
|
| 1486 |
+
"loss": 0.011,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.3947368421052633,
|
| 1491 |
+
"grad_norm": 0.1452614934065403,
|
| 1492 |
+
"learning_rate": 9.703491901596516e-05,
|
| 1493 |
+
"loss": 0.0109,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.4013157894736843,
|
| 1498 |
+
"grad_norm": 0.13843890039006693,
|
| 1499 |
+
"learning_rate": 9.697901819362721e-05,
|
| 1500 |
+
"loss": 0.0107,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.4078947368421053,
|
| 1505 |
+
"grad_norm": 0.13684375615554292,
|
| 1506 |
+
"learning_rate": 9.692261173874675e-05,
|
| 1507 |
+
"loss": 0.0105,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.4144736842105263,
|
| 1512 |
+
"grad_norm": 0.14083447160331208,
|
| 1513 |
+
"learning_rate": 9.68657002584233e-05,
|
| 1514 |
+
"loss": 0.0102,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.4210526315789473,
|
| 1519 |
+
"grad_norm": 0.13789659384952696,
|
| 1520 |
+
"learning_rate": 9.680828436519196e-05,
|
| 1521 |
+
"loss": 0.0104,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.4276315789473684,
|
| 1526 |
+
"grad_norm": 0.14516526202810717,
|
| 1527 |
+
"learning_rate": 9.675036467701674e-05,
|
| 1528 |
+
"loss": 0.0101,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.4342105263157894,
|
| 1533 |
+
"grad_norm": 0.12293537720506266,
|
| 1534 |
+
"learning_rate": 9.669194181728403e-05,
|
| 1535 |
+
"loss": 0.0099,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.4407894736842106,
|
| 1540 |
+
"grad_norm": 0.1259249188196858,
|
| 1541 |
+
"learning_rate": 9.663301641479579e-05,
|
| 1542 |
+
"loss": 0.0097,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.4473684210526316,
|
| 1547 |
+
"grad_norm": 0.12557298712724865,
|
| 1548 |
+
"learning_rate": 9.65735891037628e-05,
|
| 1549 |
+
"loss": 0.0097,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.4539473684210527,
|
| 1554 |
+
"grad_norm": 0.143039377177587,
|
| 1555 |
+
"learning_rate": 9.65136605237979e-05,
|
| 1556 |
+
"loss": 0.0094,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.4605263157894737,
|
| 1561 |
+
"grad_norm": 0.13504613441385002,
|
| 1562 |
+
"learning_rate": 9.645323131990908e-05,
|
| 1563 |
+
"loss": 0.0096,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.4671052631578947,
|
| 1568 |
+
"grad_norm": 0.13228183618353204,
|
| 1569 |
+
"learning_rate": 9.639230214249242e-05,
|
| 1570 |
+
"loss": 0.0093,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.4736842105263157,
|
| 1575 |
+
"grad_norm": 0.1480027541710592,
|
| 1576 |
+
"learning_rate": 9.633087364732527e-05,
|
| 1577 |
+
"loss": 0.0093,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.4802631578947367,
|
| 1582 |
+
"grad_norm": 0.1293455699844313,
|
| 1583 |
+
"learning_rate": 9.626894649555909e-05,
|
| 1584 |
+
"loss": 0.0094,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.486842105263158,
|
| 1589 |
+
"grad_norm": 0.1393940674869785,
|
| 1590 |
+
"learning_rate": 9.620652135371232e-05,
|
| 1591 |
+
"loss": 0.0091,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.493421052631579,
|
| 1596 |
+
"grad_norm": 0.15121559409028545,
|
| 1597 |
+
"learning_rate": 9.61435988936633e-05,
|
| 1598 |
+
"loss": 0.0091,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.5,
|
| 1603 |
+
"grad_norm": 0.1301327368552193,
|
| 1604 |
+
"learning_rate": 9.608017979264292e-05,
|
| 1605 |
+
"loss": 0.0091,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.506578947368421,
|
| 1610 |
+
"grad_norm": 0.1421460799492161,
|
| 1611 |
+
"learning_rate": 9.601626473322744e-05,
|
| 1612 |
+
"loss": 0.0089,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.513157894736842,
|
| 1617 |
+
"grad_norm": 0.17041691269286594,
|
| 1618 |
+
"learning_rate": 9.595185440333103e-05,
