Upload ensemble_teacher2_1/checkpoint-3600/trainer_state.json with huggingface_hub
Browse files
ensemble_teacher2_1/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": 0.7122366208329212,
|
| 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": 0.0019784350578692256,
|
| 14 |
+
"grad_norm": 19.637782613049417,
|
| 15 |
+
"learning_rate": 5.932762030323006e-07,
|
| 16 |
+
"loss": 8.2273,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.003956870115738451,
|
| 21 |
+
"grad_norm": 19.83226249327557,
|
| 22 |
+
"learning_rate": 1.2524719841793013e-06,
|
| 23 |
+
"loss": 8.167,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.005935305173607676,
|
| 28 |
+
"grad_norm": 17.787035299527364,
|
| 29 |
+
"learning_rate": 1.9116677653263018e-06,
|
| 30 |
+
"loss": 7.8304,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.007913740231476903,
|
| 35 |
+
"grad_norm": 6.953509182433938,
|
| 36 |
+
"learning_rate": 2.570863546473303e-06,
|
| 37 |
+
"loss": 7.0706,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.009892175289346126,
|
| 42 |
+
"grad_norm": 4.3091682221623735,
|
| 43 |
+
"learning_rate": 3.2300593276203036e-06,
|
| 44 |
+
"loss": 6.5597,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.011870610347215352,
|
| 49 |
+
"grad_norm": 3.0516932716653735,
|
| 50 |
+
"learning_rate": 3.889255108767304e-06,
|
| 51 |
+
"loss": 6.2279,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.013849045405084578,
|
| 56 |
+
"grad_norm": 2.7480308952739625,
|
| 57 |
+
"learning_rate": 4.548450889914305e-06,
|
| 58 |
+
"loss": 5.9488,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.015827480462953805,
|
| 63 |
+
"grad_norm": 3.533312550024619,
|
| 64 |
+
"learning_rate": 5.207646671061305e-06,
|
| 65 |
+
"loss": 5.6328,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.01780591552082303,
|
| 70 |
+
"grad_norm": 3.654444836350219,
|
| 71 |
+
"learning_rate": 5.866842452208306e-06,
|
| 72 |
+
"loss": 5.1732,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.019784350578692253,
|
| 77 |
+
"grad_norm": 4.086440552060253,
|
| 78 |
+
"learning_rate": 6.5260382333553066e-06,
|
| 79 |
+
"loss": 4.6653,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.02176278563656148,
|
| 84 |
+
"grad_norm": 4.138046221068996,
|
| 85 |
+
"learning_rate": 7.1852340145023085e-06,
|
| 86 |
+
"loss": 4.2061,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.023741220694430704,
|
| 91 |
+
"grad_norm": 5.7178401280922735,
|
| 92 |
+
"learning_rate": 7.844429795649308e-06,
|
| 93 |
+
"loss": 3.8123,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.02571965575229993,
|
| 98 |
+
"grad_norm": 5.106742130319203,
|
| 99 |
+
"learning_rate": 8.503625576796309e-06,
|
| 100 |
+
"loss": 3.512,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.027698090810169156,
|
| 105 |
+
"grad_norm": 4.573421695623524,
|
| 106 |
+
"learning_rate": 9.162821357943308e-06,
|
| 107 |
+
"loss": 3.2543,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.029676525868038383,
|
| 112 |
+
"grad_norm": 5.128687988140343,
|
| 113 |
+
"learning_rate": 9.82201713909031e-06,
|
| 114 |
+
"loss": 3.0517,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.03165496092590761,
|
| 119 |
+
"grad_norm": 4.880606343802238,
|
| 120 |
+
"learning_rate": 1.048121292023731e-05,
|
| 121 |
+
"loss": 2.8845,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.033633395983776834,
|
| 126 |
+
"grad_norm": 8.127044152874786,
|
| 127 |
+
"learning_rate": 1.1140408701384312e-05,
|
| 128 |
+
"loss": 2.748,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.03561183104164606,
|
| 133 |
+
"grad_norm": 8.485194090796208,
|
| 134 |
+
"learning_rate": 1.1799604482531313e-05,
|
| 135 |
+
"loss": 2.6414,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.03759026609951528,
|
| 140 |
+
"grad_norm": 6.612294272350823,
|
| 141 |
+
"learning_rate": 1.2458800263678314e-05,
|
| 142 |
+
"loss": 2.5598,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.039568701157384506,
|
| 147 |
+
"grad_norm": 12.16444744207085,
|
| 148 |
+
"learning_rate": 1.3117996044825315e-05,
|
| 149 |
+
"loss": 2.4886,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.04154713621525374,
|
| 154 |
+
"grad_norm": 18.63069790020457,
|
| 155 |
+
"learning_rate": 1.3777191825972313e-05,
|
| 156 |
+
"loss": 2.4395,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.04352557127312296,
|
| 161 |
+
"grad_norm": 11.412955192769877,
|
| 162 |
+
"learning_rate": 1.4436387607119314e-05,
|
| 163 |
+
"loss": 2.383,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.045504006330992185,
|
| 168 |
+
"grad_norm": 11.202485180848445,
|
| 169 |
+
"learning_rate": 1.5095583388266315e-05,
|
| 170 |
+
"loss": 2.3341,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.04748244138886141,
|
| 175 |
+
"grad_norm": 31.060419144311027,
|
| 176 |
+
"learning_rate": 1.5754779169413318e-05,
|
| 177 |
+
"loss": 2.3041,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.04946087644673064,
|
| 182 |
+
"grad_norm": 22.03166390729224,
|
| 183 |
+
"learning_rate": 1.6413974950560317e-05,
|
| 184 |
+
"loss": 2.2843,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.05143931150459986,
|
| 189 |
+
"grad_norm": 32.84287275099665,
|
| 190 |
+
"learning_rate": 1.707317073170732e-05,
|
| 191 |
+
"loss": 2.2874,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.05341774656246909,
|
| 196 |
+
"grad_norm": 13.792898956806326,
|
| 197 |
+
"learning_rate": 1.7732366512854316e-05,
|
| 198 |
+
"loss": 2.2134,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.05539618162033831,
|
| 203 |
+
"grad_norm": 9.703481654155626,
|
| 204 |
+
"learning_rate": 1.839156229400132e-05,
|
| 205 |
+
"loss": 2.1511,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.057374616678207535,
|
| 210 |
+
"grad_norm": 18.160904957620193,
|
| 211 |
+
"learning_rate": 1.905075807514832e-05,
|
| 212 |
+
"loss": 2.1061,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.059353051736076766,
|
| 217 |
+
"grad_norm": 31.18164400199025,
|
| 218 |
+
"learning_rate": 1.970995385629532e-05,
|
| 219 |
+
"loss": 2.0923,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.06133148679394599,
|
| 224 |
+
"grad_norm": 18.352406637861762,
|
| 225 |
+
"learning_rate": 2.0369149637442324e-05,
|
| 226 |
+
"loss": 2.0124,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.06330992185181522,
|
| 231 |
+
"grad_norm": 13.168827033502875,
|
| 232 |
+
"learning_rate": 2.102834541858932e-05,
|
| 233 |
+
"loss": 1.9455,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.06528835690968444,
|
| 238 |
+
"grad_norm": 12.98291760357499,
|
| 239 |
+
"learning_rate": 2.1687541199736323e-05,
|
| 240 |
+
"loss": 1.8702,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.06726679196755367,
|
| 245 |
+
"grad_norm": 11.475153571817428,
|
| 246 |
+
"learning_rate": 2.2346736980883322e-05,
|
| 247 |
+
"loss": 1.7829,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.06924522702542289,
|
| 252 |
+
"grad_norm": 14.551237589997115,
|
| 253 |
+
"learning_rate": 2.3005932762030325e-05,
|
| 254 |
+
"loss": 1.6901,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.07122366208329212,
|
| 259 |
+
"grad_norm": 9.667096995303945,
|
| 260 |
+
"learning_rate": 2.3665128543177324e-05,
|
| 261 |
+
"loss": 1.5745,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.07320209714116134,
|
| 266 |
+
"grad_norm": 13.163411291543234,
|
| 267 |
+
"learning_rate": 2.4324324324324327e-05,
|
| 268 |
+
"loss": 1.4658,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.07518053219903056,
|
| 273 |
+
"grad_norm": 9.940351335922,
|
| 274 |
+
"learning_rate": 2.4983520105471327e-05,
|
| 275 |
+
"loss": 1.3668,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.07715896725689979,
|
| 280 |
+
"grad_norm": 13.992341034680026,
|
| 281 |
+
"learning_rate": 2.564271588661833e-05,
|
| 282 |
+
"loss": 1.2807,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.07913740231476901,
|
| 287 |
+
"grad_norm": 7.4831699972814105,
|
| 288 |
+
"learning_rate": 2.630191166776533e-05,
|
| 289 |
+
"loss": 1.211,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.08111583737263825,
|
| 294 |
+
"grad_norm": 9.56614415520462,
|
| 295 |
+
"learning_rate": 2.6961107448912325e-05,
|
| 296 |
+
"loss": 1.1591,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.08309427243050747,
|
| 301 |
+
"grad_norm": 9.870662837309114,
|
| 302 |
+
"learning_rate": 2.762030323005933e-05,
|
| 303 |
+
"loss": 1.1007,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.0850727074883767,
|
| 308 |
+
"grad_norm": 7.296161468394743,
|
| 309 |
+
"learning_rate": 2.8279499011206327e-05,
|
| 310 |
+
"loss": 1.0605,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.08705114254624592,
|
| 315 |
+
"grad_norm": 9.934379350610987,
|
| 316 |
+
"learning_rate": 2.8938694792353333e-05,
|
| 317 |
+
"loss": 1.0221,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.08902957760411515,
|
| 322 |
+
"grad_norm": 8.322773668200096,
|
| 323 |
+
"learning_rate": 2.959789057350033e-05,
