Upload multi_student_3/checkpoint-3600/trainer_state.json with huggingface_hub
Browse files
multi_student_3/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": 2.7975126311698406,
|
| 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": 0.00777302759424796,
|
| 14 |
+
"grad_norm": 19.018803349985753,
|
| 15 |
+
"learning_rate": 8.737864077669904e-07,
|
| 16 |
+
"loss": 8.2473,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01554605518849592,
|
| 21 |
+
"grad_norm": 17.967208179671328,
|
| 22 |
+
"learning_rate": 1.8446601941747572e-06,
|
| 23 |
+
"loss": 8.1304,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.023319082782743878,
|
| 28 |
+
"grad_norm": 9.21576906708124,
|
| 29 |
+
"learning_rate": 2.8155339805825245e-06,
|
| 30 |
+
"loss": 7.4913,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.03109211037699184,
|
| 35 |
+
"grad_norm": 4.8422984662740465,
|
| 36 |
+
"learning_rate": 3.7864077669902915e-06,
|
| 37 |
+
"loss": 6.7578,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0388651379712398,
|
| 42 |
+
"grad_norm": 3.0883237528757603,
|
| 43 |
+
"learning_rate": 4.7572815533980585e-06,
|
| 44 |
+
"loss": 6.2789,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.046638165565487756,
|
| 49 |
+
"grad_norm": 3.0641013238554837,
|
| 50 |
+
"learning_rate": 5.728155339805825e-06,
|
| 51 |
+
"loss": 5.9381,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.054411193159735714,
|
| 56 |
+
"grad_norm": 3.6989178160715803,
|
| 57 |
+
"learning_rate": 6.699029126213592e-06,
|
| 58 |
+
"loss": 5.547,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.06218422075398368,
|
| 63 |
+
"grad_norm": 3.8926133061231023,
|
| 64 |
+
"learning_rate": 7.66990291262136e-06,
|
| 65 |
+
"loss": 4.9449,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06995724834823164,
|
| 70 |
+
"grad_norm": 6.005297193380389,
|
| 71 |
+
"learning_rate": 8.640776699029127e-06,
|
| 72 |
+
"loss": 4.2741,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0777302759424796,
|
| 77 |
+
"grad_norm": 4.634097942418493,
|
| 78 |
+
"learning_rate": 9.611650485436892e-06,
|
| 79 |
+
"loss": 3.692,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.08550330353672755,
|
| 84 |
+
"grad_norm": 3.9161281136802915,
|
| 85 |
+
"learning_rate": 1.0582524271844662e-05,
|
| 86 |
+
"loss": 3.3021,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.09327633113097551,
|
| 91 |
+
"grad_norm": 2.668188197386523,
|
| 92 |
+
"learning_rate": 1.1553398058252427e-05,
|
| 93 |
+
"loss": 3.0699,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10104935872522347,
|
| 98 |
+
"grad_norm": 5.67542909747789,
|
| 99 |
+
"learning_rate": 1.2524271844660197e-05,
|
| 100 |
+
"loss": 2.9232,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.10882238631947143,
|
| 105 |
+
"grad_norm": 8.358891491079559,
|
| 106 |
+
"learning_rate": 1.3495145631067962e-05,
|
| 107 |
+
"loss": 2.8142,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11659541391371939,
|
| 112 |
+
"grad_norm": 7.521330978238392,
|
| 113 |
+
"learning_rate": 1.446601941747573e-05,
|
| 114 |
+
"loss": 2.736,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.12436844150796736,
|
| 119 |
+
"grad_norm": 16.178722459983547,
|
| 120 |
+
"learning_rate": 1.5436893203883496e-05,
|
| 121 |
+
"loss": 2.6651,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1321414691022153,
|
| 126 |
+
"grad_norm": 8.335965189770683,
|
| 127 |
+
"learning_rate": 1.6407766990291263e-05,
|
| 128 |
+
"loss": 2.6021,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.13991449669646328,
|
| 133 |
+
"grad_norm": 8.022474097021126,
|
| 134 |
+
"learning_rate": 1.737864077669903e-05,
|
| 135 |
+
"loss": 2.5458,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.14768752429071122,
|
| 140 |
+
"grad_norm": 11.647238033711355,
|
| 141 |
+
"learning_rate": 1.8349514563106798e-05,
|
| 142 |
+
"loss": 2.4889,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1554605518849592,
|
| 147 |
+
"grad_norm": 11.910883080289608,
|
| 148 |
+
"learning_rate": 1.9320388349514565e-05,
|
| 149 |
+
"loss": 2.4326,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.16323357947920716,
|
| 154 |
+
"grad_norm": 11.410643498860912,
|
| 155 |
+
"learning_rate": 2.0291262135922333e-05,
|
| 156 |
+
"loss": 2.3697,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.1710066070734551,
|
| 161 |
+
"grad_norm": 11.108752154756692,
|
| 162 |
+
"learning_rate": 2.1262135922330097e-05,
|
| 163 |
+
"loss": 2.309,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.17877963466770308,
|
| 168 |
+
"grad_norm": 19.998718279627422,
|
| 169 |
+
"learning_rate": 2.2233009708737864e-05,
|
| 170 |
+
"loss": 2.241,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.18655266226195102,
|
| 175 |
+
"grad_norm": 11.758179967806427,
|
| 176 |
+
"learning_rate": 2.3203883495145632e-05,
|
| 177 |
+
"loss": 2.1862,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.194325689856199,
|
| 182 |
+
"grad_norm": 13.686852090688575,
|
| 183 |
+
"learning_rate": 2.41747572815534e-05,
|
| 184 |
+
"loss": 2.1354,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.20209871745044694,
|
| 189 |
+
"grad_norm": 7.097958211758118,
|
| 190 |
+
"learning_rate": 2.5145631067961167e-05,
|
| 191 |
+
"loss": 2.0252,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.2098717450446949,
|
| 196 |
+
"grad_norm": 12.553623894402433,
|
| 197 |
+
"learning_rate": 2.6116504854368934e-05,
|
| 198 |
+
"loss": 1.92,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.21764477263894286,
|
| 203 |
+
"grad_norm": 7.033016298250838,
|
| 204 |
+
"learning_rate": 2.7087378640776702e-05,
|
| 205 |
+
"loss": 1.832,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.22541780023319083,
|
| 210 |
+
"grad_norm": 8.05454776441266,
|
| 211 |
+
"learning_rate": 2.8058252427184466e-05,
|
| 212 |
+
"loss": 1.7505,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.23319082782743877,
|
| 217 |
+
"grad_norm": 6.32342513670527,
|
| 218 |
+
"learning_rate": 2.9029126213592233e-05,
|
| 219 |
+
"loss": 1.6596,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.24096385542168675,
|
| 224 |
+
"grad_norm": 6.0607588085936435,
|
| 225 |
+
"learning_rate": 3e-05,
|
| 226 |
+
"loss": 1.5836,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.24873688301593472,
|
| 231 |
+
"grad_norm": 7.205879956336063,
|
| 232 |
+
"learning_rate": 3.097087378640777e-05,
|
| 233 |
+
"loss": 1.5148,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.2565099106101827,
|
| 238 |
+
"grad_norm": 5.233158820350977,
|
| 239 |
+
"learning_rate": 3.194174757281553e-05,
|
| 240 |
+
"loss": 1.4655,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2642829382044306,
|
| 245 |
+
"grad_norm": 6.5668167184777495,
|
| 246 |
+
"learning_rate": 3.29126213592233e-05,
|
| 247 |
+
"loss": 1.3996,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2720559657986786,
|
| 252 |
+
"grad_norm": 5.866172867997477,
|
| 253 |
+
"learning_rate": 3.3883495145631074e-05,
|
| 254 |
+
"loss": 1.3546,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.27982899339292655,
|
| 259 |
+
"grad_norm": 7.078627326366601,
|
| 260 |
+
"learning_rate": 3.485436893203884e-05,
|
| 261 |
+
"loss": 1.2957,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2876020209871745,
|
| 266 |
+
"grad_norm": 7.013906915714377,
|
| 267 |
+
"learning_rate": 3.58252427184466e-05,
|
| 268 |
+
"loss": 1.265,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.29537504858142244,
|
| 273 |
+
"grad_norm": 5.721220928268079,
|
| 274 |
+
"learning_rate": 3.679611650485437e-05,
|
| 275 |
+
"loss": 1.2163,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3031480761756704,
|
| 280 |
+
"grad_norm": 5.571852622479908,
|
| 281 |
+
"learning_rate": 3.776699029126214e-05,
|
| 282 |
+
"loss": 1.163,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.3109211037699184,
|
| 287 |
+
"grad_norm": 5.4843565963743295,
|
| 288 |
+
"learning_rate": 3.873786407766991e-05,
|
| 289 |
+
"loss": 1.1277,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.31869413136416636,
|
| 294 |
+
"grad_norm": 7.069970365890535,
|
| 295 |
+
"learning_rate": 3.970873786407767e-05,
|
| 296 |
+
"loss": 1.1094,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3264671589584143,
|
| 301 |
+
"grad_norm": 5.425993491783754,
|
| 302 |
+
"learning_rate": 4.0679611650485436e-05,
|
| 303 |
+
"loss": 1.069,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.33424018655266224,
|
| 308 |
+
"grad_norm": 5.03054433074538,
|
| 309 |
+
"learning_rate": 4.165048543689321e-05,
|
| 310 |
+
"loss": 1.0382,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3420132141469102,
|
| 315 |
+
"grad_norm": 7.103802226586121,
|
| 316 |
+
"learning_rate": 4.262135922330097e-05,
|
| 317 |
+
"loss": 0.9988,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3497862417411582,
|
| 322 |
+
"grad_norm": 5.180871741408765,
|
| 323 |
+
"learning_rate": 4.3592233009708735e-05,
|
| 324 |
+