|
| 1619 |
+
"loss": 0.0092,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.5197368421052633,
|
| 1624 |
+
"grad_norm": 0.14376606305315906,
|
| 1625 |
+
"learning_rate": 9.588694949619851e-05,
|
| 1626 |
+
"loss": 0.0086,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.526315789473684,
|
| 1631 |
+
"grad_norm": 0.13694156071766578,
|
| 1632 |
+
"learning_rate": 9.582155071039771e-05,
|
| 1633 |
+
"loss": 0.0087,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.5328947368421053,
|
| 1638 |
+
"grad_norm": 0.12727756352863123,
|
| 1639 |
+
"learning_rate": 9.575565874981214e-05,
|
| 1640 |
+
"loss": 0.0087,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.5394736842105263,
|
| 1645 |
+
"grad_norm": 0.12785472470651071,
|
| 1646 |
+
"learning_rate": 9.568927432363327e-05,
|
| 1647 |
+
"loss": 0.0082,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.5460526315789473,
|
| 1652 |
+
"grad_norm": 0.12091981259643964,
|
| 1653 |
+
"learning_rate": 9.562239814635294e-05,
|
| 1654 |
+
"loss": 0.0083,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.5526315789473686,
|
| 1659 |
+
"grad_norm": 0.12758853427180586,
|
| 1660 |
+
"learning_rate": 9.555503093775575e-05,
|
| 1661 |
+
"loss": 0.008,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.5592105263157894,
|
| 1666 |
+
"grad_norm": 0.12354223372657464,
|
| 1667 |
+
"learning_rate": 9.548717342291115e-05,
|
| 1668 |
+
"loss": 0.0081,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.5657894736842106,
|
| 1673 |
+
"grad_norm": 0.12076843350000686,
|
| 1674 |
+
"learning_rate": 9.541882633216579e-05,
|
| 1675 |
+
"loss": 0.0079,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.5723684210526314,
|
| 1680 |
+
"grad_norm": 0.1197614697335571,
|
| 1681 |
+
"learning_rate": 9.534999040113559e-05,
|
| 1682 |
+
"loss": 0.008,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.5789473684210527,
|
| 1687 |
+
"grad_norm": 0.12153418093253635,
|
| 1688 |
+
"learning_rate": 9.52806663706978e-05,
|
| 1689 |
+
"loss": 0.0077,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.5855263157894737,
|
| 1694 |
+
"grad_norm": 0.13543828010220124,
|
| 1695 |
+
"learning_rate": 9.521085498698309e-05,
|
| 1696 |
+
"loss": 0.0079,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.5921052631578947,
|
| 1701 |
+
"grad_norm": 0.10948806505136653,
|
| 1702 |
+
"learning_rate": 9.514055700136747e-05,
|
| 1703 |
+
"loss": 0.0077,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.598684210526316,
|
| 1708 |
+
"grad_norm": 0.10980266602887555,
|
| 1709 |
+
"learning_rate": 9.506977317046424e-05,
|
| 1710 |
+
"loss": 0.0074,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.6052631578947367,
|
| 1715 |
+
"grad_norm": 0.12485717268206407,
|
| 1716 |
+
"learning_rate": 9.49985042561158e-05,
|
| 1717 |
+
"loss": 0.0075,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.611842105263158,
|
| 1722 |
+
"grad_norm": 0.11541304858534408,
|
| 1723 |
+
"learning_rate": 9.492675102538547e-05,
|
| 1724 |
+
"loss": 0.0072,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.618421052631579,
|
| 1729 |
+
"grad_norm": 0.12122451397522918,
|
| 1730 |
+
"learning_rate": 9.485451425054926e-05,
|
| 1731 |
+
"loss": 0.0072,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.625,
|
| 1736 |
+
"grad_norm": 0.4210179348904546,
|
| 1737 |
+
"learning_rate": 9.478179470908756e-05,
|
| 1738 |
+
"loss": 0.0075,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.631578947368421,
|
| 1743 |
+
"grad_norm": 0.5896074678764591,
|
| 1744 |
+
"learning_rate": 9.470859318367669e-05,
|
| 1745 |
+
"loss": 0.0086,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.638157894736842,
|
| 1750 |
+
"grad_norm": 3.7823641198616387,
|
| 1751 |
+