|
| 324 |
+
"loss": 0.9699,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.09100801266198437,
|
| 329 |
+
"grad_norm": 11.476438775344631,
|
| 330 |
+
"learning_rate": 3.025708635464733e-05,
|
| 331 |
+
"loss": 0.9331,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.0929864477198536,
|
| 336 |
+
"grad_norm": 7.972860756568456,
|
| 337 |
+
"learning_rate": 3.0916282135794335e-05,
|
| 338 |
+
"loss": 0.9203,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.09496488277772282,
|
| 343 |
+
"grad_norm": 11.520251589320083,
|
| 344 |
+
"learning_rate": 3.1575477916941334e-05,
|
| 345 |
+
"loss": 0.8832,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.09694331783559204,
|
| 350 |
+
"grad_norm": 9.465237203788785,
|
| 351 |
+
"learning_rate": 3.2234673698088334e-05,
|
| 352 |
+
"loss": 0.8484,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.09892175289346128,
|
| 357 |
+
"grad_norm": 13.258709856888437,
|
| 358 |
+
"learning_rate": 3.289386947923533e-05,
|
| 359 |
+
"loss": 0.8216,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.1009001879513305,
|
| 364 |
+
"grad_norm": 9.067024221441216,
|
| 365 |
+
"learning_rate": 3.355306526038233e-05,
|
| 366 |
+
"loss": 0.806,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.10287862300919973,
|
| 371 |
+
"grad_norm": 9.967066995989914,
|
| 372 |
+
"learning_rate": 3.421226104152934e-05,
|
| 373 |
+
"loss": 0.7768,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.10485705806706895,
|
| 378 |
+
"grad_norm": 6.990448616699275,
|
| 379 |
+
"learning_rate": 3.487145682267634e-05,
|
| 380 |
+
"loss": 0.7528,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.10683549312493817,
|
| 385 |
+
"grad_norm": 8.863709916752113,
|
| 386 |
+
"learning_rate": 3.553065260382334e-05,
|
| 387 |
+
"loss": 0.7225,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.1088139281828074,
|
| 392 |
+
"grad_norm": 8.219483179811172,
|
| 393 |
+
"learning_rate": 3.618984838497034e-05,
|
| 394 |
+
"loss": 0.6902,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.11079236324067662,
|
| 399 |
+
"grad_norm": 8.18954618815931,
|
| 400 |
+
"learning_rate": 3.684904416611734e-05,
|
| 401 |
+
"loss": 0.6664,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.11277079829854585,
|
| 406 |
+
"grad_norm": 6.99844977800662,
|
| 407 |
+
"learning_rate": 3.750823994726434e-05,
|
| 408 |
+
"loss": 0.6444,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.11474923335641507,
|
| 413 |
+
"grad_norm": 7.503046833377609,
|
| 414 |
+
"learning_rate": 3.8167435728411335e-05,
|
| 415 |
+
"loss": 0.6149,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.11672766841428431,
|
| 420 |
+
"grad_norm": 6.722656169187286,
|
| 421 |
+
"learning_rate": 3.882663150955834e-05,
|
| 422 |
+
"loss": 0.6168,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.11870610347215353,
|
| 427 |
+
"grad_norm": 7.030946593625458,
|
| 428 |
+
"learning_rate": 3.948582729070534e-05,
|
| 429 |
+
"loss": 0.5881,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.12068453853002276,
|
| 434 |
+
"grad_norm": 9.774246488301465,
|
| 435 |
+
"learning_rate": 4.014502307185235e-05,
|
| 436 |
+
"loss": 0.5607,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.12266297358789198,
|
| 441 |
+
"grad_norm": 7.078903913735779,
|
| 442 |
+
"learning_rate": 4.0804218852999346e-05,
|
| 443 |
+
"loss": 0.5405,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.1246414086457612,
|
| 448 |
+
"grad_norm": 7.402770055900705,
|
| 449 |
+
"learning_rate": 4.146341463414634e-05,
|
| 450 |
+
"loss": 0.5212,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.12661984370363044,
|
| 455 |
+
"grad_norm": 7.182019494715211,
|
| 456 |
+
"learning_rate": 4.2122610415293345e-05,
|
| 457 |
+
"loss": 0.4983,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.12859827876149965,
|
| 462 |
+
"grad_norm": 5.276303108063209,
|
| 463 |
+
"learning_rate": 4.2781806196440345e-05,
|
| 464 |
+
"loss": 0.4771,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.1305767138193689,
|
| 469 |
+
"grad_norm": 4.7737357936486235,
|
| 470 |
+
"learning_rate": 4.344100197758735e-05,
|
| 471 |
+
"loss": 0.4604,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.1325551488772381,
|
| 476 |
+
"grad_norm": 5.4871112446079655,
|
| 477 |
+
"learning_rate": 4.4100197758734343e-05,
|
| 478 |
+
"loss": 0.4532,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.13453358393510734,
|
| 483 |
+
"grad_norm": 6.595412809013484,
|
| 484 |
+
"learning_rate": 4.475939353988134e-05,
|
| 485 |
+
"loss": 0.4311,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.13651201899297655,
|
| 490 |
+
"grad_norm": 4.841829206426501,
|
| 491 |
+
"learning_rate": 4.541858932102835e-05,
|
| 492 |
+
"loss": 0.4181,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.13849045405084578,
|
| 497 |
+
"grad_norm": 4.7217934333031435,
|
| 498 |
+
"learning_rate": 4.607778510217535e-05,
|
| 499 |
+
"loss": 0.3928,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.140468889108715,
|
| 504 |
+
"grad_norm": 6.026086612922775,
|
| 505 |
+
"learning_rate": 4.673698088332235e-05,
|
| 506 |
+
"loss": 0.3897,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.14244732416658423,
|
| 511 |
+
"grad_norm": 4.396456746375336,
|
| 512 |
+
"learning_rate": 4.739617666446935e-05,
|
| 513 |
+
"loss": 0.3701,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.14442575922445347,
|
| 518 |
+
"grad_norm": 6.14790077100577,
|
| 519 |
+
"learning_rate": 4.805537244561635e-05,
|
| 520 |
+
"loss": 0.3488,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.14640419428232268,
|
| 525 |
+
"grad_norm": 5.390368007008413,
|
| 526 |
+
"learning_rate": 4.871456822676335e-05,
|
| 527 |
+
"loss": 0.3357,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.14838262934019192,
|
| 532 |
+
"grad_norm": 4.441570962528891,
|
| 533 |
+
"learning_rate": 4.937376400791035e-05,
|
| 534 |
+
"loss": 0.3203,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.15036106439806113,
|
| 539 |
+
"grad_norm": 5.270754499685052,
|
| 540 |
+
"learning_rate": 5.003295978905735e-05,
|
| 541 |
+
"loss": 0.3071,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.15233949945593037,
|
| 546 |
+
"grad_norm": 4.071551636960703,
|
| 547 |
+
"learning_rate": 5.069215557020436e-05,
|
| 548 |
+
"loss": 0.2991,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.15431793451379958,
|
| 553 |
+
"grad_norm": 3.0561647726741783,
|
| 554 |
+
"learning_rate": 5.135135135135135e-05,
|
| 555 |
+
"loss": 0.2832,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.15629636957166881,
|
| 560 |
+
"grad_norm": 2.9516027116400427,
|
| 561 |
+
"learning_rate": 5.201054713249836e-05,
|
| 562 |
+
"loss": 0.2681,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.15827480462953802,
|
| 567 |
+
"grad_norm": 4.621200447731901,
|
| 568 |
+
"learning_rate": 5.2669742913645356e-05,
|
| 569 |
+
"loss": 0.26,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.16025323968740726,
|
| 574 |
+
"grad_norm": 3.041429947319794,
|
| 575 |
+
"learning_rate": 5.332893869479235e-05,
|
| 576 |
+
"loss": 0.2471,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.1622316747452765,
|
| 581 |
+
"grad_norm": 2.995343942067045,
|
| 582 |
+
"learning_rate": 5.3988134475939355e-05,
|
| 583 |
+
"loss": 0.2369,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.1642101098031457,
|
| 588 |
+
"grad_norm": 3.1918861691745795,
|
| 589 |
+
"learning_rate": 5.464733025708636e-05,
|
| 590 |
+
"loss": 0.2279,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.16618854486101495,
|
| 595 |
+
"grad_norm": 2.1160708161237194,
|
| 596 |
+
"learning_rate": 5.5306526038233354e-05,
|
| 597 |
+
"loss": 0.2168,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.16816697991888416,
|
| 602 |
+
"grad_norm": 2.849132153355752,
|
| 603 |
+
"learning_rate": 5.596572181938036e-05,
|
| 604 |
+
"loss": 0.2107,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.1701454149767534,
|
| 609 |
+
"grad_norm": 2.5306963483355673,
|
| 610 |
+
"learning_rate": 5.662491760052736e-05,
|
| 611 |
+
"loss": 0.1974,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.1721238500346226,
|
| 616 |
+
"grad_norm": 3.134461921573245,
|
| 617 |
+
"learning_rate": 5.7284113381674366e-05,
|
| 618 |
+
"loss": 0.1898,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.17410228509249184,
|
| 623 |
+
"grad_norm": 4.467213018542406,
|
| 624 |
+
"learning_rate": 5.794330916282136e-05,
|
| 625 |
+
"loss": 0.1918,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.17608072015036105,
|
| 630 |
+
"grad_norm": 1.9384617338789845,
|
| 631 |
+
"learning_rate": 5.8602504943968364e-05,
|
| 632 |
+
"loss": 0.1846,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.1780591552082303,
|
| 637 |
+
"grad_norm": 2.22914110082071,
|
| 638 |
+
"learning_rate": 5.9261700725115364e-05,
|
| 639 |
+
"loss": 0.176,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.18003759026609953,
|
| 644 |
+
"grad_norm": 2.6195670505855104,
|
| 645 |
+