"loss": 0.979,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.35755926933540616,
|
| 329 |
+
"grad_norm": 5.168730423055745,
|
| 330 |
+
"learning_rate": 4.4563106796116506e-05,
|
| 331 |
+
"loss": 0.9442,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3653322969296541,
|
| 336 |
+
"grad_norm": 4.509772926215358,
|
| 337 |
+
"learning_rate": 4.553398058252427e-05,
|
| 338 |
+
"loss": 0.9304,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.37310532452390205,
|
| 343 |
+
"grad_norm": 5.497064414551351,
|
| 344 |
+
"learning_rate": 4.650485436893204e-05,
|
| 345 |
+
"loss": 0.8971,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.38087835211815,
|
| 350 |
+
"grad_norm": 4.2878368849547375,
|
| 351 |
+
"learning_rate": 4.747572815533981e-05,
|
| 352 |
+
"loss": 0.8754,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.388651379712398,
|
| 357 |
+
"grad_norm": 5.023778493195039,
|
| 358 |
+
"learning_rate": 4.8446601941747576e-05,
|
| 359 |
+
"loss": 0.8535,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.39642440730664597,
|
| 364 |
+
"grad_norm": 4.85429288760384,
|
| 365 |
+
"learning_rate": 4.941747572815535e-05,
|
| 366 |
+
"loss": 0.8176,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4041974349008939,
|
| 371 |
+
"grad_norm": 4.78285229843812,
|
| 372 |
+
"learning_rate": 5.0388349514563104e-05,
|
| 373 |
+
"loss": 0.8114,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.41197046249514185,
|
| 378 |
+
"grad_norm": 5.101751542601933,
|
| 379 |
+
"learning_rate": 5.135922330097087e-05,
|
| 380 |
+
"loss": 0.7752,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.4197434900893898,
|
| 385 |
+
"grad_norm": 5.448054773232898,
|
| 386 |
+
"learning_rate": 5.233009708737864e-05,
|
| 387 |
+
"loss": 0.755,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4275165176836378,
|
| 392 |
+
"grad_norm": 4.141934961389859,
|
| 393 |
+
"learning_rate": 5.3300970873786417e-05,
|
| 394 |
+
"loss": 0.7273,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.4352895452778857,
|
| 399 |
+
"grad_norm": 4.514745713617805,
|
| 400 |
+
"learning_rate": 5.427184466019418e-05,
|
| 401 |
+
"loss": 0.7143,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.4430625728721337,
|
| 406 |
+
"grad_norm": 3.852074915498448,
|
| 407 |
+
"learning_rate": 5.524271844660195e-05,
|
| 408 |
+
"loss": 0.6954,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.45083560046638166,
|
| 413 |
+
"grad_norm": 4.549795432019635,
|
| 414 |
+
"learning_rate": 5.6213592233009716e-05,
|
| 415 |
+
"loss": 0.6695,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.45860862806062963,
|
| 420 |
+
"grad_norm": 3.920314510813715,
|
| 421 |
+
"learning_rate": 5.718446601941748e-05,
|
| 422 |
+
"loss": 0.6503,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.46638165565487755,
|
| 427 |
+
"grad_norm": 5.518859711149864,
|
| 428 |
+
"learning_rate": 5.815533980582525e-05,
|
| 429 |
+
"loss": 0.6249,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4741546832491255,
|
| 434 |
+
"grad_norm": 4.876638883083892,
|
| 435 |
+
"learning_rate": 5.9126213592233015e-05,
|
| 436 |
+
"loss": 0.6064,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.4819277108433735,
|
| 441 |
+
"grad_norm": 5.2941697252063955,
|
| 442 |
+
"learning_rate": 6.009708737864078e-05,
|
| 443 |
+
"loss": 0.5911,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.48970073843762146,
|
| 448 |
+
"grad_norm": 4.245229081646686,
|
| 449 |
+
"learning_rate": 6.106796116504854e-05,
|
| 450 |
+
"loss": 0.5704,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.49747376603186944,
|
| 455 |
+
"grad_norm": 4.127680025681991,
|
| 456 |
+
"learning_rate": 6.203883495145631e-05,
|
| 457 |
+
"loss": 0.5589,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5052467936261174,
|
| 462 |
+
"grad_norm": 3.5437231540377616,
|
| 463 |
+
"learning_rate": 6.300970873786408e-05,
|
| 464 |
+
"loss": 0.5378,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5130198212203654,
|
| 469 |
+
"grad_norm": 4.180991087743002,
|
| 470 |
+
"learning_rate": 6.398058252427184e-05,
|
| 471 |
+
"loss": 0.5056,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5207928488146133,
|
| 476 |
+
"grad_norm": 3.7905222943864034,
|
| 477 |
+
"learning_rate": 6.495145631067961e-05,
|
| 478 |
+
"loss": 0.4944,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5285658764088612,
|
| 483 |
+
"grad_norm": 2.829726998778342,
|
| 484 |
+
"learning_rate": 6.592233009708738e-05,
|
| 485 |
+
"loss": 0.4706,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.5363389040031092,
|
| 490 |
+
"grad_norm": 2.9270824536877456,
|
| 491 |
+
"learning_rate": 6.689320388349515e-05,
|
| 492 |
+
"loss": 0.4539,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5441119315973572,
|
| 497 |
+
"grad_norm": 2.742603316305724,
|
| 498 |
+
"learning_rate": 6.786407766990291e-05,
|
| 499 |
+
"loss": 0.4376,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5518849591916052,
|
| 504 |
+
"grad_norm": 2.896538805797215,
|
| 505 |
+
"learning_rate": 6.883495145631068e-05,
|
| 506 |
+
"loss": 0.4113,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5596579867858531,
|
| 511 |
+
"grad_norm": 3.4815145675652013,
|
| 512 |
+
"learning_rate": 6.980582524271845e-05,
|
| 513 |
+
"loss": 0.4005,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.567431014380101,
|
| 518 |
+
"grad_norm": 2.270621987351258,
|
| 519 |
+
"learning_rate": 7.077669902912621e-05,
|
| 520 |
+
"loss": 0.3886,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.575204041974349,
|
| 525 |
+
"grad_norm": 2.7187166457028344,
|
| 526 |
+
"learning_rate": 7.174757281553398e-05,
|
| 527 |
+
"loss": 0.3765,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.582977069568597,
|
| 532 |
+
"grad_norm": 3.184495607766821,
|
| 533 |
+
"learning_rate": 7.271844660194175e-05,
|
| 534 |
+
"loss": 0.3514,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5907500971628449,
|
| 539 |
+
"grad_norm": 2.379059028542611,
|
| 540 |
+
"learning_rate": 7.368932038834951e-05,
|
| 541 |
+
"loss": 0.341,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.5985231247570929,
|
| 546 |
+
"grad_norm": 2.4310101684060705,
|
| 547 |
+
"learning_rate": 7.466019417475728e-05,
|
| 548 |
+
"loss": 0.3263,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6062961523513408,
|
| 553 |
+
"grad_norm": 2.5710567898956005,
|
| 554 |
+
"learning_rate": 7.563106796116505e-05,
|
| 555 |
+
"loss": 0.3117,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6140691799455888,
|
| 560 |
+
"grad_norm": 2.416082850923419,
|
| 561 |
+
"learning_rate": 7.660194174757281e-05,
|
| 562 |
+
"loss": 0.2944,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.6218422075398368,
|
| 567 |
+
"grad_norm": 2.109447352556029,
|
| 568 |
+
"learning_rate": 7.757281553398059e-05,
|
| 569 |
+
"loss": 0.2832,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6296152351340847,
|
| 574 |
+
"grad_norm": 2.2994104123752988,
|
| 575 |
+
"learning_rate": 7.854368932038836e-05,
|
| 576 |
+
"loss": 0.2726,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6373882627283327,
|
| 581 |
+
"grad_norm": 2.9267817958414524,
|
| 582 |
+
"learning_rate": 7.951456310679612e-05,
|
| 583 |
+
"loss": 0.2606,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.6451612903225806,
|
| 588 |
+
"grad_norm": 2.1699995560860197,
|
| 589 |
+
"learning_rate": 8.048543689320389e-05,
|
| 590 |
+
"loss": 0.2471,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.6529343179168287,
|
| 595 |
+
"grad_norm": 1.680681697729957,
|
| 596 |
+
"learning_rate": 8.145631067961166e-05,
|
| 597 |
+
"loss": 0.2354,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6607073455110766,
|
| 602 |
+
"grad_norm": 1.88393966323043,
|
| 603 |
+
"learning_rate": 8.242718446601942e-05,
|
| 604 |
+
"loss": 0.2263,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6684803731053245,
|
| 609 |
+
"grad_norm": 2.7385999995385957,
|
| 610 |
+
"learning_rate": 8.339805825242719e-05,
|
| 611 |
+
"loss": 0.2109,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6762534006995725,
|
| 616 |
+
"grad_norm": 1.71125390353767,
|
| 617 |
+
"learning_rate": 8.436893203883496e-05,
|
| 618 |
+
"loss": 0.2034,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.6840264282938204,
|
| 623 |
+
"grad_norm": 1.8856456915119455,
|
| 624 |
+
"learning_rate": 8.533980582524272e-05,
|
| 625 |
+
"loss": 0.1928,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6917994558880683,
|
| 630 |
+
"grad_norm": 1.54640226468132,
|
| 631 |
+
"learning_rate": 8.631067961165049e-05,
|
| 632 |
+
"loss": 0.1797,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.6995724834823164,
|
| 637 |
+
"grad_norm": 1.3309710354478357,
|
| 638 |
+
"learning_rate": 8.728155339805826e-05,
|
| 639 |
+
"loss": 0.1712,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7073455110765643,
|
| 644 |
+
"grad_norm": 1.2897601826602498,
|
| 645 |
+
"learning_rate": 8.825242718446602e-05,
|
| 646 |
+