"learning_rate": 9.463491046218058e-05,
|
| 1752 |
+
"loss": 0.0186,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.6447368421052633,
|
| 1757 |
+
"grad_norm": 5.014692495425624,
|
| 1758 |
+
"learning_rate": 9.456074733764224e-05,
|
| 1759 |
+
"loss": 0.0827,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.651315789473684,
|
| 1764 |
+
"grad_norm": 0.30328471004984053,
|
| 1765 |
+
"learning_rate": 9.448610460827522e-05,
|
| 1766 |
+
"loss": 0.0328,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.6578947368421053,
|
| 1771 |
+
"grad_norm": 0.17555996108962088,
|
| 1772 |
+
"learning_rate": 9.441098307745507e-05,
|
| 1773 |
+
"loss": 0.0119,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.6644736842105263,
|
| 1778 |
+
"grad_norm": 0.12830924181654307,
|
| 1779 |
+
"learning_rate": 9.433538355371057e-05,
|
| 1780 |
+
"loss": 0.0096,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.6710526315789473,
|
| 1785 |
+
"grad_norm": 0.1306978986667964,
|
| 1786 |
+
"learning_rate": 9.425930685071517e-05,
|
| 1787 |
+
"loss": 0.0083,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.6776315789473686,
|
| 1792 |
+
"grad_norm": 0.1507331325454933,
|
| 1793 |
+
"learning_rate": 9.418275378727819e-05,
|
| 1794 |
+
"loss": 0.0083,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.6842105263157894,
|
| 1799 |
+
"grad_norm": 0.11592835345613971,
|
| 1800 |
+
"learning_rate": 9.410572518733596e-05,
|
| 1801 |
+
"loss": 0.0079,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.6907894736842106,
|
| 1806 |
+
"grad_norm": 0.10808538914723062,
|
| 1807 |
+
"learning_rate": 9.402822187994298e-05,
|
| 1808 |
+
"loss": 0.0073,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.6973684210526314,
|
| 1813 |
+
"grad_norm": 0.11722096305528684,
|
| 1814 |
+
"learning_rate": 9.395024469926302e-05,
|
| 1815 |
+
"loss": 0.0073,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.7039473684210527,
|
| 1820 |
+
"grad_norm": 0.10458331688166168,
|
| 1821 |
+
"learning_rate": 9.387179448456009e-05,
|
| 1822 |
+
"loss": 0.0072,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.7105263157894737,
|
| 1827 |
+
"grad_norm": 0.11291592102103867,
|
| 1828 |
+
"learning_rate": 9.379287208018945e-05,
|
| 1829 |
+
"loss": 0.0067,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.7171052631578947,
|
| 1834 |
+
"grad_norm": 0.08646582566807197,
|
| 1835 |
+
"learning_rate": 9.371347833558856e-05,
|
| 1836 |
+
"loss": 0.0065,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.723684210526316,
|
| 1841 |
+
"grad_norm": 0.10553066382224695,
|
| 1842 |
+
"learning_rate": 9.363361410526783e-05,
|
| 1843 |
+
"loss": 0.0067,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.7302631578947367,
|
| 1848 |
+
"grad_norm": 0.09811144117737546,
|
| 1849 |
+
"learning_rate": 9.35532802488015e-05,
|
| 1850 |
+
"loss": 0.0065,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.736842105263158,
|
| 1855 |
+
"grad_norm": 0.09975550040113049,
|
| 1856 |
+
"learning_rate": 9.347247763081835e-05,
|
| 1857 |
+
"loss": 0.0064,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.743421052631579,
|
| 1862 |
+
"grad_norm": 0.11209372785035113,
|
| 1863 |
+
"learning_rate": 9.339120712099245e-05,
|
| 1864 |
+
"loss": 0.0065,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.75,
|
| 1869 |
+
"grad_norm": 0.09335458400468914,
|
| 1870 |
+
"learning_rate": 9.330946959403375e-05,
|
| 1871 |
+
"loss": 0.0061,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.756578947368421,
|
| 1876 |
+
"grad_norm": 0.0975073527993663,
|
| 1877 |
+
"learning_rate": 9.322726592967868e-05,
|
| 1878 |
+
"loss": 0.006,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.763157894736842,
|
| 1883 |
+
"grad_norm": 0.09366196020997528,