"learning_rate": 5.9920896506262356e-05,
|
| 646 |
+
"loss": 0.1645,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.18201602532396874,
|
| 651 |
+
"grad_norm": 5.623481408126527,
|
| 652 |
+
"learning_rate": 6.058009228740936e-05,
|
| 653 |
+
"loss": 0.1614,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.18399446038183798,
|
| 658 |
+
"grad_norm": 2.541850203721254,
|
| 659 |
+
"learning_rate": 6.123928806855637e-05,
|
| 660 |
+
"loss": 0.1562,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.1859728954397072,
|
| 665 |
+
"grad_norm": 2.530270886035028,
|
| 666 |
+
"learning_rate": 6.189848384970337e-05,
|
| 667 |
+
"loss": 0.1498,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.18795133049757642,
|
| 672 |
+
"grad_norm": 1.8747639348929863,
|
| 673 |
+
"learning_rate": 6.255767963085037e-05,
|
| 674 |
+
"loss": 0.1418,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.18992976555544563,
|
| 679 |
+
"grad_norm": 1.6302710016776405,
|
| 680 |
+
"learning_rate": 6.321687541199737e-05,
|
| 681 |
+
"loss": 0.1353,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.19190820061331487,
|
| 686 |
+
"grad_norm": 1.374949112823051,
|
| 687 |
+
"learning_rate": 6.387607119314437e-05,
|
| 688 |
+
"loss": 0.1303,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.19388663567118408,
|
| 693 |
+
"grad_norm": 2.300554535314371,
|
| 694 |
+
"learning_rate": 6.453526697429137e-05,
|
| 695 |
+
"loss": 0.1262,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.19586507072905332,
|
| 700 |
+
"grad_norm": 1.6596926456190018,
|
| 701 |
+
"learning_rate": 6.519446275543837e-05,
|
| 702 |
+
"loss": 0.1338,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.19784350578692256,
|
| 707 |
+
"grad_norm": 1.6448857589126298,
|
| 708 |
+
"learning_rate": 6.585365853658538e-05,
|
| 709 |
+
"loss": 0.1282,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.19982194084479177,
|
| 714 |
+
"grad_norm": 1.4480849991974878,
|
| 715 |
+
"learning_rate": 6.651285431773236e-05,
|
| 716 |
+
"loss": 0.116,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.201800375902661,
|
| 721 |
+
"grad_norm": 1.8869991728949147,
|
| 722 |
+
"learning_rate": 6.717205009887936e-05,
|
| 723 |
+
"loss": 0.1104,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.20377881096053022,
|
| 728 |
+
"grad_norm": 1.6293318832625847,
|
| 729 |
+
"learning_rate": 6.783124588002638e-05,
|
| 730 |
+
"loss": 0.1061,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.20575724601839945,
|
| 735 |
+
"grad_norm": 1.2427464212813581,
|
| 736 |
+
"learning_rate": 6.849044166117338e-05,
|
| 737 |
+
"loss": 0.1022,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.20773568107626866,
|
| 742 |
+
"grad_norm": 1.2404060214766448,
|
| 743 |
+
"learning_rate": 6.914963744232038e-05,
|
| 744 |
+
"loss": 0.0974,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.2097141161341379,
|
| 749 |
+
"grad_norm": 6.9065034819409625,
|
| 750 |
+
"learning_rate": 6.980883322346737e-05,
|
| 751 |
+
"loss": 0.0977,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.2116925511920071,
|
| 756 |
+
"grad_norm": 2.190513083820012,
|
| 757 |
+
"learning_rate": 7.046802900461437e-05,
|
| 758 |
+
"loss": 0.1328,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.21367098624987635,
|
| 763 |
+
"grad_norm": 1.263188586172193,
|
| 764 |
+
"learning_rate": 7.112722478576137e-05,
|
| 765 |
+
"loss": 0.0917,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.2156494213077456,
|
| 770 |
+
"grad_norm": 1.9784069526825092,
|
| 771 |
+
"learning_rate": 7.178642056690837e-05,
|
| 772 |
+
"loss": 0.0871,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.2176278563656148,
|
| 777 |
+
"grad_norm": 0.8523617558822155,
|
| 778 |
+
"learning_rate": 7.244561634805537e-05,
|
| 779 |
+
"loss": 0.085,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.21960629142348403,
|
| 784 |
+
"grad_norm": 0.8144860765656456,
|
| 785 |
+
"learning_rate": 7.310481212920237e-05,
|
| 786 |
+
"loss": 0.0775,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.22158472648135324,
|
| 791 |
+
"grad_norm": 0.8322446467787168,
|
| 792 |
+
"learning_rate": 7.376400791034937e-05,
|
| 793 |
+
"loss": 0.072,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.22356316153922248,
|
| 798 |
+
"grad_norm": 0.8840235823881915,
|
| 799 |
+
"learning_rate": 7.442320369149638e-05,
|
| 800 |
+
"loss": 0.0781,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.2255415965970917,
|
| 805 |
+
"grad_norm": 0.7406476619212958,
|
| 806 |
+
"learning_rate": 7.508239947264338e-05,
|
| 807 |
+
"loss": 0.067,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.22752003165496093,
|
| 812 |
+
"grad_norm": 0.6135846645545082,
|
| 813 |
+
"learning_rate": 7.574159525379037e-05,
|
| 814 |
+
"loss": 0.0625,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.22949846671283014,
|
| 819 |
+
"grad_norm": 0.8497087947713994,
|
| 820 |
+
"learning_rate": 7.640079103493738e-05,
|
| 821 |
+
"loss": 0.0588,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.23147690177069938,
|
| 826 |
+
"grad_norm": 0.5684521607051449,
|
| 827 |
+
"learning_rate": 7.705998681608438e-05,
|
| 828 |
+
"loss": 0.0559,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.23345533682856862,
|
| 833 |
+
"grad_norm": 0.44642133269859746,
|
| 834 |
+
"learning_rate": 7.771918259723138e-05,
|
| 835 |
+
"loss": 0.0524,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.23543377188643783,
|
| 840 |
+
"grad_norm": 0.7417253246437843,
|
| 841 |
+
"learning_rate": 7.837837837837838e-05,
|
| 842 |
+
"loss": 0.049,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.23741220694430706,
|
| 847 |
+
"grad_norm": 0.6699824985744847,
|
| 848 |
+
"learning_rate": 7.903757415952538e-05,
|
| 849 |
+
"loss": 0.0489,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.23939064200217627,
|
| 854 |
+
"grad_norm": 0.4584409365864605,
|
| 855 |
+
"learning_rate": 7.969676994067239e-05,
|
| 856 |
+
"loss": 0.0459,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.2413690770600455,
|
| 861 |
+
"grad_norm": 0.44232745755322544,
|
| 862 |
+
"learning_rate": 8.035596572181938e-05,
|
| 863 |
+
"loss": 0.0436,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.24334751211791472,
|
| 868 |
+
"grad_norm": 0.3867957221399885,
|
| 869 |
+
"learning_rate": 8.101516150296639e-05,
|
| 870 |
+
"loss": 0.0407,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.24532594717578396,
|
| 875 |
+
"grad_norm": 0.41432418911839947,
|
| 876 |
+
"learning_rate": 8.167435728411339e-05,
|
| 877 |
+
"loss": 0.0401,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.24730438223365317,
|
| 882 |
+
"grad_norm": 0.43958767606215066,
|
| 883 |
+
"learning_rate": 8.233355306526038e-05,
|
| 884 |
+
"loss": 0.0385,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.2492828172915224,
|
| 889 |
+
"grad_norm": 2.3478935687105587,
|
| 890 |
+
"learning_rate": 8.299274884640739e-05,
|
| 891 |
+
"loss": 0.0391,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.2512612523493916,
|
| 896 |
+
"grad_norm": 0.7921257388969846,
|
| 897 |
+
"learning_rate": 8.365194462755439e-05,
|
| 898 |
+
"loss": 0.0391,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.2532396874072609,
|
| 903 |
+
"grad_norm": 0.4159095258664668,
|
| 904 |
+
"learning_rate": 8.431114040870139e-05,
|
| 905 |
+
"loss": 0.0364,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.2552181224651301,
|
| 910 |
+
"grad_norm": 0.3574794027980809,
|
| 911 |
+
"learning_rate": 8.497033618984839e-05,
|
| 912 |
+
"loss": 0.0347,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.2571965575229993,
|
| 917 |
+
"grad_norm": 0.3793357762723568,
|
| 918 |
+
"learning_rate": 8.562953197099539e-05,
|
| 919 |
+
"loss": 0.0338,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.2591749925808685,
|
| 924 |
+
"grad_norm": 0.39061129654899224,
|
| 925 |
+
"learning_rate": 8.62887277521424e-05,
|
| 926 |
+
"loss": 0.0322,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.2611534276387378,
|
| 931 |
+
"grad_norm": 0.3507422200583872,
|
| 932 |
+
"learning_rate": 8.694792353328939e-05,
|
| 933 |
+
"loss": 0.0318,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.263131862696607,
|
| 938 |
+
"grad_norm": 0.3316497467819614,
|
| 939 |
+
"learning_rate": 8.760711931443639e-05,
|
| 940 |
+
"loss": 0.0309,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.2651102977544762,
|
| 945 |
+
"grad_norm": 0.4422978758582334,
|
| 946 |
+
"learning_rate": 8.82663150955834e-05,
|
| 947 |
+
"loss": 0.0302,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.2670887328123454,
|
| 952 |
+
"grad_norm": 0.42532269482645996,
|
| 953 |
+
"learning_rate": 8.892551087673038e-05,
|
| 954 |
+
"loss": 0.0292,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.2690671678702147,
|
| 959 |
+
"grad_norm": 1.2751835724526872,
|
| 960 |
+
"learning_rate": 8.95847066578774e-05,
|
| 961 |
+
"loss": 0.0322,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.2710456029280839,
|
| 966 |
+
"grad_norm": 0.41974258023847216,