"loss": 0.1624,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7151185386708123,
|
| 651 |
+
"grad_norm": 1.1022252566548159,
|
| 652 |
+
"learning_rate": 8.922330097087379e-05,
|
| 653 |
+
"loss": 0.1509,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7228915662650602,
|
| 658 |
+
"grad_norm": 1.4797407171136665,
|
| 659 |
+
"learning_rate": 9.019417475728156e-05,
|
| 660 |
+
"loss": 0.1495,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7306645938593082,
|
| 665 |
+
"grad_norm": 1.4708443604696282,
|
| 666 |
+
"learning_rate": 9.116504854368933e-05,
|
| 667 |
+
"loss": 0.1407,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.7384376214535562,
|
| 672 |
+
"grad_norm": 1.2062811965627402,
|
| 673 |
+
"learning_rate": 9.213592233009709e-05,
|
| 674 |
+
"loss": 0.1409,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.7462106490478041,
|
| 679 |
+
"grad_norm": 0.991340571431122,
|
| 680 |
+
"learning_rate": 9.310679611650486e-05,
|
| 681 |
+
"loss": 0.1267,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.7539836766420521,
|
| 686 |
+
"grad_norm": 1.0799976482859923,
|
| 687 |
+
"learning_rate": 9.407766990291263e-05,
|
| 688 |
+
"loss": 0.1294,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.7617567042363,
|
| 693 |
+
"grad_norm": 1.0055612260520772,
|
| 694 |
+
"learning_rate": 9.504854368932039e-05,
|
| 695 |
+
"loss": 0.1167,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.769529731830548,
|
| 700 |
+
"grad_norm": 0.99809772380525,
|
| 701 |
+
"learning_rate": 9.601941747572816e-05,
|
| 702 |
+
"loss": 0.1104,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.777302759424796,
|
| 707 |
+
"grad_norm": 0.8341966391057081,
|
| 708 |
+
"learning_rate": 9.699029126213593e-05,
|
| 709 |
+
"loss": 0.104,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7850757870190439,
|
| 714 |
+
"grad_norm": 0.7440675821604489,
|
| 715 |
+
"learning_rate": 9.796116504854369e-05,
|
| 716 |
+
"loss": 0.1,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.7928488146132919,
|
| 721 |
+
"grad_norm": 1.0089217362175902,
|
| 722 |
+
"learning_rate": 9.893203883495146e-05,
|
| 723 |
+
"loss": 0.0926,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.8006218422075398,
|
| 728 |
+
"grad_norm": 0.669475479205569,
|
| 729 |
+
"learning_rate": 9.990291262135923e-05,
|
| 730 |
+
"loss": 0.0855,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8083948698017878,
|
| 735 |
+
"grad_norm": 0.7508700947786369,
|
| 736 |
+
"learning_rate": 9.999976722311928e-05,
|
| 737 |
+
"loss": 0.0824,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8161678973960358,
|
| 742 |
+
"grad_norm": 0.5833904781662174,
|
| 743 |
+
"learning_rate": 9.999896256507963e-05,
|
| 744 |
+
"loss": 0.0794,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8239409249902837,
|
| 749 |
+
"grad_norm": 0.6285661168460489,
|
| 750 |
+
"learning_rate": 9.99975831613399e-05,
|
| 751 |
+
"loss": 0.0737,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.8317139525845316,
|
| 756 |
+
"grad_norm": 0.574291945207083,
|
| 757 |
+
"learning_rate": 9.99956290277565e-05,
|
| 758 |
+
"loss": 0.0689,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.8394869801787797,
|
| 763 |
+
"grad_norm": 0.4876511188851532,
|
| 764 |
+
"learning_rate": 9.999310018679253e-05,
|
| 765 |
+
"loss": 0.0659,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.8472600077730276,
|
| 770 |
+
"grad_norm": 0.553792085705227,
|
| 771 |
+
"learning_rate": 9.998999666751738e-05,
|
| 772 |
+
"loss": 0.064,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.8550330353672756,
|
| 777 |
+
"grad_norm": 0.4302541179456994,
|
| 778 |
+
"learning_rate": 9.998631850560647e-05,
|
| 779 |
+
"loss": 0.0612,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.8628060629615235,
|
| 784 |
+
"grad_norm": 0.47710435973221127,
|
| 785 |
+
"learning_rate": 9.998206574334089e-05,
|
| 786 |
+
"loss": 0.0572,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.8705790905557714,
|
| 791 |
+
"grad_norm": 0.6437234961571133,
|
| 792 |
+
"learning_rate": 9.997723842960671e-05,
|
| 793 |
+
"loss": 0.0556,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8783521181500195,
|
| 798 |
+
"grad_norm": 0.44297654610119586,
|
| 799 |
+
"learning_rate": 9.997183661989468e-05,
|
| 800 |
+
"loss": 0.0551,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8861251457442674,
|
| 805 |
+
"grad_norm": 0.40789308058272417,
|
| 806 |
+
"learning_rate": 9.996586037629944e-05,
|
| 807 |
+
"loss": 0.052,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8938981733385154,
|
| 812 |
+
"grad_norm": 0.412683747214122,
|
| 813 |
+
"learning_rate": 9.99593097675188e-05,
|
| 814 |
+
"loss": 0.0503,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9016712009327633,
|
| 819 |
+
"grad_norm": 0.4042307583726218,
|
| 820 |
+
"learning_rate": 9.995218486885302e-05,
|
| 821 |
+
"loss": 0.0491,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9094442285270112,
|
| 826 |
+
"grad_norm": 0.3917357425568075,
|
| 827 |
+
"learning_rate": 9.994448576220388e-05,
|
| 828 |
+
"loss": 0.0472,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9172172561212593,
|
| 833 |
+
"grad_norm": 0.42847580686284464,
|
| 834 |
+
"learning_rate": 9.99362125360738e-05,
|
| 835 |
+
"loss": 0.0461,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.9249902837155072,
|
| 840 |
+
"grad_norm": 0.36275912435050334,
|
| 841 |
+
"learning_rate": 9.992736528556474e-05,
|
| 842 |
+
"loss": 0.045,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.9327633113097551,
|
| 847 |
+
"grad_norm": 0.29922320469821373,
|
| 848 |
+
"learning_rate": 9.991794411237725e-05,
|
| 849 |
+
"loss": 0.0429,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.9405363389040031,
|
| 854 |
+
"grad_norm": 0.3914552553052167,
|
| 855 |
+
"learning_rate": 9.990794912480913e-05,
|
| 856 |
+
"loss": 0.0426,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.948309366498251,
|
| 861 |
+
"grad_norm": 0.38509264588619113,
|
| 862 |
+
"learning_rate": 9.989738043775428e-05,
|
| 863 |
+
"loss": 0.0423,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.9560823940924991,
|
| 868 |
+
"grad_norm": 0.3259823714473278,
|
| 869 |
+
"learning_rate": 9.988623817270137e-05,
|
| 870 |
+
"loss": 0.0406,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.963855421686747,
|
| 875 |
+
"grad_norm": 0.31452752346688945,
|
| 876 |
+
"learning_rate": 9.987452245773243e-05,
|
| 877 |
+
"loss": 0.0398,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.9716284492809949,
|
| 882 |
+
"grad_norm": 0.338353347661472,
|
| 883 |
+
"learning_rate": 9.986223342752141e-05,
|
| 884 |
+
"loss": 0.0386,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.9794014768752429,
|
| 889 |
+
"grad_norm": 0.3002478510776732,
|
| 890 |
+
"learning_rate": 9.984937122333256e-05,
|
| 891 |
+
"loss": 0.0382,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.9871745044694908,
|
| 896 |
+
"grad_norm": 0.40670865355319846,
|
| 897 |
+
"learning_rate": 9.983593599301891e-05,
|
| 898 |
+
"loss": 0.037,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9949475320637389,
|
| 903 |
+
"grad_norm": 0.6429283417769656,
|
| 904 |
+
"learning_rate": 9.982192789102047e-05,
|
| 905 |
+
"loss": 0.0373,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0023319082782745,
|
| 910 |
+
"grad_norm": 0.2801176127856459,
|
| 911 |
+
"learning_rate": 9.980734707836248e-05,
|
| 912 |
+
"loss": 0.0354,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0101049358725223,
|
| 917 |
+
"grad_norm": 0.3724656663427553,
|
| 918 |
+
"learning_rate": 9.979219372265365e-05,
|
| 919 |
+
"loss": 0.0344,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.0178779634667703,
|
| 924 |
+
"grad_norm": 0.2573842211275922,
|
| 925 |
+
"learning_rate": 9.97764679980841e-05,
|
| 926 |
+
"loss": 0.0333,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.0256509910610183,
|
| 931 |
+
"grad_norm": 0.24713329711260812,
|
| 932 |
+
"learning_rate": 9.97601700854234e-05,
|
| 933 |
+
"loss": 0.0336,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.0334240186552661,
|
| 938 |
+
"grad_norm": 0.23958650404178872,
|
| 939 |
+
"learning_rate": 9.974330017201859e-05,
|
| 940 |
+
"loss": 0.0328,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.0411970462495141,
|
| 945 |
+
"grad_norm": 0.22856953483452255,
|
| 946 |
+
"learning_rate": 9.972585845179188e-05,
|
| 947 |
+
"loss": 0.0316,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.0489700738437622,
|
| 952 |
+
"grad_norm": 0.25194563553211635,
|
| 953 |
+
"learning_rate": 9.970784512523848e-05,
|
| 954 |
+
"loss": 0.0311,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.0567431014380102,
|
| 959 |
+
"grad_norm": 0.23842502807724422,
|
| 960 |
+
"learning_rate": 9.968926039942435e-05,
|
| 961 |
+
"loss": 0.0306,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.064516129032258,
|
| 966 |
+
"grad_norm": 0.2701262980707877,
|
| 967 |
+
"learning_rate": 9.967010448798375e-05,
|
| 968 |
+