|
| 1884 |
+
"learning_rate": 9.314459701268065e-05,
|
| 1885 |
+
"loss": 0.006,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.7697368421052633,
|
| 1890 |
+
"grad_norm": 0.09562732934182246,
|
| 1891 |
+
"learning_rate": 9.306146373280061e-05,
|
| 1892 |
+
"loss": 0.0059,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.776315789473684,
|
| 1897 |
+
"grad_norm": 0.09966326146415878,
|
| 1898 |
+
"learning_rate": 9.297786698479739e-05,
|
| 1899 |
+
"loss": 0.006,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.7828947368421053,
|
| 1904 |
+
"grad_norm": 0.10122383105599052,
|
| 1905 |
+
"learning_rate": 9.289380766841811e-05,
|
| 1906 |
+
"loss": 0.0059,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.7894736842105263,
|
| 1911 |
+
"grad_norm": 0.10095110671102905,
|
| 1912 |
+
"learning_rate": 9.280928668838848e-05,
|
| 1913 |
+
"loss": 0.0057,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.7960526315789473,
|
| 1918 |
+
"grad_norm": 0.091482492267746,
|
| 1919 |
+
"learning_rate": 9.27243049544031e-05,
|
| 1920 |
+
"loss": 0.0057,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.8026315789473686,
|
| 1925 |
+
"grad_norm": 0.08891652643763134,
|
| 1926 |
+
"learning_rate": 9.263886338111559e-05,
|
| 1927 |
+
"loss": 0.0055,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.8092105263157894,
|
| 1932 |
+
"grad_norm": 0.09263144069565209,
|
| 1933 |
+
"learning_rate": 9.25529628881288e-05,
|
| 1934 |
+
"loss": 0.0055,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.8157894736842106,
|
| 1939 |
+
"grad_norm": 0.11137233685500829,
|
| 1940 |
+
"learning_rate": 9.246660439998497e-05,
|
| 1941 |
+
"loss": 0.0055,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.8223684210526314,
|
| 1946 |
+
"grad_norm": 0.11114880917458508,
|
| 1947 |
+
"learning_rate": 9.237978884615565e-05,
|
| 1948 |
+
"loss": 0.0054,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.8289473684210527,
|
| 1953 |
+
"grad_norm": 0.09841409593938921,
|
| 1954 |
+
"learning_rate": 9.229251716103179e-05,
|
| 1955 |
+
"loss": 0.0055,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.8355263157894737,
|
| 1960 |
+
"grad_norm": 0.11223577944844708,
|
| 1961 |
+
"learning_rate": 9.220479028391364e-05,
|
| 1962 |
+
"loss": 0.0054,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.8421052631578947,
|
| 1967 |
+
"grad_norm": 0.09110216546608073,
|
| 1968 |
+
"learning_rate": 9.211660915900069e-05,
|
| 1969 |
+
"loss": 0.0053,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.848684210526316,
|
| 1974 |
+
"grad_norm": 0.09506804998323148,
|
| 1975 |
+
"learning_rate": 9.202797473538143e-05,
|
| 1976 |
+
"loss": 0.0053,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.8552631578947367,
|
| 1981 |
+
"grad_norm": 0.09167591406498078,
|
| 1982 |
+
"learning_rate": 9.193888796702321e-05,
|
| 1983 |
+
"loss": 0.0052,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.861842105263158,
|
| 1988 |
+
"grad_norm": 0.0910480942341495,
|
| 1989 |
+
"learning_rate": 9.184934981276192e-05,
|
| 1990 |
+
"loss": 0.0053,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.868421052631579,
|
| 1995 |
+
"grad_norm": 0.08389589500624173,
|
| 1996 |
+
"learning_rate": 9.175936123629171e-05,
|
| 1997 |
+
"loss": 0.0053,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.875,
|
| 2002 |
+
"grad_norm": 0.08751212493753141,
|
| 2003 |
+
"learning_rate": 9.16689232061546e-05,
|
| 2004 |
+
"loss": 0.0052,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.881578947368421,
|
| 2009 |
+
"grad_norm": 0.08925947798147087,
|
| 2010 |
+
"learning_rate": 9.157803669573e-05,
|
| 2011 |
+
"loss": 0.005,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.888157894736842,
|
| 2016 |
+