|
| 967 |
+
"learning_rate": 9.02439024390244e-05,
|
| 968 |
+
"loss": 0.0299,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.2730240379859531,
|
| 973 |
+
"grad_norm": 0.32580553825660563,
|
| 974 |
+
"learning_rate": 9.09030982201714e-05,
|
| 975 |
+
"loss": 0.0285,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.27500247304382236,
|
| 980 |
+
"grad_norm": 0.362899277572937,
|
| 981 |
+
"learning_rate": 9.15622940013184e-05,
|
| 982 |
+
"loss": 0.0272,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.27698090810169157,
|
| 987 |
+
"grad_norm": 0.31286682503756197,
|
| 988 |
+
"learning_rate": 9.22214897824654e-05,
|
| 989 |
+
"loss": 0.0263,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.2789593431595608,
|
| 994 |
+
"grad_norm": 0.26773424358272674,
|
| 995 |
+
"learning_rate": 9.288068556361241e-05,
|
| 996 |
+
"loss": 0.0255,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.28093777821743,
|
| 1001 |
+
"grad_norm": 0.40448389034455734,
|
| 1002 |
+
"learning_rate": 9.35398813447594e-05,
|
| 1003 |
+
"loss": 0.0252,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.28291621327529926,
|
| 1008 |
+
"grad_norm": 0.2783819383288086,
|
| 1009 |
+
"learning_rate": 9.41990771259064e-05,
|
| 1010 |
+
"loss": 0.0241,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.28489464833316847,
|
| 1015 |
+
"grad_norm": 0.29961272536024675,
|
| 1016 |
+
"learning_rate": 9.48582729070534e-05,
|
| 1017 |
+
"loss": 0.0236,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.2868730833910377,
|
| 1022 |
+
"grad_norm": 0.24347322009574,
|
| 1023 |
+
"learning_rate": 9.551746868820039e-05,
|
| 1024 |
+
"loss": 0.0228,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.28885151844890694,
|
| 1029 |
+
"grad_norm": 0.23839323438206578,
|
| 1030 |
+
"learning_rate": 9.61766644693474e-05,
|
| 1031 |
+
"loss": 0.0226,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.29082995350677615,
|
| 1036 |
+
"grad_norm": 0.26473311552442963,
|
| 1037 |
+
"learning_rate": 9.68358602504944e-05,
|
| 1038 |
+
"loss": 0.0217,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.29280838856464536,
|
| 1043 |
+
"grad_norm": 0.25862281572683987,
|
| 1044 |
+
"learning_rate": 9.74950560316414e-05,
|
| 1045 |
+
"loss": 0.0211,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.29478682362251457,
|
| 1050 |
+
"grad_norm": 0.2277270114727975,
|
| 1051 |
+
"learning_rate": 9.81542518127884e-05,
|
| 1052 |
+
"loss": 0.0205,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.29676525868038384,
|
| 1057 |
+
"grad_norm": 0.24087395405359055,
|
| 1058 |
+
"learning_rate": 9.88134475939354e-05,
|
| 1059 |
+
"loss": 0.02,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.29874369373825305,
|
| 1064 |
+
"grad_norm": 0.22063387099417836,
|
| 1065 |
+
"learning_rate": 9.94726433750824e-05,
|
| 1066 |
+
"loss": 0.0193,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.30072212879612226,
|
| 1071 |
+
"grad_norm": 0.20714882477980437,
|
| 1072 |
+
"learning_rate": 9.999999470139298e-05,
|
| 1073 |
+
"loss": 0.0187,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.30270056385399147,
|
| 1078 |
+
"grad_norm": 0.2302853547965489,
|
| 1079 |
+
"learning_rate": 9.99998092502654e-05,
|
| 1080 |
+
"loss": 0.0184,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.30467899891186073,
|
| 1085 |
+
"grad_norm": 0.24697567028746728,
|
| 1086 |
+
"learning_rate": 9.999935886991012e-05,
|
| 1087 |
+
"loss": 0.0184,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.30665743396972994,
|
| 1092 |
+
"grad_norm": 0.2162360843440974,
|
| 1093 |
+
"learning_rate": 9.99986435627135e-05,
|
| 1094 |
+
"loss": 0.0179,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.30863586902759915,
|
| 1099 |
+
"grad_norm": 0.22785434831280413,
|
| 1100 |
+
"learning_rate": 9.99976633324657e-05,
|
| 1101 |
+
"loss": 0.0174,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.3106143040854684,
|
| 1106 |
+
"grad_norm": 0.21011992583277403,
|
| 1107 |
+
"learning_rate": 9.999641818436056e-05,
|
| 1108 |
+
"loss": 0.0169,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.31259273914333763,
|
| 1113 |
+
"grad_norm": 0.23882667469933117,
|
| 1114 |
+
"learning_rate": 9.999490812499562e-05,
|
| 1115 |
+
"loss": 0.0169,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.31457117420120684,
|
| 1120 |
+
"grad_norm": 0.22869025276994928,
|
| 1121 |
+
"learning_rate": 9.999313316237211e-05,
|
| 1122 |
+
"loss": 0.0162,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.31654960925907605,
|
| 1127 |
+
"grad_norm": 0.19936558362244672,
|
| 1128 |
+
"learning_rate": 9.999109330589484e-05,
|
| 1129 |
+
"loss": 0.0158,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.3185280443169453,
|
| 1134 |
+
"grad_norm": 0.18776511028399503,
|
| 1135 |
+
"learning_rate": 9.998878856637221e-05,
|
| 1136 |
+
"loss": 0.0157,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.3205064793748145,
|
| 1141 |
+
"grad_norm": 0.22047831859103678,
|
| 1142 |
+
"learning_rate": 9.998621895601612e-05,
|
| 1143 |
+
"loss": 0.0151,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.32248491443268373,
|
| 1148 |
+
"grad_norm": 0.21909712933711048,
|
| 1149 |
+
"learning_rate": 9.998338448844192e-05,
|
| 1150 |
+
"loss": 0.0153,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.324463349490553,
|
| 1155 |
+
"grad_norm": 0.18342955578512224,
|
| 1156 |
+
"learning_rate": 9.998028517866833e-05,
|
| 1157 |
+
"loss": 0.0147,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.3264417845484222,
|
| 1162 |
+
"grad_norm": 0.19614528373745874,
|
| 1163 |
+
"learning_rate": 9.997692104311738e-05,
|
| 1164 |
+
"loss": 0.0144,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.3284202196062914,
|
| 1169 |
+
"grad_norm": 0.1910028678576799,
|
| 1170 |
+
"learning_rate": 9.997329209961429e-05,
|
| 1171 |
+
"loss": 0.0141,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.33039865466416063,
|
| 1176 |
+
"grad_norm": 0.1875581912159569,
|
| 1177 |
+
"learning_rate": 9.996939836738737e-05,
|
| 1178 |
+
"loss": 0.0138,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.3323770897220299,
|
| 1183 |
+
"grad_norm": 0.21318059322872368,
|
| 1184 |
+
"learning_rate": 9.9965239867068e-05,
|
| 1185 |
+
"loss": 0.0138,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.3343555247798991,
|
| 1190 |
+
"grad_norm": 0.15100583828371544,
|
| 1191 |
+
"learning_rate": 9.996081662069043e-05,
|
| 1192 |
+
"loss": 0.0138,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.3363339598377683,
|
| 1197 |
+
"grad_norm": 0.16665345852934302,
|
| 1198 |
+
"learning_rate": 9.995612865169168e-05,
|
| 1199 |
+
"loss": 0.0136,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.3383123948956375,
|
| 1204 |
+
"grad_norm": 0.9719045950170202,
|
| 1205 |
+
"learning_rate": 9.995117598491145e-05,
|
| 1206 |
+
"loss": 0.0277,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.3402908299535068,
|
| 1211 |
+
"grad_norm": 3.161728848159066,
|
| 1212 |
+
"learning_rate": 9.994595864659197e-05,
|
| 1213 |
+
"loss": 0.0406,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.342269265011376,
|
| 1218 |
+
"grad_norm": 3.6147429552746417,
|
| 1219 |
+
"learning_rate": 9.994047666437785e-05,
|
| 1220 |
+
"loss": 0.0542,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.3442477000692452,
|
| 1225 |
+
"grad_norm": 0.2735534000567824,
|
| 1226 |
+
"learning_rate": 9.993473006731592e-05,
|
| 1227 |
+
"loss": 0.022,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.3462261351271145,
|
| 1232 |
+
"grad_norm": 0.1810803504663682,
|
| 1233 |
+
"learning_rate": 9.992871888585518e-05,
|
| 1234 |
+
"loss": 0.016,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.3482045701849837,
|
| 1239 |
+
"grad_norm": 0.14853009791786845,
|
| 1240 |
+
"learning_rate": 9.992244315184647e-05,
|
| 1241 |
+
"loss": 0.0143,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.3501830052428529,
|
| 1246 |
+
"grad_norm": 0.1508942913302995,
|
| 1247 |
+
"learning_rate": 9.991590289854243e-05,
|
| 1248 |
+
"loss": 0.0136,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.3521614403007221,
|
| 1253 |
+
"grad_norm": 0.14224740250692022,
|
| 1254 |
+
"learning_rate": 9.990909816059725e-05,
|
| 1255 |
+
"loss": 0.0126,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.35413987535859137,
|
| 1260 |
+
"grad_norm": 0.13888816460678802,
|
| 1261 |
+
"learning_rate": 9.99020289740666e-05,
|
| 1262 |
+
"loss": 0.0124,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.3561183104164606,
|
| 1267 |
+
"grad_norm": 0.14941569438589294,
|
| 1268 |
+
"learning_rate": 9.989469537640728e-05,
|
| 1269 |
+
"loss": 0.0122,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.3580967454743298,
|
| 1274 |
+
"grad_norm": 0.14008372469353464,
|
| 1275 |
+
"learning_rate": 9.988709740647713e-05,
|
| 1276 |
+
"loss": 0.0118,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.36007518053219906,
|
| 1281 |
+
"grad_norm": 0.1471251068367602,
|
| 1282 |
+
"learning_rate": 9.987923510453478e-05,