"loss": 0.0303,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.072289156626506,
|
| 973 |
+
"grad_norm": 0.25970806872919033,
|
| 974 |
+
"learning_rate": 9.965037761111681e-05,
|
| 975 |
+
"loss": 0.0295,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.080062184220754,
|
| 980 |
+
"grad_norm": 0.24331306995976767,
|
| 981 |
+
"learning_rate": 9.963007999558696e-05,
|
| 982 |
+
"loss": 0.0291,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.0878352118150019,
|
| 987 |
+
"grad_norm": 0.24100978412084842,
|
| 988 |
+
"learning_rate": 9.96092118747184e-05,
|
| 989 |
+
"loss": 0.0287,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.09560823940925,
|
| 994 |
+
"grad_norm": 0.2184207676934967,
|
| 995 |
+
"learning_rate": 9.958777348839336e-05,
|
| 996 |
+
"loss": 0.0277,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.103381267003498,
|
| 1001 |
+
"grad_norm": 0.21996045383271537,
|
| 1002 |
+
"learning_rate": 9.956576508304938e-05,
|
| 1003 |
+
"loss": 0.0272,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.1111542945977457,
|
| 1008 |
+
"grad_norm": 0.21628224837569832,
|
| 1009 |
+
"learning_rate": 9.954318691167641e-05,
|
| 1010 |
+
"loss": 0.0273,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.1189273221919938,
|
| 1015 |
+
"grad_norm": 0.22298883167746006,
|
| 1016 |
+
"learning_rate": 9.952003923381399e-05,
|
| 1017 |
+
"loss": 0.0264,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.1267003497862418,
|
| 1022 |
+
"grad_norm": 0.2246150118171153,
|
| 1023 |
+
"learning_rate": 9.949632231554819e-05,
|
| 1024 |
+
"loss": 0.0265,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1344733773804898,
|
| 1029 |
+
"grad_norm": 0.23186180017701422,
|
| 1030 |
+
"learning_rate": 9.947203642950862e-05,
|
| 1031 |
+
"loss": 0.0262,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.1422464049747376,
|
| 1036 |
+
"grad_norm": 0.24648849817769855,
|
| 1037 |
+
"learning_rate": 9.944718185486523e-05,
|
| 1038 |
+
"loss": 0.0256,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.1500194325689856,
|
| 1043 |
+
"grad_norm": 0.30413187225809407,
|
| 1044 |
+
"learning_rate": 9.942175887732517e-05,
|
| 1045 |
+
"loss": 0.025,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.1577924601632337,
|
| 1050 |
+
"grad_norm": 0.23138441255979614,
|
| 1051 |
+
"learning_rate": 9.93957677891294e-05,
|
| 1052 |
+
"loss": 0.0251,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.1655654877574815,
|
| 1057 |
+
"grad_norm": 0.24336635292704567,
|
| 1058 |
+
"learning_rate": 9.93692088890495e-05,
|
| 1059 |
+
"loss": 0.0244,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.1733385153517295,
|
| 1064 |
+
"grad_norm": 0.21185931343408212,
|
| 1065 |
+
"learning_rate": 9.934208248238403e-05,
|
| 1066 |
+
"loss": 0.0243,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.1811115429459775,
|
| 1071 |
+
"grad_norm": 0.19638440276738353,
|
| 1072 |
+
"learning_rate": 9.93143888809552e-05,
|
| 1073 |
+
"loss": 0.0239,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.1888845705402253,
|
| 1078 |
+
"grad_norm": 0.21756730427462398,
|
| 1079 |
+
"learning_rate": 9.928612840310514e-05,
|
| 1080 |
+
"loss": 0.0232,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.1966575981344734,
|
| 1085 |
+
"grad_norm": 0.21305245786411112,
|
| 1086 |
+
"learning_rate": 9.925730137369242e-05,
|
| 1087 |
+
"loss": 0.023,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.2044306257287214,
|
| 1092 |
+
"grad_norm": 0.22828714542453502,
|
| 1093 |
+
"learning_rate": 9.922790812408808e-05,
|
| 1094 |
+
"loss": 0.0226,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.2122036533229692,
|
| 1099 |
+
"grad_norm": 0.19930203197693303,
|
| 1100 |
+
"learning_rate": 9.919794899217203e-05,
|
| 1101 |
+
"loss": 0.0224,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.2199766809172172,
|
| 1106 |
+
"grad_norm": 0.20433748074735467,
|
| 1107 |
+
"learning_rate": 9.916742432232909e-05,
|
| 1108 |
+
"loss": 0.0222,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2277497085114653,
|
| 1113 |
+
"grad_norm": 0.20184151603700423,
|
| 1114 |
+
"learning_rate": 9.913633446544493e-05,
|
| 1115 |
+
"loss": 0.0215,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2355227361057133,
|
| 1120 |
+
"grad_norm": 0.2081529716929725,
|
| 1121 |
+
"learning_rate": 9.910467977890225e-05,
|
| 1122 |
+
"loss": 0.0216,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.243295763699961,
|
| 1127 |
+
"grad_norm": 0.27882658958977674,
|
| 1128 |
+
"learning_rate": 9.907246062657648e-05,
|
| 1129 |
+
"loss": 0.0234,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.2510687912942091,
|
| 1134 |
+
"grad_norm": 0.30986570505669037,
|
| 1135 |
+
"learning_rate": 9.903967737883168e-05,
|
| 1136 |
+
"loss": 0.0248,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.258841818888457,
|
| 1141 |
+
"grad_norm": 0.2205524563203657,
|
| 1142 |
+
"learning_rate": 9.900633041251631e-05,
|
| 1143 |
+
"loss": 0.0222,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.266614846482705,
|
| 1148 |
+
"grad_norm": 0.19645809273796086,
|
| 1149 |
+
"learning_rate": 9.897242011095883e-05,
|
| 1150 |
+
"loss": 0.0208,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.274387874076953,
|
| 1155 |
+
"grad_norm": 0.20220041423279006,
|
| 1156 |
+
"learning_rate": 9.893794686396329e-05,
|
| 1157 |
+
"loss": 0.0203,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.282160901671201,
|
| 1162 |
+
"grad_norm": 0.19701852522226898,
|
| 1163 |
+
"learning_rate": 9.890291106780495e-05,
|
| 1164 |
+
"loss": 0.0199,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.289933929265449,
|
| 1169 |
+
"grad_norm": 0.19055264078653414,
|
| 1170 |
+
"learning_rate": 9.886731312522563e-05,
|
| 1171 |
+
"loss": 0.0196,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.2977069568596968,
|
| 1176 |
+
"grad_norm": 0.17498611562411595,
|
| 1177 |
+
"learning_rate": 9.883115344542911e-05,
|
| 1178 |
+
"loss": 0.0189,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.3054799844539449,
|
| 1183 |
+
"grad_norm": 0.18664040460319856,
|
| 1184 |
+
"learning_rate": 9.879443244407643e-05,
|
| 1185 |
+
"loss": 0.019,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.3132530120481927,
|
| 1190 |
+
"grad_norm": 0.186899833087349,
|
| 1191 |
+
"learning_rate": 9.875715054328108e-05,
|
| 1192 |
+
"loss": 0.0186,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.3210260396424407,
|
| 1197 |
+
"grad_norm": 0.18960588899249967,
|
| 1198 |
+
"learning_rate": 9.871930817160422e-05,
|
| 1199 |
+
"loss": 0.018,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3287990672366887,
|
| 1204 |
+
"grad_norm": 0.2246998119833756,
|
| 1205 |
+
"learning_rate": 9.868090576404966e-05,
|
| 1206 |
+
"loss": 0.0185,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.3365720948309368,
|
| 1211 |
+
"grad_norm": 0.18612335330028418,
|
| 1212 |
+
"learning_rate": 9.864194376205895e-05,
|
| 1213 |
+
"loss": 0.0181,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.3443451224251846,
|
| 1218 |
+
"grad_norm": 0.19504144084562025,
|
| 1219 |
+
"learning_rate": 9.860242261350623e-05,
|
| 1220 |
+
"loss": 0.0174,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.3521181500194326,
|
| 1225 |
+
"grad_norm": 0.1940460381388096,
|
| 1226 |
+
"learning_rate": 9.856234277269313e-05,
|
| 1227 |
+
"loss": 0.0173,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.3598911776136804,
|
| 1232 |
+
"grad_norm": 0.1882305169943444,
|
| 1233 |
+
"learning_rate": 9.852170470034356e-05,
|
| 1234 |
+
"loss": 0.0173,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.3676642052079284,
|
| 1239 |
+
"grad_norm": 0.1844609631309131,
|
| 1240 |
+
"learning_rate": 9.848050886359835e-05,
|
| 1241 |
+
"loss": 0.0171,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.3754372328021764,
|
| 1246 |
+
"grad_norm": 0.16755276537647784,
|
| 1247 |
+
"learning_rate": 9.843875573600989e-05,
|
| 1248 |
+
"loss": 0.0169,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.3832102603964245,
|
| 1253 |
+
"grad_norm": 0.16696100318285242,
|
| 1254 |
+
"learning_rate": 9.839644579753677e-05,
|
| 1255 |
+
"loss": 0.0164,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.3909832879906725,
|
| 1260 |
+
"grad_norm": 0.1818918991872044,
|
| 1261 |
+
"learning_rate": 9.835357953453821e-05,
|
| 1262 |
+
"loss": 0.0163,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.3987563155849203,
|
| 1267 |
+
"grad_norm": 0.18963889299339048,
|
| 1268 |
+
"learning_rate": 9.831015743976843e-05,
|
| 1269 |
+
"loss": 0.0158,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.4065293431791683,
|
| 1274 |
+
"grad_norm": 0.1634818998306774,
|
| 1275 |
+
"learning_rate": 9.8266180012371e-05,
|
| 1276 |
+
"loss": 0.0154,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.4143023707734161,
|
| 1281 |
+
"grad_norm": 0.197090617938791,
|
| 1282 |
+
"learning_rate": 9.82216477578732e-05,
|
| 1283 |
+