"grad_norm": 0.0896619316816315,
|
| 2017 |
+
"learning_rate": 9.148670268322438e-05,
|
| 2018 |
+
"loss": 0.005,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.8947368421052633,
|
| 2023 |
+
"grad_norm": 0.09175226088849178,
|
| 2024 |
+
"learning_rate": 9.139492215166059e-05,
|
| 2025 |
+
"loss": 0.0048,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.901315789473684,
|
| 2030 |
+
"grad_norm": 0.08885389457486895,
|
| 2031 |
+
"learning_rate": 9.130269608886735e-05,
|
| 2032 |
+
"loss": 0.005,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.9078947368421053,
|
| 2037 |
+
"grad_norm": 0.09883908696946943,
|
| 2038 |
+
"learning_rate": 9.12100254874686e-05,
|
| 2039 |
+
"loss": 0.0051,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.9144736842105263,
|
| 2044 |
+
"grad_norm": 0.09233528269274262,
|
| 2045 |
+
"learning_rate": 9.111691134487286e-05,
|
| 2046 |
+
"loss": 0.005,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.9210526315789473,
|
| 2051 |
+
"grad_norm": 0.08544980345629175,
|
| 2052 |
+
"learning_rate": 9.102335466326241e-05,
|
| 2053 |
+
"loss": 0.0049,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.9276315789473686,
|
| 2058 |
+
"grad_norm": 0.10003966542752647,
|
| 2059 |
+
"learning_rate": 9.092935644958257e-05,
|
| 2060 |
+
"loss": 0.0048,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.9342105263157894,
|
| 2065 |
+
"grad_norm": 0.08168055005388938,
|
| 2066 |
+
"learning_rate": 9.083491771553089e-05,
|
| 2067 |
+
"loss": 0.0047,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.9407894736842106,
|
| 2072 |
+
"grad_norm": 0.0992409697866901,
|
| 2073 |
+
"learning_rate": 9.074003947754615e-05,
|
| 2074 |
+
"loss": 0.0047,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.9473684210526314,
|
| 2079 |
+
"grad_norm": 0.08492119209203541,
|
| 2080 |
+
"learning_rate": 9.06447227567975e-05,
|
| 2081 |
+
"loss": 0.0048,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.9539473684210527,
|
| 2086 |
+
"grad_norm": 0.08129120975425032,
|
| 2087 |
+
"learning_rate": 9.054896857917349e-05,
|
| 2088 |
+
"loss": 0.0048,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.9605263157894737,
|
| 2093 |
+
"grad_norm": 0.08471582707226462,
|
| 2094 |
+
"learning_rate": 9.045277797527096e-05,
|
| 2095 |
+
"loss": 0.0047,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.9671052631578947,
|
| 2100 |
+
"grad_norm": 0.08915074149762475,
|
| 2101 |
+
"learning_rate": 9.035615198038401e-05,
|
| 2102 |
+
"loss": 0.0047,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.973684210526316,
|
| 2107 |
+
"grad_norm": 0.08922804724560857,
|
| 2108 |
+
"learning_rate": 9.02590916344928e-05,
|
| 2109 |
+
"loss": 0.0046,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.9802631578947367,
|
| 2114 |
+
"grad_norm": 0.08357488105287375,
|
| 2115 |
+
"learning_rate": 9.016159798225239e-05,
|
| 2116 |
+
"loss": 0.0045,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.986842105263158,
|
| 2121 |
+
"grad_norm": 0.09089197660267276,
|
| 2122 |
+
"learning_rate": 9.006367207298152e-05,
|
| 2123 |
+
"loss": 0.0045,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.993421052631579,
|
| 2128 |
+
"grad_norm": 0.07872067497811534,
|
| 2129 |
+
"learning_rate": 8.996531496065128e-05,
|
| 2130 |
+
"loss": 0.0046,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.0,
|
| 2135 |
+
"grad_norm": 0.08439906072392114,
|
| 2136 |
+
"learning_rate": 8.986652770387372e-05,
|
| 2137 |
+
"loss": 0.0044,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.0065789473684212,
|
| 2142 |
+
"grad_norm": 0.06961910527253773,
|
| 2143 |
+
"learning_rate": 8.976731136589059e-05,
|
| 2144 |
+
"loss": 0.0033,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.013157894736842,
|
| 2149 |