|
| 1283 |
+
"loss": 0.0119,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.36205361559006827,
|
| 1288 |
+
"grad_norm": 0.11583487666072863,
|
| 1289 |
+
"learning_rate": 9.987110851223947e-05,
|
| 1290 |
+
"loss": 0.0118,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.3640320506479375,
|
| 1295 |
+
"grad_norm": 0.14009606580200387,
|
| 1296 |
+
"learning_rate": 9.986271767265079e-05,
|
| 1297 |
+
"loss": 0.0117,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.3660104857058067,
|
| 1302 |
+
"grad_norm": 0.15359018530615084,
|
| 1303 |
+
"learning_rate": 9.985406263022849e-05,
|
| 1304 |
+
"loss": 0.0115,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.36798892076367595,
|
| 1309 |
+
"grad_norm": 0.13794481098483324,
|
| 1310 |
+
"learning_rate": 9.984514343083221e-05,
|
| 1311 |
+
"loss": 0.0115,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.36996735582154516,
|
| 1316 |
+
"grad_norm": 0.1338257809163921,
|
| 1317 |
+
"learning_rate": 9.983596012172127e-05,
|
| 1318 |
+
"loss": 0.0113,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.3719457908794144,
|
| 1323 |
+
"grad_norm": 0.11194861485256917,
|
| 1324 |
+
"learning_rate": 9.982651275155441e-05,
|
| 1325 |
+
"loss": 0.011,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.3739242259372836,
|
| 1330 |
+
"grad_norm": 0.13570239156331954,
|
| 1331 |
+
"learning_rate": 9.981680137038949e-05,
|
| 1332 |
+
"loss": 0.0108,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.37590266099515285,
|
| 1337 |
+
"grad_norm": 0.12947791555980262,
|
| 1338 |
+
"learning_rate": 9.980682602968329e-05,
|
| 1339 |
+
"loss": 0.0107,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.37788109605302206,
|
| 1344 |
+
"grad_norm": 0.11846270267188591,
|
| 1345 |
+
"learning_rate": 9.97965867822912e-05,
|
| 1346 |
+
"loss": 0.0105,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.37985953111089127,
|
| 1351 |
+
"grad_norm": 0.12775236471738088,
|
| 1352 |
+
"learning_rate": 9.978608368246693e-05,
|
| 1353 |
+
"loss": 0.0108,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.38183796616876053,
|
| 1358 |
+
"grad_norm": 0.11510206230282105,
|
| 1359 |
+
"learning_rate": 9.977531678586227e-05,
|
| 1360 |
+
"loss": 0.0102,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.38381640122662974,
|
| 1365 |
+
"grad_norm": 0.1167495841995747,
|
| 1366 |
+
"learning_rate": 9.976428614952675e-05,
|
| 1367 |
+
"loss": 0.0101,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.38579483628449895,
|
| 1372 |
+
"grad_norm": 0.1371366840889005,
|
| 1373 |
+
"learning_rate": 9.975299183190734e-05,
|
| 1374 |
+
"loss": 0.0102,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.38777327134236816,
|
| 1379 |
+
"grad_norm": 0.12495466419939394,
|
| 1380 |
+
"learning_rate": 9.974143389284818e-05,
|
| 1381 |
+
"loss": 0.0099,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.38975170640023743,
|
| 1386 |
+
"grad_norm": 0.1222265321586572,
|
| 1387 |
+
"learning_rate": 9.97296123935902e-05,
|
| 1388 |
+
"loss": 0.0099,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.39173014145810664,
|
| 1393 |
+
"grad_norm": 0.10581669983906558,
|
| 1394 |
+
"learning_rate": 9.971752739677088e-05,
|
| 1395 |
+
"loss": 0.0098,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.39370857651597585,
|
| 1400 |
+
"grad_norm": 0.1031261631363871,
|
| 1401 |
+
"learning_rate": 9.970517896642382e-05,
|
| 1402 |
+
"loss": 0.0096,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.3956870115738451,
|
| 1407 |
+
"grad_norm": 0.1098854853995119,
|
| 1408 |
+
"learning_rate": 9.969256716797848e-05,
|
| 1409 |
+
"loss": 0.0098,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.3976654466317143,
|
| 1414 |
+
"grad_norm": 0.11173021888171392,
|
| 1415 |
+
"learning_rate": 9.967969206825977e-05,
|
| 1416 |
+
"loss": 0.0093,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.39964388168958354,
|
| 1421 |
+
"grad_norm": 0.11939922842820136,
|
| 1422 |
+
"learning_rate": 9.96665537354878e-05,
|
| 1423 |
+
"loss": 0.0096,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.40162231674745275,
|
| 1428 |
+
"grad_norm": 0.12434579253503734,
|
| 1429 |
+
"learning_rate": 9.965315223927741e-05,
|
| 1430 |
+
"loss": 0.0095,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.403600751805322,
|
| 1435 |
+
"grad_norm": 0.10397011769498186,
|
| 1436 |
+
"learning_rate": 9.963948765063778e-05,
|
| 1437 |
+
"loss": 0.0096,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.4055791868631912,
|
| 1442 |
+
"grad_norm": 0.10874306421551215,
|
| 1443 |
+
"learning_rate": 9.962556004197217e-05,
|
| 1444 |
+
"loss": 0.0093,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.40755762192106043,
|
| 1449 |
+
"grad_norm": 0.13220931485486098,
|
| 1450 |
+
"learning_rate": 9.961136948707752e-05,
|
| 1451 |
+
"loss": 0.0093,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.40953605697892964,
|
| 1456 |
+
"grad_norm": 0.1296809440842965,
|
| 1457 |
+
"learning_rate": 9.959691606114393e-05,
|
| 1458 |
+
"loss": 0.0093,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.4115144920367989,
|
| 1463 |
+
"grad_norm": 0.1179090438361135,
|
| 1464 |
+
"learning_rate": 9.958219984075441e-05,
|
| 1465 |
+
"loss": 0.0093,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.4134929270946681,
|
| 1470 |
+
"grad_norm": 0.11884656631643735,
|
| 1471 |
+
"learning_rate": 9.956722090388438e-05,
|
| 1472 |
+
"loss": 0.009,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.4154713621525373,
|
| 1477 |
+
"grad_norm": 0.11716687338707497,
|
| 1478 |
+
"learning_rate": 9.955197932990132e-05,
|
| 1479 |
+
"loss": 0.009,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.4174497972104066,
|
| 1484 |
+
"grad_norm": 0.10664497146695569,
|
| 1485 |
+
"learning_rate": 9.953647519956433e-05,
|
| 1486 |
+
"loss": 0.0091,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.4194282322682758,
|
| 1491 |
+
"grad_norm": 0.11903304887750939,
|
| 1492 |
+
"learning_rate": 9.952070859502361e-05,
|
| 1493 |
+
"loss": 0.0088,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.421406667326145,
|
| 1498 |
+
"grad_norm": 0.09738834474502249,
|
| 1499 |
+
"learning_rate": 9.950467959982021e-05,
|
| 1500 |
+
"loss": 0.0088,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.4233851023840142,
|
| 1505 |
+
"grad_norm": 0.09920220236467324,
|
| 1506 |
+
"learning_rate": 9.948838829888543e-05,
|
| 1507 |
+
"loss": 0.0088,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.4253635374418835,
|
| 1512 |
+
"grad_norm": 0.09619743321048915,
|
| 1513 |
+
"learning_rate": 9.947183477854042e-05,
|
| 1514 |
+
"loss": 0.0088,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.4273419724997527,
|
| 1519 |
+
"grad_norm": 0.0974254141180528,
|
| 1520 |
+
"learning_rate": 9.945501912649579e-05,
|
| 1521 |
+
"loss": 0.0087,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.4293204075576219,
|
| 1526 |
+
"grad_norm": 0.09443148747455482,
|
| 1527 |
+
"learning_rate": 9.943794143185096e-05,
|
| 1528 |
+
"loss": 0.0087,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.4312988426154912,
|
| 1533 |
+
"grad_norm": 0.09350468571118675,
|
| 1534 |
+
"learning_rate": 9.942060178509396e-05,
|
| 1535 |
+
"loss": 0.0086,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.4332772776733604,
|
| 1540 |
+
"grad_norm": 0.09070704335772732,
|
| 1541 |
+
"learning_rate": 9.940300027810067e-05,
|
| 1542 |
+
"loss": 0.0085,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.4352557127312296,
|
| 1547 |
+
"grad_norm": 0.12110216964097328,
|
| 1548 |
+
"learning_rate": 9.938513700413454e-05,
|
| 1549 |
+
"loss": 0.0087,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.4372341477890988,
|
| 1554 |
+
"grad_norm": 0.09890163894885512,
|
| 1555 |
+
"learning_rate": 9.9367012057846e-05,
|
| 1556 |
+
"loss": 0.0085,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.43921258284696807,
|
| 1561 |
+
"grad_norm": 0.11128753159509124,
|
| 1562 |
+
"learning_rate": 9.934862553527198e-05,
|
| 1563 |
+
"loss": 0.0084,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.4411910179048373,
|
| 1568 |
+
"grad_norm": 0.10195595147527749,
|
| 1569 |
+
"learning_rate": 9.932997753383537e-05,
|
| 1570 |
+
"loss": 0.0084,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.4431694529627065,
|
| 1575 |
+
"grad_norm": 0.1035635647276319,
|
| 1576 |
+
"learning_rate": 9.931106815234459e-05,
|
| 1577 |
+
"loss": 0.0084,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.4451478880205757,
|
| 1582 |
+
"grad_norm": 0.12526375331576606,
|
| 1583 |
+
"learning_rate": 9.929189749099296e-05,
|
| 1584 |
+
"loss": 0.0082,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.44712632307844496,
|
| 1589 |
+
"grad_norm": 0.10124687119197032,
|
| 1590 |
+
"learning_rate": 9.927246565135826e-05,
|
| 1591 |
+
"loss": 0.0085,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.4491047581363142,
|
| 1596 |
+
"grad_norm": 0.12655923047390247,
|
| 1597 |
+
"learning_rate": 9.92527727364021e-05,