"loss": 0.0152,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.4220753983676642,
|
| 1288 |
+
"grad_norm": 0.15850580732892314,
|
| 1289 |
+
"learning_rate": 9.817656118818006e-05,
|
| 1290 |
+
"loss": 0.0148,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.4298484259619122,
|
| 1295 |
+
"grad_norm": 0.15186976089409565,
|
| 1296 |
+
"learning_rate": 9.813092082156858e-05,
|
| 1297 |
+
"loss": 0.0146,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.4376214535561602,
|
| 1302 |
+
"grad_norm": 0.17352561780495993,
|
| 1303 |
+
"learning_rate": 9.808472718268174e-05,
|
| 1304 |
+
"loss": 0.0146,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.445394481150408,
|
| 1309 |
+
"grad_norm": 0.16575574275890378,
|
| 1310 |
+
"learning_rate": 9.803798080252245e-05,
|
| 1311 |
+
"loss": 0.014,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.453167508744656,
|
| 1316 |
+
"grad_norm": 0.16700742476995556,
|
| 1317 |
+
"learning_rate": 9.79906822184475e-05,
|
| 1318 |
+
"loss": 0.0139,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.460940536338904,
|
| 1323 |
+
"grad_norm": 0.15966930327908196,
|
| 1324 |
+
"learning_rate": 9.79428319741613e-05,
|
| 1325 |
+
"loss": 0.0136,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.4687135639331519,
|
| 1330 |
+
"grad_norm": 0.16653090767039122,
|
| 1331 |
+
"learning_rate": 9.789443061970973e-05,
|
| 1332 |
+
"loss": 0.0134,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.4764865915274,
|
| 1337 |
+
"grad_norm": 0.190626948029195,
|
| 1338 |
+
"learning_rate": 9.784547871147374e-05,
|
| 1339 |
+
"loss": 0.0129,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.484259619121648,
|
| 1344 |
+
"grad_norm": 0.1763435524084487,
|
| 1345 |
+
"learning_rate": 9.779597681216297e-05,
|
| 1346 |
+
"loss": 0.0127,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.492032646715896,
|
| 1351 |
+
"grad_norm": 0.1672345466206139,
|
| 1352 |
+
"learning_rate": 9.774592549080932e-05,
|
| 1353 |
+
"loss": 0.0126,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.4998056743101438,
|
| 1358 |
+
"grad_norm": 0.15678075113388967,
|
| 1359 |
+
"learning_rate": 9.769532532276034e-05,
|
| 1360 |
+
"loss": 0.0123,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.5075787019043918,
|
| 1365 |
+
"grad_norm": 0.14761977546971236,
|
| 1366 |
+
"learning_rate": 9.764417688967268e-05,
|
| 1367 |
+
"loss": 0.0118,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.5153517294986396,
|
| 1372 |
+
"grad_norm": 0.17063805371316892,
|
| 1373 |
+
"learning_rate": 9.75924807795054e-05,
|
| 1374 |
+
"loss": 0.0118,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.5231247570928876,
|
| 1379 |
+
"grad_norm": 0.14571436924477155,
|
| 1380 |
+
"learning_rate": 9.754023758651314e-05,
|
| 1381 |
+
"loss": 0.0115,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.5308977846871357,
|
| 1386 |
+
"grad_norm": 0.15118080081237256,
|
| 1387 |
+
"learning_rate": 9.74874479112394e-05,
|
| 1388 |
+
"loss": 0.0112,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.5386708122813837,
|
| 1393 |
+
"grad_norm": 0.1524883133010242,
|
| 1394 |
+
"learning_rate": 9.743411236050953e-05,
|
| 1395 |
+
"loss": 0.0112,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.5464438398756317,
|
| 1400 |
+
"grad_norm": 0.15389453717185902,
|
| 1401 |
+
"learning_rate": 9.738023154742383e-05,
|
| 1402 |
+
"loss": 0.0114,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.5542168674698795,
|
| 1407 |
+
"grad_norm": 0.14981753056593858,
|
| 1408 |
+
"learning_rate": 9.732580609135047e-05,
|
| 1409 |
+
"loss": 0.0108,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.5619898950641273,
|
| 1414 |
+
"grad_norm": 1.9198802022850896,
|
| 1415 |
+
"learning_rate": 9.727083661791836e-05,
|
| 1416 |
+
"loss": 0.0165,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.5697629226583754,
|
| 1421 |
+
"grad_norm": 0.26855012872950995,
|
| 1422 |
+
"learning_rate": 9.721532375900997e-05,
|
| 1423 |
+
"loss": 0.0144,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.5775359502526234,
|
| 1428 |
+
"grad_norm": 0.2071283206940206,
|
| 1429 |
+
"learning_rate": 9.71592681527541e-05,
|
| 1430 |
+
"loss": 0.0117,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.5853089778468714,
|
| 1435 |
+
"grad_norm": 0.14770534677344455,
|
| 1436 |
+
"learning_rate": 9.710267044351846e-05,
|
| 1437 |
+
"loss": 0.0109,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.5930820054411194,
|
| 1442 |
+
"grad_norm": 0.14368540945251085,
|
| 1443 |
+
"learning_rate": 9.704553128190239e-05,
|
| 1444 |
+
"loss": 0.0104,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.6008550330353672,
|
| 1449 |
+
"grad_norm": 0.13894193794675433,
|
| 1450 |
+
"learning_rate": 9.698785132472926e-05,
|
| 1451 |
+
"loss": 0.01,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.6086280606296153,
|
| 1456 |
+
"grad_norm": 0.13441313900735385,
|
| 1457 |
+
"learning_rate": 9.692963123503894e-05,
|
| 1458 |
+
"loss": 0.0094,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.616401088223863,
|
| 1463 |
+
"grad_norm": 0.12841760115701012,
|
| 1464 |
+
"learning_rate": 9.687087168208026e-05,
|
| 1465 |
+
"loss": 0.0094,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.624174115818111,
|
| 1470 |
+
"grad_norm": 0.1378414358921271,
|
| 1471 |
+
"learning_rate": 9.681157334130326e-05,
|
| 1472 |
+
"loss": 0.0092,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6319471434123591,
|
| 1477 |
+
"grad_norm": 0.13875742382402598,
|
| 1478 |
+
"learning_rate": 9.67517368943514e-05,
|
| 1479 |
+
"loss": 0.0093,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6397201710066072,
|
| 1484 |
+
"grad_norm": 0.12150978897988295,
|
| 1485 |
+
"learning_rate": 9.669136302905374e-05,
|
| 1486 |
+
"loss": 0.0094,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.6474931986008552,
|
| 1491 |
+
"grad_norm": 0.13386154431399386,
|
| 1492 |
+
"learning_rate": 9.663045243941711e-05,
|
| 1493 |
+
"loss": 0.009,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.655266226195103,
|
| 1498 |
+
"grad_norm": 0.14353227081750533,
|
| 1499 |
+
"learning_rate": 9.656900582561801e-05,
|
| 1500 |
+
"loss": 0.009,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.6630392537893508,
|
| 1505 |
+
"grad_norm": 0.13602583881387148,
|
| 1506 |
+
"learning_rate": 9.650702389399461e-05,
|
| 1507 |
+
"loss": 0.0088,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.6708122813835988,
|
| 1512 |
+
"grad_norm": 0.13201111610838254,
|
| 1513 |
+
"learning_rate": 9.644450735703872e-05,
|
| 1514 |
+
"loss": 0.0087,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.6785853089778469,
|
| 1519 |
+
"grad_norm": 0.11982302853617144,
|
| 1520 |
+
"learning_rate": 9.638145693338743e-05,
|
| 1521 |
+
"loss": 0.0087,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.6863583365720949,
|
| 1526 |
+
"grad_norm": 0.11312867242927008,
|
| 1527 |
+
"learning_rate": 9.631787334781497e-05,
|
| 1528 |
+
"loss": 0.0087,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.694131364166343,
|
| 1533 |
+
"grad_norm": 0.11767848177751597,
|
| 1534 |
+
"learning_rate": 9.625375733122439e-05,
|
| 1535 |
+
"loss": 0.0085,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.7019043917605907,
|
| 1540 |
+
"grad_norm": 0.11041145208342112,
|
| 1541 |
+
"learning_rate": 9.618910962063903e-05,
|
| 1542 |
+
"loss": 0.0082,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.7096774193548387,
|
| 1547 |
+
"grad_norm": 0.10799469734884444,
|
| 1548 |
+
"learning_rate": 9.612393095919423e-05,
|
| 1549 |
+
"loss": 0.0078,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.7174504469490866,
|
| 1554 |
+
"grad_norm": 0.11477500796146126,
|
| 1555 |
+
"learning_rate": 9.605822209612862e-05,
|
| 1556 |
+
"loss": 0.0078,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.7252234745433346,
|
| 1561 |
+
"grad_norm": 0.17423934102739813,
|
| 1562 |
+
"learning_rate": 9.599198378677559e-05,
|
| 1563 |
+
"loss": 0.0079,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.7329965021375826,
|
| 1568 |
+
"grad_norm": 0.11632697553641431,
|
| 1569 |
+
"learning_rate": 9.592521679255464e-05,
|
| 1570 |
+
"loss": 0.008,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.7407695297318306,
|
| 1575 |
+
"grad_norm": 0.12288031599865465,
|
| 1576 |
+
"learning_rate": 9.585792188096252e-05,
|
| 1577 |
+
"loss": 0.008,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.7485425573260787,
|
| 1582 |
+
"grad_norm": 0.15984027115213267,
|
| 1583 |
+
"learning_rate": 9.579009982556452e-05,
|
| 1584 |
+
"loss": 0.0079,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.7563155849203265,
|
| 1589 |
+
"grad_norm": 0.11306629622107851,
|
| 1590 |
+
"learning_rate": 9.572175140598553e-05,
|
| 1591 |
+
"loss": 0.0076,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.7640886125145743,
|
| 1596 |
+
"grad_norm": 0.1278172766126942,
|
| 1597 |
+
"learning_rate": 9.565287740790101e-05,
|
| 1598 |
+