+
"grad_norm": 0.07154791151160267,
|
| 2150 |
+
"learning_rate": 8.966766701456177e-05,
|
| 2151 |
+
"loss": 0.0033,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.0197368421052633,
|
| 2156 |
+
"grad_norm": 0.07772175093467776,
|
| 2157 |
+
"learning_rate": 8.956759572235383e-05,
|
| 2158 |
+
"loss": 0.0031,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.026315789473684,
|
| 2163 |
+
"grad_norm": 0.0733292062504736,
|
| 2164 |
+
"learning_rate": 8.946709856632848e-05,
|
| 2165 |
+
"loss": 0.0031,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.0328947368421053,
|
| 2170 |
+
"grad_norm": 0.07783736387463817,
|
| 2171 |
+
"learning_rate": 8.9366176628131e-05,
|
| 2172 |
+
"loss": 0.003,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.039473684210526,
|
| 2177 |
+
"grad_norm": 0.06633555943399455,
|
| 2178 |
+
"learning_rate": 8.926483099397855e-05,
|
| 2179 |
+
"loss": 0.003,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.0460526315789473,
|
| 2184 |
+
"grad_norm": 0.07782294604433761,
|
| 2185 |
+
"learning_rate": 8.916306275464852e-05,
|
| 2186 |
+
"loss": 0.0029,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.0526315789473686,
|
| 2191 |
+
"grad_norm": 0.07670901138238165,
|
| 2192 |
+
"learning_rate": 8.906087300546675e-05,
|
| 2193 |
+
"loss": 0.0031,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.0592105263157894,
|
| 2198 |
+
"grad_norm": 0.07524758423152544,
|
| 2199 |
+
"learning_rate": 8.895826284629582e-05,
|
| 2200 |
+
"loss": 0.0029,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.0657894736842106,
|
| 2205 |
+
"grad_norm": 0.07695970302800449,
|
| 2206 |
+
"learning_rate": 8.885523338152313e-05,
|
| 2207 |
+
"loss": 0.0031,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.0723684210526314,
|
| 2212 |
+
"grad_norm": 0.0745426942514364,
|
| 2213 |
+
"learning_rate": 8.875178572004903e-05,
|
| 2214 |
+
"loss": 0.0032,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.0789473684210527,
|
| 2219 |
+
"grad_norm": 0.0723930470222858,
|
| 2220 |
+
"learning_rate": 8.864792097527489e-05,
|
| 2221 |
+
"loss": 0.0031,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.085526315789474,
|
| 2226 |
+
"grad_norm": 0.08361406797403782,
|
| 2227 |
+
"learning_rate": 8.854364026509117e-05,
|
| 2228 |
+
"loss": 0.0031,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.0921052631578947,
|
| 2233 |
+
"grad_norm": 0.06955666676258866,
|
| 2234 |
+
"learning_rate": 8.843894471186529e-05,
|
| 2235 |
+
"loss": 0.0031,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.098684210526316,
|
| 2240 |
+
"grad_norm": 0.07227264207927699,
|
| 2241 |
+
"learning_rate": 8.833383544242966e-05,
|
| 2242 |
+
"loss": 0.0031,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.1052631578947367,
|
| 2247 |
+
"grad_norm": 0.06914653302815033,
|
| 2248 |
+
"learning_rate": 8.822831358806942e-05,
|
| 2249 |
+
"loss": 0.0029,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.111842105263158,
|
| 2254 |
+
"grad_norm": 0.07323401762292578,
|
| 2255 |
+
"learning_rate": 8.812238028451043e-05,
|
| 2256 |
+
"loss": 0.003,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.1184210526315788,
|
| 2261 |
+
"grad_norm": 0.07393279181869614,
|
| 2262 |
+
"learning_rate": 8.801603667190686e-05,
|
| 2263 |
+
"loss": 0.003,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.125,
|
| 2268 |
+
"grad_norm": 0.07096487949532827,
|
| 2269 |
+
"learning_rate": 8.790928389482909e-05,
|
| 2270 |
+
"loss": 0.003,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.1315789473684212,
|
| 2275 |
+
"grad_norm": 0.0728089853059884,
|
| 2276 |
+
"learning_rate": 8.780212310225126e-05,
|
| 2277 |
+
"loss": 0.0031,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.138157894736842,