|
| 1598 |
+
"loss": 0.0084,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.4510831931941834,
|
| 1603 |
+
"grad_norm": 0.12234433416748676,
|
| 1604 |
+
"learning_rate": 9.923281885046949e-05,
|
| 1605 |
+
"loss": 0.0082,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.45306162825205265,
|
| 1610 |
+
"grad_norm": 0.10038433023346938,
|
| 1611 |
+
"learning_rate": 9.921260409928814e-05,
|
| 1612 |
+
"loss": 0.0081,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.45504006330992186,
|
| 1617 |
+
"grad_norm": 0.10592374911299951,
|
| 1618 |
+
"learning_rate": 9.919212858996804e-05,
|
| 1619 |
+
"loss": 0.0082,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.45701849836779107,
|
| 1624 |
+
"grad_norm": 0.11256516289439368,
|
| 1625 |
+
"learning_rate": 9.917139243100087e-05,
|
| 1626 |
+
"loss": 0.0083,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.4589969334256603,
|
| 1631 |
+
"grad_norm": 0.10941118440635397,
|
| 1632 |
+
"learning_rate": 9.915039573225926e-05,
|
| 1633 |
+
"loss": 0.0083,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.46097536848352955,
|
| 1638 |
+
"grad_norm": 0.09076166231590097,
|
| 1639 |
+
"learning_rate": 9.912913860499649e-05,
|
| 1640 |
+
"loss": 0.0081,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.46295380354139876,
|
| 1645 |
+
"grad_norm": 0.09539136003050565,
|
| 1646 |
+
"learning_rate": 9.910762116184562e-05,
|
| 1647 |
+
"loss": 0.0079,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.46493223859926797,
|
| 1652 |
+
"grad_norm": 0.09920800558172575,
|
| 1653 |
+
"learning_rate": 9.908584351681911e-05,
|
| 1654 |
+
"loss": 0.0079,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.46691067365713723,
|
| 1659 |
+
"grad_norm": 0.07883437880669862,
|
| 1660 |
+
"learning_rate": 9.906380578530809e-05,
|
| 1661 |
+
"loss": 0.0078,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.46888910871500644,
|
| 1666 |
+
"grad_norm": 0.08844392359513846,
|
| 1667 |
+
"learning_rate": 9.904150808408178e-05,
|
| 1668 |
+
"loss": 0.0076,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.47086754377287565,
|
| 1673 |
+
"grad_norm": 0.08782650023677235,
|
| 1674 |
+
"learning_rate": 9.901895053128689e-05,
|
| 1675 |
+
"loss": 0.0079,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.47284597883074486,
|
| 1680 |
+
"grad_norm": 0.10263452740495278,
|
| 1681 |
+
"learning_rate": 9.899613324644698e-05,
|
| 1682 |
+
"loss": 0.0078,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.4748244138886141,
|
| 1687 |
+
"grad_norm": 0.07539712980685903,
|
| 1688 |
+
"learning_rate": 9.897305635046183e-05,
|
| 1689 |
+
"loss": 0.0077,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.47680284894648334,
|
| 1694 |
+
"grad_norm": 0.09154103611980138,
|
| 1695 |
+
"learning_rate": 9.894971996560678e-05,
|
| 1696 |
+
"loss": 0.0078,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.47878128400435255,
|
| 1701 |
+
"grad_norm": 0.0949233826823644,
|
| 1702 |
+
"learning_rate": 9.89261242155321e-05,
|
| 1703 |
+
"loss": 0.0075,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.48075971906222176,
|
| 1708 |
+
"grad_norm": 0.08791399663054178,
|
| 1709 |
+
"learning_rate": 9.890226922526237e-05,
|
| 1710 |
+
"loss": 0.0076,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.482738154120091,
|
| 1715 |
+
"grad_norm": 0.08812362893437024,
|
| 1716 |
+
"learning_rate": 9.887815512119574e-05,
|
| 1717 |
+
"loss": 0.0077,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.48471658917796023,
|
| 1722 |
+
"grad_norm": 0.08865882789841384,
|
| 1723 |
+
"learning_rate": 9.885378203110331e-05,
|
| 1724 |
+
"loss": 0.0077,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.48669502423582944,
|
| 1729 |
+
"grad_norm": 0.08302609862666924,
|
| 1730 |
+
"learning_rate": 9.882915008412847e-05,
|
| 1731 |
+
"loss": 0.0075,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.4886734592936987,
|
| 1736 |
+
"grad_norm": 0.0894044049025858,
|
| 1737 |
+
"learning_rate": 9.880425941078617e-05,
|
| 1738 |
+
"loss": 0.0076,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.4906518943515679,
|
| 1743 |
+
"grad_norm": 0.0875976993567813,
|
| 1744 |
+
"learning_rate": 9.877911014296223e-05,
|
| 1745 |
+
"loss": 0.0075,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.49263032940943713,
|
| 1750 |
+
"grad_norm": 0.07532701277591465,
|
| 1751 |
+
"learning_rate": 9.875370241391269e-05,
|
| 1752 |
+
"loss": 0.0075,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.49460876446730634,
|
| 1757 |
+
"grad_norm": 0.07966759962069439,
|
| 1758 |
+
"learning_rate": 9.872803635826307e-05,
|
| 1759 |
+
"loss": 0.0075,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.4965871995251756,
|
| 1764 |
+
"grad_norm": 0.09858397509750241,
|
| 1765 |
+
"learning_rate": 9.870211211200766e-05,
|
| 1766 |
+
"loss": 0.0073,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.4985656345830448,
|
| 1771 |
+
"grad_norm": 0.09319554063457361,
|
| 1772 |
+
"learning_rate": 9.867592981250878e-05,
|
| 1773 |
+
"loss": 0.0074,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.500544069640914,
|
| 1778 |
+
"grad_norm": 0.12781117254542,
|
| 1779 |
+
"learning_rate": 9.864948959849609e-05,
|
| 1780 |
+
"loss": 0.0074,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.5025225046987832,
|
| 1785 |
+
"grad_norm": 0.26482548077665524,
|
| 1786 |
+
"learning_rate": 9.862279161006585e-05,
|
| 1787 |
+
"loss": 0.0098,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.5045009397566524,
|
| 1792 |
+
"grad_norm": 1.3380179010337696,
|
| 1793 |
+
"learning_rate": 9.859583598868012e-05,
|
| 1794 |
+
"loss": 0.0138,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.5064793748145218,
|
| 1799 |
+
"grad_norm": 0.5324160366432114,
|
| 1800 |
+
"learning_rate": 9.856862287716613e-05,
|
| 1801 |
+
"loss": 0.0651,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.508457809872391,
|
| 1806 |
+
"grad_norm": 0.20795257366814654,
|
| 1807 |
+
"learning_rate": 9.854115241971535e-05,
|
| 1808 |
+
"loss": 0.0156,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.5104362449302602,
|
| 1813 |
+
"grad_norm": 0.1358973049313345,
|
| 1814 |
+
"learning_rate": 9.85134247618829e-05,
|
| 1815 |
+
"loss": 0.0104,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.5124146799881294,
|
| 1820 |
+
"grad_norm": 0.11303985103375866,
|
| 1821 |
+
"learning_rate": 9.848544005058668e-05,
|
| 1822 |
+
"loss": 0.0088,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.5143931150459986,
|
| 1827 |
+
"grad_norm": 0.07561229737159056,
|
| 1828 |
+
"learning_rate": 9.845719843410663e-05,
|
| 1829 |
+
"loss": 0.0081,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.5163715501038678,
|
| 1834 |
+
"grad_norm": 0.08491679400058451,
|
| 1835 |
+
"learning_rate": 9.842870006208389e-05,
|
| 1836 |
+
"loss": 0.0078,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.518349985161737,
|
| 1841 |
+
"grad_norm": 0.08114020479286879,
|
| 1842 |
+
"learning_rate": 9.839994508552006e-05,
|
| 1843 |
+
"loss": 0.0077,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.5203284202196063,
|
| 1848 |
+
"grad_norm": 0.07439429710508837,
|
| 1849 |
+
"learning_rate": 9.837093365677644e-05,
|
| 1850 |
+
"loss": 0.0073,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.5223068552774756,
|
| 1855 |
+
"grad_norm": 0.07588934608576213,
|
| 1856 |
+
"learning_rate": 9.83416659295731e-05,
|
| 1857 |
+
"loss": 0.0075,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.5242852903353448,
|
| 1862 |
+
"grad_norm": 0.07727254521424115,
|
| 1863 |
+
"learning_rate": 9.831214205898814e-05,
|
| 1864 |
+
"loss": 0.0074,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.526263725393214,
|
| 1869 |
+
"grad_norm": 0.0770544746274172,
|
| 1870 |
+
"learning_rate": 9.82823622014569e-05,
|
| 1871 |
+
"loss": 0.0073,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.5282421604510832,
|
| 1876 |
+
"grad_norm": 0.07507052238833033,
|
| 1877 |
+
"learning_rate": 9.825232651477109e-05,
|
| 1878 |
+
"loss": 0.0072,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.5302205955089524,
|
| 1883 |
+
"grad_norm": 0.08038501538455785,
|
| 1884 |
+
"learning_rate": 9.82220351580779e-05,
|
| 1885 |
+
"loss": 0.0071,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.5321990305668216,
|
| 1890 |
+
"grad_norm": 0.07064608286766139,
|
| 1891 |
+
"learning_rate": 9.819148829187932e-05,
|
| 1892 |
+
"loss": 0.007,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.5341774656246908,
|
| 1897 |
+
"grad_norm": 0.06202757033774752,
|
| 1898 |
+
"learning_rate": 9.816068607803107e-05,
|
| 1899 |
+
"loss": 0.0071,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.5361559006825601,
|
| 1904 |
+
"grad_norm": 0.06498278906559196,
|
| 1905 |
+
"learning_rate": 9.812962867974191e-05,
|
| 1906 |
+
"loss": 0.007,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.5381343357404293,
|
| 1911 |
+
"grad_norm": 0.06889168223262838,
|
| 1912 |
+
"learning_rate": 9.809831626157274e-05,