"loss": 0.0077,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.7718616401088223,
|
| 1603 |
+
"grad_norm": 0.1251067207191811,
|
| 1604 |
+
"learning_rate": 9.55834786230281e-05,
|
| 1605 |
+
"loss": 0.0074,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.7796346677030703,
|
| 1610 |
+
"grad_norm": 0.12072123910224757,
|
| 1611 |
+
"learning_rate": 9.551355584911641e-05,
|
| 1612 |
+
"loss": 0.0077,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.7874076952973184,
|
| 1617 |
+
"grad_norm": 0.13420830427564445,
|
| 1618 |
+
"learning_rate": 9.544310988993889e-05,
|
| 1619 |
+
"loss": 0.0076,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.7951807228915664,
|
| 1624 |
+
"grad_norm": 0.11275927835211168,
|
| 1625 |
+
"learning_rate": 9.53721415552826e-05,
|
| 1626 |
+
"loss": 0.0072,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.8029537504858142,
|
| 1631 |
+
"grad_norm": 0.11663410063710995,
|
| 1632 |
+
"learning_rate": 9.530065166093936e-05,
|
| 1633 |
+
"loss": 0.007,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.8107267780800622,
|
| 1638 |
+
"grad_norm": 0.11515297587501505,
|
| 1639 |
+
"learning_rate": 9.52286410286964e-05,
|
| 1640 |
+
"loss": 0.007,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.81849980567431,
|
| 1645 |
+
"grad_norm": 0.12034539299056037,
|
| 1646 |
+
"learning_rate": 9.515611048632692e-05,
|
| 1647 |
+
"loss": 0.0071,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.826272833268558,
|
| 1652 |
+
"grad_norm": 0.13525219378529196,
|
| 1653 |
+
"learning_rate": 9.508306086758059e-05,
|
| 1654 |
+
"loss": 0.0069,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.834045860862806,
|
| 1659 |
+
"grad_norm": 0.11355987672855966,
|
| 1660 |
+
"learning_rate": 9.500949301217388e-05,
|
| 1661 |
+
"loss": 0.0066,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.841818888457054,
|
| 1666 |
+
"grad_norm": 0.10553177998244642,
|
| 1667 |
+
"learning_rate": 9.493540776578053e-05,
|
| 1668 |
+
"loss": 0.0067,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.8495919160513021,
|
| 1673 |
+
"grad_norm": 0.10733467006153838,
|
| 1674 |
+
"learning_rate": 9.486080598002176e-05,
|
| 1675 |
+
"loss": 0.0066,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.85736494364555,
|
| 1680 |
+
"grad_norm": 0.11088682968161348,
|
| 1681 |
+
"learning_rate": 9.478568851245643e-05,
|
| 1682 |
+
"loss": 0.0064,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.8651379712397977,
|
| 1687 |
+
"grad_norm": 0.11788349609407499,
|
| 1688 |
+
"learning_rate": 9.471005622657134e-05,
|
| 1689 |
+
"loss": 0.0064,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.8729109988340458,
|
| 1694 |
+
"grad_norm": 0.1382091673100836,
|
| 1695 |
+
"learning_rate": 9.463390999177113e-05,
|
| 1696 |
+
"loss": 0.0066,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.8806840264282938,
|
| 1701 |
+
"grad_norm": 0.11080598206299531,
|
| 1702 |
+
"learning_rate": 9.455725068336837e-05,
|
| 1703 |
+
"loss": 0.0067,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.8884570540225418,
|
| 1708 |
+
"grad_norm": 0.11345029521361591,
|
| 1709 |
+
"learning_rate": 9.44800791825735e-05,
|
| 1710 |
+
"loss": 0.0063,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.8962300816167899,
|
| 1715 |
+
"grad_norm": 0.12000127996735861,
|
| 1716 |
+
"learning_rate": 9.440239637648473e-05,
|
| 1717 |
+
"loss": 0.0064,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.9040031092110377,
|
| 1722 |
+
"grad_norm": 0.11742977407222017,
|
| 1723 |
+
"learning_rate": 9.432420315807771e-05,
|
| 1724 |
+
"loss": 0.0063,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.9117761368052857,
|
| 1729 |
+
"grad_norm": 0.10552545531273427,
|
| 1730 |
+
"learning_rate": 9.424550042619544e-05,
|
| 1731 |
+
"loss": 0.0062,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.9195491643995335,
|
| 1736 |
+
"grad_norm": 0.0996572096889176,
|
| 1737 |
+
"learning_rate": 9.416628908553784e-05,
|
| 1738 |
+
"loss": 0.0059,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.9273221919937815,
|
| 1743 |
+
"grad_norm": 0.09626416174818198,
|
| 1744 |
+
"learning_rate": 9.408657004665131e-05,
|
| 1745 |
+
"loss": 0.0058,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.9350952195880295,
|
| 1750 |
+
"grad_norm": 0.09783061431201936,
|
| 1751 |
+
"learning_rate": 9.400634422591838e-05,
|
| 1752 |
+
"loss": 0.0058,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.9428682471822776,
|
| 1757 |
+
"grad_norm": 0.11102429789832584,
|
| 1758 |
+
"learning_rate": 9.392561254554713e-05,
|
| 1759 |
+
"loss": 0.0059,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.9506412747765256,
|
| 1764 |
+
"grad_norm": 0.10011364069857738,
|
| 1765 |
+
"learning_rate": 9.384437593356048e-05,
|
| 1766 |
+
"loss": 0.0059,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.9584143023707734,
|
| 1771 |
+
"grad_norm": 0.10584225350181801,
|
| 1772 |
+
"learning_rate": 9.37626353237857e-05,
|
| 1773 |
+
"loss": 0.0061,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.9661873299650212,
|
| 1778 |
+
"grad_norm": 0.10444281925647403,
|
| 1779 |
+
"learning_rate": 9.368039165584359e-05,
|
| 1780 |
+
"loss": 0.0061,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.9739603575592692,
|
| 1785 |
+
"grad_norm": 0.10604191503186843,
|
| 1786 |
+
"learning_rate": 9.359764587513761e-05,
|
| 1787 |
+
"loss": 0.0057,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.9817333851535173,
|
| 1792 |
+
"grad_norm": 0.09211496249761421,
|
| 1793 |
+
"learning_rate": 9.351439893284317e-05,
|
| 1794 |
+
"loss": 0.0056,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.9895064127477653,
|
| 1799 |
+
"grad_norm": 0.14501544032188865,
|
| 1800 |
+
"learning_rate": 9.343065178589654e-05,
|
| 1801 |
+
"loss": 0.0056,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.9972794403420133,
|
| 1806 |
+
"grad_norm": 0.11456289818413888,
|
| 1807 |
+
"learning_rate": 9.334640539698396e-05,
|
| 1808 |
+
"loss": 0.0058,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.004663816556549,
|
| 1813 |
+
"grad_norm": 0.08068921187905599,
|
| 1814 |
+
"learning_rate": 9.326166073453051e-05,
|
| 1815 |
+
"loss": 0.0052,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.012436844150797,
|
| 1820 |
+
"grad_norm": 0.10032295518621465,
|
| 1821 |
+
"learning_rate": 9.317641877268901e-05,
|
| 1822 |
+
"loss": 0.0045,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.0202098717450445,
|
| 1827 |
+
"grad_norm": 0.0906748520799375,
|
| 1828 |
+
"learning_rate": 9.309068049132882e-05,
|
| 1829 |
+
"loss": 0.0044,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.0279828993392925,
|
| 1834 |
+
"grad_norm": 0.07982904566291607,
|
| 1835 |
+
"learning_rate": 9.300444687602457e-05,
|
| 1836 |
+
"loss": 0.0041,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.0357559269335406,
|
| 1841 |
+
"grad_norm": 0.1253437061299033,
|
| 1842 |
+
"learning_rate": 9.291771891804478e-05,
|
| 1843 |
+
"loss": 0.004,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.0435289545277886,
|
| 1848 |
+
"grad_norm": 0.09180533228491213,
|
| 1849 |
+
"learning_rate": 9.283049761434059e-05,
|
| 1850 |
+
"loss": 0.0041,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.0513019821220366,
|
| 1855 |
+
"grad_norm": 0.09238927797437002,
|
| 1856 |
+
"learning_rate": 9.27427839675342e-05,
|
| 1857 |
+
"loss": 0.004,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.0590750097162847,
|
| 1862 |
+
"grad_norm": 0.08085544322712232,
|
| 1863 |
+
"learning_rate": 9.26545789859073e-05,
|
| 1864 |
+
"loss": 0.004,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.0668480373105322,
|
| 1869 |
+
"grad_norm": 0.08278933444675111,
|
| 1870 |
+
"learning_rate": 9.256588368338968e-05,
|
| 1871 |
+
"loss": 0.0039,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.0746210649047803,
|
| 1876 |
+
"grad_norm": 0.0821491544443607,
|
| 1877 |
+
"learning_rate": 9.24766990795473e-05,
|
| 1878 |
+
"loss": 0.004,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.0823940924990283,
|
| 1883 |
+
"grad_norm": 0.0820536970017854,
|
| 1884 |
+
"learning_rate": 9.238702619957083e-05,
|
| 1885 |
+
"loss": 0.0039,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.0901671200932763,
|
| 1890 |
+
"grad_norm": 0.07783393489314461,
|
| 1891 |
+
"learning_rate": 9.229686607426368e-05,
|
| 1892 |
+
"loss": 0.0038,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.0979401476875243,
|
| 1897 |
+
"grad_norm": 0.08972295811517289,
|
| 1898 |
+
"learning_rate": 9.220621974003025e-05,
|
| 1899 |
+
"loss": 0.0038,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.1057131752817724,
|
| 1904 |
+
"grad_norm": 0.10850037565865495,
|
| 1905 |
+
"learning_rate": 9.211508823886398e-05,
|
| 1906 |
+
"loss": 0.0038,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.1134862028760204,
|
| 1911 |
+
"grad_norm": 0.1165050619921317,
|
| 1912 |
+
"learning_rate": 9.202347261833536e-05,
|
| 1913 |
+