|
| 2282 |
+
"grad_norm": 0.0695501122917682,
|
| 2283 |
+
"learning_rate": 8.7694555447539e-05,
|
| 2284 |
+
"loss": 0.0031,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.1447368421052633,
|
| 2289 |
+
"grad_norm": 0.07705485434544583,
|
| 2290 |
+
"learning_rate": 8.758658208843694e-05,
|
| 2291 |
+
"loss": 0.0031,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.151315789473684,
|
| 2296 |
+
"grad_norm": 0.0845273469079357,
|
| 2297 |
+
"learning_rate": 8.74782041870563e-05,
|
| 2298 |
+
"loss": 0.0031,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.1578947368421053,
|
| 2303 |
+
"grad_norm": 0.08068281928940782,
|
| 2304 |
+
"learning_rate": 8.736942290986239e-05,
|
| 2305 |
+
"loss": 0.0031,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.1644736842105265,
|
| 2310 |
+
"grad_norm": 0.07961859812048024,
|
| 2311 |
+
"learning_rate": 8.726023942766196e-05,
|
| 2312 |
+
"loss": 0.003,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.1710526315789473,
|
| 2317 |
+
"grad_norm": 0.08758099933665693,
|
| 2318 |
+
"learning_rate": 8.715065491559076e-05,
|
| 2319 |
+
"loss": 0.0029,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.1776315789473686,
|
| 2324 |
+
"grad_norm": 0.07188296495338724,
|
| 2325 |
+
"learning_rate": 8.704067055310072e-05,
|
| 2326 |
+
"loss": 0.0031,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.1842105263157894,
|
| 2331 |
+
"grad_norm": 0.07663306624675915,
|
| 2332 |
+
"learning_rate": 8.693028752394741e-05,
|
| 2333 |
+
"loss": 0.003,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.1907894736842106,
|
| 2338 |
+
"grad_norm": 0.07462874387461102,
|
| 2339 |
+
"learning_rate": 8.681950701617715e-05,
|
| 2340 |
+
"loss": 0.003,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.1973684210526314,
|
| 2345 |
+
"grad_norm": 0.07336225552850846,
|
| 2346 |
+
"learning_rate": 8.67083302221144e-05,
|
| 2347 |
+
"loss": 0.003,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.2039473684210527,
|
| 2352 |
+
"grad_norm": 0.0752519347418371,
|
| 2353 |
+
"learning_rate": 8.659675833834872e-05,
|
| 2354 |
+
"loss": 0.0029,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.2105263157894735,
|
| 2359 |
+
"grad_norm": 0.0727931975094885,
|
| 2360 |
+
"learning_rate": 8.648479256572208e-05,
|
| 2361 |
+
"loss": 0.003,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.2171052631578947,
|
| 2366 |
+
"grad_norm": 0.07511952648108541,
|
| 2367 |
+
"learning_rate": 8.637243410931582e-05,
|
| 2368 |
+
"loss": 0.0029,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.223684210526316,
|
| 2373 |
+
"grad_norm": 0.07882019908319284,
|
| 2374 |
+
"learning_rate": 8.625968417843769e-05,
|
| 2375 |
+
"loss": 0.0029,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.2302631578947367,
|
| 2380 |
+
"grad_norm": 0.08265112815093757,
|
| 2381 |
+
"learning_rate": 8.614654398660889e-05,
|
| 2382 |
+
"loss": 0.0029,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.236842105263158,
|
| 2387 |
+
"grad_norm": 0.0797172772654838,
|
| 2388 |
+
"learning_rate": 8.603301475155098e-05,
|
| 2389 |
+
"loss": 0.003,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.2434210526315788,
|
| 2394 |
+
"grad_norm": 0.07058513328217333,
|
| 2395 |
+
"learning_rate": 8.591909769517277e-05,
|
| 2396 |
+
"loss": 0.0031,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.25,
|
| 2401 |
+
"grad_norm": 0.0806856826321758,
|
| 2402 |
+
"learning_rate": 8.580479404355714e-05,
|
| 2403 |
+
"loss": 0.003,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.2565789473684212,
|
| 2408 |
+
"grad_norm": 0.08816650076302643,
|
| 2409 |
+
"learning_rate": 8.56901050269479e-05,
|
| 2410 |
+
"loss": 0.003,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.263157894736842,