|
| 1913 |
+
"loss": 0.0069,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.5401127707982986,
|
| 1918 |
+
"grad_norm": 0.07658902219040475,
|
| 1919 |
+
"learning_rate": 9.806674898943569e-05,
|
| 1920 |
+
"loss": 0.007,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.5420912058561678,
|
| 1925 |
+
"grad_norm": 0.065227283289949,
|
| 1926 |
+
"learning_rate": 9.803492703059323e-05,
|
| 1927 |
+
"loss": 0.007,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.544069640914037,
|
| 1932 |
+
"grad_norm": 0.07285925886476977,
|
| 1933 |
+
"learning_rate": 9.800285055365736e-05,
|
| 1934 |
+
"loss": 0.0069,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.5460480759719062,
|
| 1939 |
+
"grad_norm": 0.07094206575560595,
|
| 1940 |
+
"learning_rate": 9.797051972858867e-05,
|
| 1941 |
+
"loss": 0.0069,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.5480265110297754,
|
| 1946 |
+
"grad_norm": 0.07226595292182501,
|
| 1947 |
+
"learning_rate": 9.79379347266954e-05,
|
| 1948 |
+
"loss": 0.0068,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.5500049460876447,
|
| 1953 |
+
"grad_norm": 0.07467132239176189,
|
| 1954 |
+
"learning_rate": 9.79050957206326e-05,
|
| 1955 |
+
"loss": 0.007,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.5519833811455139,
|
| 1960 |
+
"grad_norm": 0.10178890401636641,
|
| 1961 |
+
"learning_rate": 9.787200288440119e-05,
|
| 1962 |
+
"loss": 0.0069,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.5539618162033831,
|
| 1967 |
+
"grad_norm": 0.08092189483997364,
|
| 1968 |
+
"learning_rate": 9.783865639334705e-05,
|
| 1969 |
+
"loss": 0.0069,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.5559402512612523,
|
| 1974 |
+
"grad_norm": 0.05878477071038792,
|
| 1975 |
+
"learning_rate": 9.780505642416001e-05,
|
| 1976 |
+
"loss": 0.0067,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.5579186863191216,
|
| 1981 |
+
"grad_norm": 0.05530654925471683,
|
| 1982 |
+
"learning_rate": 9.777120315487306e-05,
|
| 1983 |
+
"loss": 0.0066,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.5598971213769908,
|
| 1988 |
+
"grad_norm": 0.0686536401780331,
|
| 1989 |
+
"learning_rate": 9.773709676486129e-05,
|
| 1990 |
+
"loss": 0.0067,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.56187555643486,
|
| 1995 |
+
"grad_norm": 0.06035416233845031,
|
| 1996 |
+
"learning_rate": 9.770273743484097e-05,
|
| 1997 |
+
"loss": 0.0068,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.5638539914927293,
|
| 2002 |
+
"grad_norm": 0.05803243981437283,
|
| 2003 |
+
"learning_rate": 9.766812534686863e-05,
|
| 2004 |
+
"loss": 0.0066,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.5658324265505985,
|
| 2009 |
+
"grad_norm": 0.06995933649531104,
|
| 2010 |
+
"learning_rate": 9.763326068434e-05,
|
| 2011 |
+
"loss": 0.0066,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.5678108616084677,
|
| 2016 |
+
"grad_norm": 0.0605332765649623,
|
| 2017 |
+
"learning_rate": 9.75981436319892e-05,
|
| 2018 |
+
"loss": 0.0066,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.5697892966663369,
|
| 2023 |
+
"grad_norm": 0.06737579928910571,
|
| 2024 |
+
"learning_rate": 9.756277437588759e-05,
|
| 2025 |
+
"loss": 0.0067,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.5717677317242061,
|
| 2030 |
+
"grad_norm": 0.06189558705755954,
|
| 2031 |
+
"learning_rate": 9.752715310344288e-05,
|
| 2032 |
+
"loss": 0.0066,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.5737461667820754,
|
| 2037 |
+
"grad_norm": 0.06854647227258914,
|
| 2038 |
+
"learning_rate": 9.74912800033981e-05,
|
| 2039 |
+
"loss": 0.0067,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.5757246018399446,
|
| 2044 |
+
"grad_norm": 0.06334049765668723,
|
| 2045 |
+
"learning_rate": 9.745515526583065e-05,
|
| 2046 |
+
"loss": 0.0067,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.5777030368978139,
|
| 2051 |
+
"grad_norm": 0.07265687833634417,
|
| 2052 |
+
"learning_rate": 9.741877908215123e-05,
|
| 2053 |
+
"loss": 0.0067,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.5796814719556831,
|
| 2058 |
+
"grad_norm": 0.06912150801140689,
|
| 2059 |
+
"learning_rate": 9.738215164510286e-05,
|
| 2060 |
+
"loss": 0.0067,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.5816599070135523,
|
| 2065 |
+
"grad_norm": 0.06380076622359117,
|
| 2066 |
+
"learning_rate": 9.734527314875985e-05,
|
| 2067 |
+
"loss": 0.0066,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.5836383420714215,
|
| 2072 |
+
"grad_norm": 0.0748767974202622,
|
| 2073 |
+
"learning_rate": 9.730814378852678e-05,
|
| 2074 |
+
"loss": 0.0067,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.5856167771292907,
|
| 2079 |
+
"grad_norm": 0.06800504603033558,
|
| 2080 |
+
"learning_rate": 9.727076376113745e-05,
|
| 2081 |
+
"loss": 0.0065,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.5875952121871599,
|
| 2086 |
+
"grad_norm": 0.059588811804270385,
|
| 2087 |
+
"learning_rate": 9.723313326465384e-05,
|
| 2088 |
+
"loss": 0.0065,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.5895736472450291,
|
| 2093 |
+
"grad_norm": 0.05354398140471879,
|
| 2094 |
+
"learning_rate": 9.71952524984651e-05,
|
| 2095 |
+
"loss": 0.0064,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.5915520823028984,
|
| 2100 |
+
"grad_norm": 0.06992506261058357,
|
| 2101 |
+
"learning_rate": 9.715712166328644e-05,
|
| 2102 |
+
"loss": 0.0065,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.5935305173607677,
|
| 2107 |
+
"grad_norm": 0.07328603458194191,
|
| 2108 |
+
"learning_rate": 9.711874096115805e-05,
|
| 2109 |
+
"loss": 0.0064,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.5955089524186369,
|
| 2114 |
+
"grad_norm": 0.05746748382423511,
|
| 2115 |
+
"learning_rate": 9.708011059544415e-05,
|
| 2116 |
+
"loss": 0.0064,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.5974873874765061,
|
| 2121 |
+
"grad_norm": 0.05877461678841951,
|
| 2122 |
+
"learning_rate": 9.704123077083176e-05,
|
| 2123 |
+
"loss": 0.0065,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.5994658225343753,
|
| 2128 |
+
"grad_norm": 0.051196419888608125,
|
| 2129 |
+
"learning_rate": 9.700210169332968e-05,
|
| 2130 |
+
"loss": 0.0063,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.6014442575922445,
|
| 2135 |
+
"grad_norm": 0.06331233148415444,
|
| 2136 |
+
"learning_rate": 9.696272357026746e-05,
|
| 2137 |
+
"loss": 0.0063,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.6034226926501137,
|
| 2142 |
+
"grad_norm": 0.061860321869425514,
|
| 2143 |
+
"learning_rate": 9.692309661029421e-05,
|
| 2144 |
+
"loss": 0.0064,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.6054011277079829,
|
| 2149 |
+
"grad_norm": 0.067519120947699,
|
| 2150 |
+
"learning_rate": 9.68832210233775e-05,
|
| 2151 |
+
"loss": 0.0066,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.6073795627658523,
|
| 2156 |
+
"grad_norm": 0.07524005887985293,
|
| 2157 |
+
"learning_rate": 9.684309702080233e-05,
|
| 2158 |
+
"loss": 0.0065,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.6093579978237215,
|
| 2163 |
+
"grad_norm": 0.06070609348728366,
|
| 2164 |
+
"learning_rate": 9.680272481516992e-05,
|
| 2165 |
+
"loss": 0.0065,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.6113364328815907,
|
| 2170 |
+
"grad_norm": 0.12817129560256088,
|
| 2171 |
+
"learning_rate": 9.676210462039664e-05,
|
| 2172 |
+
"loss": 0.0064,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.6133148679394599,
|
| 2177 |
+
"grad_norm": 0.06537155344505334,
|
| 2178 |
+
"learning_rate": 9.672123665171285e-05,
|
| 2179 |
+
"loss": 0.0065,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.6152933029973291,
|
| 2184 |
+
"grad_norm": 0.08187746090238779,
|
| 2185 |
+
"learning_rate": 9.668012112566174e-05,
|
| 2186 |
+
"loss": 0.0065,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.6172717380551983,
|
| 2191 |
+
"grad_norm": 0.06284585792273872,
|
| 2192 |
+
"learning_rate": 9.663875826009824e-05,
|
| 2193 |
+
"loss": 0.0063,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.6192501731130675,
|
| 2198 |
+
"grad_norm": 0.058579617677076326,
|
| 2199 |
+
"learning_rate": 9.659714827418786e-05,
|
| 2200 |
+
"loss": 0.0064,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.6212286081709368,
|
| 2205 |
+
"grad_norm": 0.05671718811117095,
|
| 2206 |
+
"learning_rate": 9.655529138840543e-05,
|
| 2207 |
+
"loss": 0.0063,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.623207043228806,
|
| 2212 |
+
"grad_norm": 0.058903454628710904,
|
| 2213 |
+
"learning_rate": 9.651318782453407e-05,
|
| 2214 |
+
"loss": 0.0062,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.6251854782866753,
|
| 2219 |
+
"grad_norm": 0.05670298829502456,
|
| 2220 |
+
"learning_rate": 9.64708378056639e-05,
|
| 2221 |
+
"loss": 0.0062,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.6271639133445445,
|
| 2226 |
+
"grad_norm": 0.05428876166111194,
|
| 2227 |
+
"learning_rate": 9.642824155619095e-05,
|
| 2228 |
+