"loss": 0.004,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.121259230470268,
|
| 1918 |
+
"grad_norm": 0.09613835587740172,
|
| 1919 |
+
"learning_rate": 9.193137393157994e-05,
|
| 1920 |
+
"loss": 0.004,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.129032258064516,
|
| 1925 |
+
"grad_norm": 0.08222637851192974,
|
| 1926 |
+
"learning_rate": 9.183879323728613e-05,
|
| 1927 |
+
"loss": 0.0041,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.136805285658764,
|
| 1932 |
+
"grad_norm": 0.09364974494864589,
|
| 1933 |
+
"learning_rate": 9.174573159968319e-05,
|
| 1934 |
+
"loss": 0.0041,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.144578313253012,
|
| 1939 |
+
"grad_norm": 0.09079410215789208,
|
| 1940 |
+
"learning_rate": 9.165219008852879e-05,
|
| 1941 |
+
"loss": 0.0037,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.15235134084726,
|
| 1946 |
+
"grad_norm": 0.08693641953192356,
|
| 1947 |
+
"learning_rate": 9.155816977909688e-05,
|
| 1948 |
+
"loss": 0.0038,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.160124368441508,
|
| 1953 |
+
"grad_norm": 0.08407058222727168,
|
| 1954 |
+
"learning_rate": 9.146367175216527e-05,
|
| 1955 |
+
"loss": 0.0038,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.167897396035756,
|
| 1960 |
+
"grad_norm": 0.08864727152361798,
|
| 1961 |
+
"learning_rate": 9.136869709400318e-05,
|
| 1962 |
+
"loss": 0.0037,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.1756704236300037,
|
| 1967 |
+
"grad_norm": 0.08585633309380411,
|
| 1968 |
+
"learning_rate": 9.127324689635879e-05,
|
| 1969 |
+
"loss": 0.0038,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.1834434512242518,
|
| 1974 |
+
"grad_norm": 0.09079038707160886,
|
| 1975 |
+
"learning_rate": 9.117732225644668e-05,
|
| 1976 |
+
"loss": 0.0037,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.1912164788185,
|
| 1981 |
+
"grad_norm": 0.08945120074693964,
|
| 1982 |
+
"learning_rate": 9.108092427693521e-05,
|
| 1983 |
+
"loss": 0.0039,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.198989506412748,
|
| 1988 |
+
"grad_norm": 0.09181352080010166,
|
| 1989 |
+
"learning_rate": 9.098405406593387e-05,
|
| 1990 |
+
"loss": 0.0038,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.206762534006996,
|
| 1995 |
+
"grad_norm": 0.08585652120551285,
|
| 1996 |
+
"learning_rate": 9.088671273698049e-05,
|
| 1997 |
+
"loss": 0.0038,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.214535561601244,
|
| 2002 |
+
"grad_norm": 0.08716662155761687,
|
| 2003 |
+
"learning_rate": 9.078890140902851e-05,
|
| 2004 |
+
"loss": 0.0037,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.2223085891954915,
|
| 2009 |
+
"grad_norm": 0.08274857092352111,
|
| 2010 |
+
"learning_rate": 9.069062120643406e-05,
|
| 2011 |
+
"loss": 0.0036,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.2300816167897395,
|
| 2016 |
+
"grad_norm": 0.08054936545881469,
|
| 2017 |
+
"learning_rate": 9.059187325894305e-05,
|
| 2018 |
+
"loss": 0.0037,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.2378546443839875,
|
| 2023 |
+
"grad_norm": 0.08928250602229335,
|
| 2024 |
+
"learning_rate": 9.04926587016782e-05,
|
| 2025 |
+
"loss": 0.0037,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.2456276719782355,
|
| 2030 |
+
"grad_norm": 0.14580475649103322,
|
| 2031 |
+
"learning_rate": 9.039297867512598e-05,
|
| 2032 |
+
"loss": 0.0039,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.2534006995724836,
|
| 2037 |
+
"grad_norm": 0.09684044113116728,
|
| 2038 |
+
"learning_rate": 9.02928343251235e-05,
|
| 2039 |
+
"loss": 0.0036,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.2611737271667316,
|
| 2044 |
+
"grad_norm": 4.610227312049106,
|
| 2045 |
+
"learning_rate": 9.019222680284531e-05,
|
| 2046 |
+
"loss": 0.0177,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.2689467547609796,
|
| 2051 |
+
"grad_norm": 3.9441098766180716,
|
| 2052 |
+
"learning_rate": 9.009115726479028e-05,
|
| 2053 |
+
"loss": 0.0845,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.276719782355227,
|
| 2058 |
+
"grad_norm": 0.30362597840942473,
|
| 2059 |
+
"learning_rate": 8.998962687276809e-05,
|
| 2060 |
+
"loss": 0.0336,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.2844928099494752,
|
| 2065 |
+
"grad_norm": 0.17872746957322608,
|
| 2066 |
+
"learning_rate": 8.988763679388614e-05,
|
| 2067 |
+
"loss": 0.0106,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.2922658375437233,
|
| 2072 |
+
"grad_norm": 0.11891620122269164,
|
| 2073 |
+
"learning_rate": 8.97851882005359e-05,
|
| 2074 |
+
"loss": 0.0069,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.3000388651379713,
|
| 2079 |
+
"grad_norm": 0.11931811737114671,
|
| 2080 |
+
"learning_rate": 8.968228227037961e-05,
|
| 2081 |
+
"loss": 0.0054,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.3078118927322193,
|
| 2086 |
+
"grad_norm": 0.11297205800033525,
|
| 2087 |
+
"learning_rate": 8.957892018633662e-05,
|
| 2088 |
+
"loss": 0.005,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.3155849203264673,
|
| 2093 |
+
"grad_norm": 0.08607374863983304,
|
| 2094 |
+
"learning_rate": 8.947510313656983e-05,
|
| 2095 |
+
"loss": 0.0046,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.323357947920715,
|
| 2100 |
+
"grad_norm": 0.08646791211129752,
|
| 2101 |
+
"learning_rate": 8.93708323144721e-05,
|
| 2102 |
+
"loss": 0.0043,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.331130975514963,
|
| 2107 |
+
"grad_norm": 0.09633865197527709,
|
| 2108 |
+
"learning_rate": 8.926610891865239e-05,
|
| 2109 |
+
"loss": 0.0042,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.338904003109211,
|
| 2114 |
+
"grad_norm": 0.07797950909850374,
|
| 2115 |
+
"learning_rate": 8.916093415292212e-05,
|
| 2116 |
+
"loss": 0.004,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.346677030703459,
|
| 2121 |
+
"grad_norm": 0.08803265774925108,
|
| 2122 |
+
"learning_rate": 8.905530922628129e-05,
|
| 2123 |
+
"loss": 0.0039,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.354450058297707,
|
| 2128 |
+
"grad_norm": 0.08578035815680632,
|
| 2129 |
+
"learning_rate": 8.894923535290452e-05,
|
| 2130 |
+
"loss": 0.0038,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.362223085891955,
|
| 2135 |
+
"grad_norm": 0.0723027097191447,
|
| 2136 |
+
"learning_rate": 8.884271375212713e-05,
|
| 2137 |
+
"loss": 0.0037,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.369996113486203,
|
| 2142 |
+
"grad_norm": 0.07933890872564231,
|
| 2143 |
+
"learning_rate": 8.873574564843123e-05,
|
| 2144 |
+
"loss": 0.0037,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.3777691410804507,
|
| 2149 |
+
"grad_norm": 0.08309597358808056,
|
| 2150 |
+
"learning_rate": 8.862833227143143e-05,
|
| 2151 |
+
"loss": 0.0037,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.3855421686746987,
|
| 2156 |
+
"grad_norm": 0.0796136928178535,
|
| 2157 |
+
"learning_rate": 8.852047485586093e-05,
|
| 2158 |
+
"loss": 0.0034,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.3933151962689467,
|
| 2163 |
+
"grad_norm": 0.0906501106792225,
|
| 2164 |
+
"learning_rate": 8.841217464155714e-05,
|
| 2165 |
+
"loss": 0.0036,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.4010882238631948,
|
| 2170 |
+
"grad_norm": 0.09232238343629308,
|
| 2171 |
+
"learning_rate": 8.830343287344753e-05,
|
| 2172 |
+
"loss": 0.0036,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.408861251457443,
|
| 2177 |
+
"grad_norm": 0.07677160549185906,
|
| 2178 |
+
"learning_rate": 8.819425080153533e-05,
|
| 2179 |
+
"loss": 0.0036,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.416634279051691,
|
| 2184 |
+
"grad_norm": 0.07965459624459517,
|
| 2185 |
+
"learning_rate": 8.808462968088512e-05,
|
| 2186 |
+
"loss": 0.0036,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.4244073066459384,
|
| 2191 |
+
"grad_norm": 0.08152518289735053,
|
| 2192 |
+
"learning_rate": 8.797457077160836e-05,
|
| 2193 |
+
"loss": 0.0036,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.4321803342401864,
|
| 2198 |
+
"grad_norm": 0.07840524642090889,
|
| 2199 |
+
"learning_rate": 8.786407533884902e-05,
|
| 2200 |
+
"loss": 0.0035,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.4399533618344345,
|
| 2205 |
+
"grad_norm": 0.0753906754915079,
|
| 2206 |
+
"learning_rate": 8.775314465276896e-05,
|
| 2207 |
+
"loss": 0.0034,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.4477263894286825,
|
| 2212 |
+
"grad_norm": 0.07553857495501508,
|
| 2213 |
+
"learning_rate": 8.764177998853334e-05,
|
| 2214 |
+
"loss": 0.0034,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.4554994170229305,
|
| 2219 |
+
"grad_norm": 0.07989895676116511,
|
| 2220 |
+
"learning_rate": 8.752998262629593e-05,
|
| 2221 |
+
"loss": 0.0034,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.4632724446171785,
|
| 2226 |
+
"grad_norm": 0.08604538052038023,
|
| 2227 |
+
"learning_rate": 8.741775385118447e-05,
|
| 2228 |
+