|
| 2415 |
+
"grad_norm": 0.08576523991244484,
|
| 2416 |
+
"learning_rate": 8.557503187973651e-05,
|
| 2417 |
+
"loss": 0.0031,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.2697368421052633,
|
| 2422 |
+
"grad_norm": 0.07817935240746017,
|
| 2423 |
+
"learning_rate": 8.545957584044882e-05,
|
| 2424 |
+
"loss": 0.003,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.276315789473684,
|
| 2429 |
+
"grad_norm": 0.07132102691438429,
|
| 2430 |
+
"learning_rate": 8.534373815173173e-05,
|
| 2431 |
+
"loss": 0.0028,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.2828947368421053,
|
| 2436 |
+
"grad_norm": 0.06907666862127639,
|
| 2437 |
+
"learning_rate": 8.522752006033977e-05,
|
| 2438 |
+
"loss": 0.0029,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.2894736842105265,
|
| 2443 |
+
"grad_norm": 0.07888763534903219,
|
| 2444 |
+
"learning_rate": 8.511092281712174e-05,
|
| 2445 |
+
"loss": 0.0028,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.2960526315789473,
|
| 2450 |
+
"grad_norm": 0.07104911749539242,
|
| 2451 |
+
"learning_rate": 8.499394767700728e-05,
|
| 2452 |
+
"loss": 0.0028,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.3026315789473686,
|
| 2457 |
+
"grad_norm": 0.07161589717548758,
|
| 2458 |
+
"learning_rate": 8.487659589899325e-05,
|
| 2459 |
+
"loss": 0.003,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.3092105263157894,
|
| 2464 |
+
"grad_norm": 0.07545738438626039,
|
| 2465 |
+
"learning_rate": 8.475886874613027e-05,
|
| 2466 |
+
"loss": 0.003,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.3157894736842106,
|
| 2471 |
+
"grad_norm": 0.06960029822263444,
|
| 2472 |
+
"learning_rate": 8.464076748550911e-05,
|
| 2473 |
+
"loss": 0.0028,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.3223684210526314,
|
| 2478 |
+
"grad_norm": 0.09053565638501836,
|
| 2479 |
+
"learning_rate": 8.4522293388247e-05,
|
| 2480 |
+
"loss": 0.003,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.3289473684210527,
|
| 2485 |
+
"grad_norm": 0.08111502757412685,
|
| 2486 |
+
"learning_rate": 8.440344772947403e-05,
|
| 2487 |
+
"loss": 0.0029,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.3355263157894735,
|
| 2492 |
+
"grad_norm": 0.07907419444465598,
|
| 2493 |
+
"learning_rate": 8.428423178831937e-05,
|
| 2494 |
+
"loss": 0.0029,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.3421052631578947,
|
| 2499 |
+
"grad_norm": 0.07530055852137314,
|
| 2500 |
+
"learning_rate": 8.416464684789751e-05,
|
| 2501 |
+
"loss": 0.0033,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.348684210526316,
|
| 2506 |
+
"grad_norm": 0.08578659904895668,
|
| 2507 |
+
"learning_rate": 8.404469419529448e-05,
|
| 2508 |
+
"loss": 0.0032,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.3552631578947367,
|
| 2513 |
+
"grad_norm": 0.07749694254198063,
|
| 2514 |
+
"learning_rate": 8.392437512155397e-05,
|
| 2515 |
+
"loss": 0.003,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.361842105263158,
|
| 2520 |
+
"grad_norm": 0.09613810033382267,
|
| 2521 |
+
"learning_rate": 8.380369092166343e-05,
|
| 2522 |
+
"loss": 0.0031,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 2.3684210526315788,
|
| 2527 |
+
"grad_norm": 0.10312166758776871,
|
| 2528 |
+
"learning_rate": 8.368264289454012e-05,
|
| 2529 |
+
"loss": 0.003,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
}
|
| 2532 |
+
],
|
| 2533 |
+
"logging_steps": 10,
|
| 2534 |
+
"max_steps": 10640,
|
| 2535 |
+
"num_input_tokens_seen": 0,
|
| 2536 |
+
"num_train_epochs": 7,
|
| 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": 286751327256576.0,
|
| 2551 |
+
"train_batch_size": 128,
|
| 2552 |
+
"trial_name": null,
|
| 2553 |
+
"trial_params": null
|
| 2554 |
+
}
|