"loss": 0.0062,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.6291423484024137,
|
| 2233 |
+
"grad_norm": 0.06254574460432358,
|
| 2234 |
+
"learning_rate": 9.638539930181592e-05,
|
| 2235 |
+
"loss": 0.0063,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.6311207834602829,
|
| 2240 |
+
"grad_norm": 0.06672441168101174,
|
| 2241 |
+
"learning_rate": 9.634231126954296e-05,
|
| 2242 |
+
"loss": 0.0064,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.6330992185181521,
|
| 2247 |
+
"grad_norm": 0.05090452979591465,
|
| 2248 |
+
"learning_rate": 9.629897768767853e-05,
|
| 2249 |
+
"loss": 0.0062,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.6350776535760214,
|
| 2254 |
+
"grad_norm": 0.05765847904822037,
|
| 2255 |
+
"learning_rate": 9.625539878583016e-05,
|
| 2256 |
+
"loss": 0.0063,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.6370560886338906,
|
| 2261 |
+
"grad_norm": 0.05041445650810007,
|
| 2262 |
+
"learning_rate": 9.621157479490521e-05,
|
| 2263 |
+
"loss": 0.0063,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.6390345236917598,
|
| 2268 |
+
"grad_norm": 0.04947260615273494,
|
| 2269 |
+
"learning_rate": 9.616750594710973e-05,
|
| 2270 |
+
"loss": 0.0064,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.641012958749629,
|
| 2275 |
+
"grad_norm": 0.052930154093284505,
|
| 2276 |
+
"learning_rate": 9.612319247594707e-05,
|
| 2277 |
+
"loss": 0.0062,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.6429913938074983,
|
| 2282 |
+
"grad_norm": 0.054125747626247174,
|
| 2283 |
+
"learning_rate": 9.607863461621685e-05,
|
| 2284 |
+
"loss": 0.0062,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.6449698288653675,
|
| 2289 |
+
"grad_norm": 0.05710421262162926,
|
| 2290 |
+
"learning_rate": 9.603383260401352e-05,
|
| 2291 |
+
"loss": 0.0063,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.6469482639232367,
|
| 2296 |
+
"grad_norm": 0.05398629342840845,
|
| 2297 |
+
"learning_rate": 9.598878667672526e-05,
|
| 2298 |
+
"loss": 0.0063,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.648926698981106,
|
| 2303 |
+
"grad_norm": 0.051946230834141204,
|
| 2304 |
+
"learning_rate": 9.594349707303262e-05,
|
| 2305 |
+
"loss": 0.0062,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.6509051340389752,
|
| 2310 |
+
"grad_norm": 0.05510427993338031,
|
| 2311 |
+
"learning_rate": 9.589796403290732e-05,
|
| 2312 |
+
"loss": 0.0062,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.6528835690968444,
|
| 2317 |
+
"grad_norm": 0.05585402519049233,
|
| 2318 |
+
"learning_rate": 9.585218779761093e-05,
|
| 2319 |
+
"loss": 0.0062,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.6548620041547136,
|
| 2324 |
+
"grad_norm": 0.05801028715682649,
|
| 2325 |
+
"learning_rate": 9.580616860969364e-05,
|
| 2326 |
+
"loss": 0.0062,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.6568404392125828,
|
| 2331 |
+
"grad_norm": 0.053932109311021714,
|
| 2332 |
+
"learning_rate": 9.575990671299295e-05,
|
| 2333 |
+
"loss": 0.0062,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.658818874270452,
|
| 2338 |
+
"grad_norm": 0.05924132673349397,
|
| 2339 |
+
"learning_rate": 9.571340235263236e-05,
|
| 2340 |
+
"loss": 0.0061,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.6607973093283213,
|
| 2345 |
+
"grad_norm": 0.06107219838029144,
|
| 2346 |
+
"learning_rate": 9.566665577502008e-05,
|
| 2347 |
+
"loss": 0.0062,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.6627757443861905,
|
| 2352 |
+
"grad_norm": 0.055948574313692906,
|
| 2353 |
+
"learning_rate": 9.561966722784774e-05,
|
| 2354 |
+
"loss": 0.0063,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.6647541794440598,
|
| 2359 |
+
"grad_norm": 0.056699774353642714,
|
| 2360 |
+
"learning_rate": 9.557243696008912e-05,
|
| 2361 |
+
"loss": 0.0062,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.666732614501929,
|
| 2366 |
+
"grad_norm": 0.06647034643900732,
|
| 2367 |
+
"learning_rate": 9.552496522199872e-05,
|
| 2368 |
+
"loss": 0.0062,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.6687110495597982,
|
| 2373 |
+
"grad_norm": 0.062109360688086826,
|
| 2374 |
+
"learning_rate": 9.54772522651105e-05,
|
| 2375 |
+
"loss": 0.0062,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.6706894846176674,
|
| 2380 |
+
"grad_norm": 0.0443074645474432,
|
| 2381 |
+
"learning_rate": 9.54292983422366e-05,
|
| 2382 |
+
"loss": 0.0062,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.6726679196755366,
|
| 2387 |
+
"grad_norm": 0.7126545771408139,
|
| 2388 |
+
"learning_rate": 9.538110370746587e-05,
|
| 2389 |
+
"loss": 0.0135,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.6746463547334058,
|
| 2394 |
+
"grad_norm": 1.838740056177238,
|
| 2395 |
+
"learning_rate": 9.533266861616264e-05,
|
| 2396 |
+
"loss": 0.0157,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.676624789791275,
|
| 2401 |
+
"grad_norm": 2.313205467450454,
|
| 2402 |
+
"learning_rate": 9.528399332496533e-05,
|
| 2403 |
+
"loss": 0.0903,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.6786032248491444,
|
| 2408 |
+
"grad_norm": 0.19913027405735842,
|
| 2409 |
+
"learning_rate": 9.523507809178506e-05,
|
| 2410 |
+
"loss": 0.0211,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.6805816599070136,
|
| 2415 |
+
"grad_norm": 0.11775151842160198,
|
| 2416 |
+
"learning_rate": 9.518592317580432e-05,
|
| 2417 |
+
"loss": 0.0103,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.6825600949648828,
|
| 2422 |
+
"grad_norm": 0.07434810113025182,
|
| 2423 |
+
"learning_rate": 9.513652883747558e-05,
|
| 2424 |
+
"loss": 0.008,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.684538530022752,
|
| 2429 |
+
"grad_norm": 0.07557817167064859,
|
| 2430 |
+
"learning_rate": 9.508689533851991e-05,
|
| 2431 |
+
"loss": 0.0073,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.6865169650806212,
|
| 2436 |
+
"grad_norm": 0.06071379140394132,
|
| 2437 |
+
"learning_rate": 9.503702294192563e-05,
|
| 2438 |
+
"loss": 0.0068,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.6884954001384904,
|
| 2443 |
+
"grad_norm": 0.063060210146848,
|
| 2444 |
+
"learning_rate": 9.498691191194682e-05,
|
| 2445 |
+
"loss": 0.0067,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.6904738351963596,
|
| 2450 |
+
"grad_norm": 0.05653952131526833,
|
| 2451 |
+
"learning_rate": 9.493656251410204e-05,
|
| 2452 |
+
"loss": 0.0066,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.692452270254229,
|
| 2457 |
+
"grad_norm": 0.07485922990062185,
|
| 2458 |
+
"learning_rate": 9.488597501517287e-05,
|
| 2459 |
+
"loss": 0.0066,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.6944307053120982,
|
| 2464 |
+
"grad_norm": 0.05282320165364128,
|
| 2465 |
+
"learning_rate": 9.483514968320243e-05,
|
| 2466 |
+
"loss": 0.0064,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.6964091403699674,
|
| 2471 |
+
"grad_norm": 0.05066162976905024,
|
| 2472 |
+
"learning_rate": 9.478408678749412e-05,
|
| 2473 |
+
"loss": 0.0062,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.6983875754278366,
|
| 2478 |
+
"grad_norm": 0.06086114944598497,
|
| 2479 |
+
"learning_rate": 9.473278659861e-05,
|
| 2480 |
+
"loss": 0.0063,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.7003660104857058,
|
| 2485 |
+
"grad_norm": 0.0585517952529397,
|
| 2486 |
+
"learning_rate": 9.46812493883695e-05,
|
| 2487 |
+
"loss": 0.0062,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.702344445543575,
|
| 2492 |
+
"grad_norm": 0.05465672190075681,
|
| 2493 |
+
"learning_rate": 9.462947542984795e-05,
|
| 2494 |
+
"loss": 0.0061,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.7043228806014442,
|
| 2499 |
+
"grad_norm": 0.05460020875531365,
|
| 2500 |
+
"learning_rate": 9.457746499737508e-05,
|
| 2501 |
+
"loss": 0.0063,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.7063013156593135,
|
| 2506 |
+
"grad_norm": 0.05449513153115839,
|
| 2507 |
+
"learning_rate": 9.45252183665336e-05,
|
| 2508 |
+
"loss": 0.0061,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.7082797507171827,
|
| 2513 |
+
"grad_norm": 0.05912074293349835,
|
| 2514 |
+
"learning_rate": 9.44727358141578e-05,
|
| 2515 |
+
"loss": 0.0063,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.710258185775052,
|
| 2520 |
+
"grad_norm": 0.05139678969575524,
|
| 2521 |
+
"learning_rate": 9.442001761833194e-05,
|
| 2522 |
+
"loss": 0.0062,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.7122366208329212,
|
| 2527 |
+
"grad_norm": 0.042382018800827065,
|
| 2528 |
+
"learning_rate": 9.436706405838892e-05,
|
| 2529 |
+
"loss": 0.006,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
}
|
| 2532 |
+
],
|
| 2533 |
+
"logging_steps": 10,
|
| 2534 |
+
"max_steps": 15165,
|
| 2535 |
+
"num_input_tokens_seen": 0,
|
| 2536 |
+
"num_train_epochs": 3,
|
| 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": 326036163133440.0,
|
| 2551 |
+
"train_batch_size": 128,
|
| 2552 |
+
"trial_name": null,
|
| 2553 |
+
"trial_params": null
|
| 2554 |
+
}
|