"loss": 0.0035,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.4710454722114266,
|
| 2233 |
+
"grad_norm": 0.07301243329178161,
|
| 2234 |
+
"learning_rate": 8.730509495328587e-05,
|
| 2235 |
+
"loss": 0.0034,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.478818499805674,
|
| 2240 |
+
"grad_norm": 0.07831526315789164,
|
| 2241 |
+
"learning_rate": 8.71920072276313e-05,
|
| 2242 |
+
"loss": 0.0034,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.486591527399922,
|
| 2247 |
+
"grad_norm": 0.10194734979674533,
|
| 2248 |
+
"learning_rate": 8.707849197418147e-05,
|
| 2249 |
+
"loss": 0.0034,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.49436455499417,
|
| 2254 |
+
"grad_norm": 0.07574798451139898,
|
| 2255 |
+
"learning_rate": 8.696455049781147e-05,
|
| 2256 |
+
"loss": 0.0035,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.5021375825884182,
|
| 2261 |
+
"grad_norm": 0.07681960051751799,
|
| 2262 |
+
"learning_rate": 8.6850184108296e-05,
|
| 2263 |
+
"loss": 0.0033,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.5099106101826663,
|
| 2268 |
+
"grad_norm": 0.07836534928409306,
|
| 2269 |
+
"learning_rate": 8.673539412029413e-05,
|
| 2270 |
+
"loss": 0.0033,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.517683637776914,
|
| 2275 |
+
"grad_norm": 0.08548600777071955,
|
| 2276 |
+
"learning_rate": 8.662018185333425e-05,
|
| 2277 |
+
"loss": 0.0034,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.525456665371162,
|
| 2282 |
+
"grad_norm": 0.0873567409210325,
|
| 2283 |
+
"learning_rate": 8.650454863179899e-05,
|
| 2284 |
+
"loss": 0.0032,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.53322969296541,
|
| 2289 |
+
"grad_norm": 0.08119732193506358,
|
| 2290 |
+
"learning_rate": 8.638849578490983e-05,
|
| 2291 |
+
"loss": 0.0032,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.541002720559658,
|
| 2296 |
+
"grad_norm": 0.07579199273478096,
|
| 2297 |
+
"learning_rate": 8.627202464671196e-05,
|
| 2298 |
+
"loss": 0.0031,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.548775748153906,
|
| 2303 |
+
"grad_norm": 0.07826696555904925,
|
| 2304 |
+
"learning_rate": 8.615513655605888e-05,
|
| 2305 |
+
"loss": 0.0031,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.556548775748154,
|
| 2310 |
+
"grad_norm": 0.07982815545083491,
|
| 2311 |
+
"learning_rate": 8.6037832856597e-05,
|
| 2312 |
+
"loss": 0.0031,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.564321803342402,
|
| 2317 |
+
"grad_norm": 0.08481539873916785,
|
| 2318 |
+
"learning_rate": 8.592011489675025e-05,
|
| 2319 |
+
"loss": 0.0031,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.57209483093665,
|
| 2324 |
+
"grad_norm": 0.07916097799005735,
|
| 2325 |
+
"learning_rate": 8.580198402970456e-05,
|
| 2326 |
+
"loss": 0.003,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.579867858530898,
|
| 2331 |
+
"grad_norm": 0.07473474197283404,
|
| 2332 |
+
"learning_rate": 8.56834416133922e-05,
|
| 2333 |
+
"loss": 0.003,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.5876408861251456,
|
| 2338 |
+
"grad_norm": 0.09130974953799981,
|
| 2339 |
+
"learning_rate": 8.556448901047638e-05,
|
| 2340 |
+
"loss": 0.0031,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.5954139137193937,
|
| 2345 |
+
"grad_norm": 0.1047401105445018,
|
| 2346 |
+
"learning_rate": 8.544512758833537e-05,
|
| 2347 |
+
"loss": 0.003,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.6031869413136417,
|
| 2352 |
+
"grad_norm": 0.06673230506522639,
|
| 2353 |
+
"learning_rate": 8.532535871904695e-05,
|
| 2354 |
+
"loss": 0.003,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.6109599689078897,
|
| 2359 |
+
"grad_norm": 0.07876904924601609,
|
| 2360 |
+
"learning_rate": 8.520518377937253e-05,
|
| 2361 |
+
"loss": 0.0031,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.6187329965021373,
|
| 2366 |
+
"grad_norm": 0.07009255329786931,
|
| 2367 |
+
"learning_rate": 8.508460415074137e-05,
|
| 2368 |
+
"loss": 0.0029,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.6265060240963853,
|
| 2373 |
+
"grad_norm": 0.07519024159981848,
|
| 2374 |
+
"learning_rate": 8.49636212192347e-05,
|
| 2375 |
+
"loss": 0.003,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.6342790516906334,
|
| 2380 |
+
"grad_norm": 0.08094145785537127,
|
| 2381 |
+
"learning_rate": 8.484223637556974e-05,
|
| 2382 |
+
"loss": 0.0031,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.6420520792848814,
|
| 2387 |
+
"grad_norm": 0.07736222204990759,
|
| 2388 |
+
"learning_rate": 8.472045101508382e-05,
|
| 2389 |
+
"loss": 0.003,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.6498251068791294,
|
| 2394 |
+
"grad_norm": 0.0776215180930073,
|
| 2395 |
+
"learning_rate": 8.459826653771821e-05,
|
| 2396 |
+
"loss": 0.003,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.6575981344733774,
|
| 2401 |
+
"grad_norm": 0.06742420975546093,
|
| 2402 |
+
"learning_rate": 8.44756843480021e-05,
|
| 2403 |
+
"loss": 0.0028,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.6653711620676255,
|
| 2408 |
+
"grad_norm": 0.0698399904847748,
|
| 2409 |
+
"learning_rate": 8.435270585503647e-05,
|
| 2410 |
+
"loss": 0.0029,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.6731441896618735,
|
| 2415 |
+
"grad_norm": 0.07591621006160705,
|
| 2416 |
+
"learning_rate": 8.422933247247786e-05,
|
| 2417 |
+
"loss": 0.0029,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.6809172172561215,
|
| 2422 |
+
"grad_norm": 0.06930363699655918,
|
| 2423 |
+
"learning_rate": 8.410556561852212e-05,
|
| 2424 |
+
"loss": 0.0029,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.688690244850369,
|
| 2429 |
+
"grad_norm": 0.07067953717607145,
|
| 2430 |
+
"learning_rate": 8.39814067158881e-05,
|
| 2431 |
+
"loss": 0.0028,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.696463272444617,
|
| 2436 |
+
"grad_norm": 0.077677315745107,
|
| 2437 |
+
"learning_rate": 8.385685719180132e-05,
|
| 2438 |
+
"loss": 0.0029,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.704236300038865,
|
| 2443 |
+
"grad_norm": 0.07570167649828695,
|
| 2444 |
+
"learning_rate": 8.373191847797754e-05,
|
| 2445 |
+
"loss": 0.0028,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.712009327633113,
|
| 2450 |
+
"grad_norm": 0.06955310603007175,
|
| 2451 |
+
"learning_rate": 8.360659201060639e-05,
|
| 2452 |
+
"loss": 0.0029,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.719782355227361,
|
| 2457 |
+
"grad_norm": 0.0659051677019647,
|
| 2458 |
+
"learning_rate": 8.348087923033465e-05,
|
| 2459 |
+
"loss": 0.0029,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.727555382821609,
|
| 2464 |
+
"grad_norm": 0.07181363122453262,
|
| 2465 |
+
"learning_rate": 8.335478158224996e-05,
|
| 2466 |
+
"loss": 0.0027,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.735328410415857,
|
| 2471 |
+
"grad_norm": 0.06585918912737933,
|
| 2472 |
+
"learning_rate": 8.3228300515864e-05,
|
| 2473 |
+
"loss": 0.0027,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.743101438010105,
|
| 2478 |
+
"grad_norm": 0.08664632265940041,
|
| 2479 |
+
"learning_rate": 8.310143748509588e-05,
|
| 2480 |
+
"loss": 0.0029,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.750874465604353,
|
| 2485 |
+
"grad_norm": 0.0773656650662347,
|
| 2486 |
+
"learning_rate": 8.297419394825554e-05,
|
| 2487 |
+
"loss": 0.0028,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.758647493198601,
|
| 2492 |
+
"grad_norm": 0.07477928292429042,
|
| 2493 |
+
"learning_rate": 8.28465713680268e-05,
|
| 2494 |
+
"loss": 0.0029,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.766420520792849,
|
| 2499 |
+
"grad_norm": 0.06614018293609181,
|
| 2500 |
+
"learning_rate": 8.271857121145073e-05,
|
| 2501 |
+
"loss": 0.0028,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.774193548387097,
|
| 2506 |
+
"grad_norm": 0.07953850815353782,
|
| 2507 |
+
"learning_rate": 8.25901949499086e-05,
|
| 2508 |
+
"loss": 0.0028,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.781966575981345,
|
| 2513 |
+
"grad_norm": 0.06857404401820201,
|
| 2514 |
+
"learning_rate": 8.246144405910517e-05,
|
| 2515 |
+
"loss": 0.0028,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.7897396035755926,
|
| 2520 |
+
"grad_norm": 0.06706667623803336,
|
| 2521 |
+
"learning_rate": 8.233232001905151e-05,
|
| 2522 |
+
"loss": 0.0029,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 2.7975126311698406,
|
| 2527 |
+
"grad_norm": 0.06548223623701806,
|
| 2528 |
+
"learning_rate": 8.220282431404818e-05,
|
| 2529 |
+
"loss": 0.0027,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
}
|
| 2532 |
+
],
|
| 2533 |
+
"logging_steps": 10,
|
| 2534 |
+
"max_steps": 10296,
|
| 2535 |
+
"num_input_tokens_seen": 0,
|
| 2536 |
+
"num_train_epochs": 8,
|
| 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": 265508976328704.0,
|
| 2551 |
+
"train_batch_size": 128,
|
| 2552 |
+
"trial_name": null,
|
| 2553 |
+
"trial_params": null
|
| 2554 |
+
}
|