tmnam20 commited on
Commit
bdd0e88
1 Parent(s): a2886b1

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +4155 -0
trainer_state.json ADDED
@@ -0,0 +1,4155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 10000,
6
+ "global_step": 34113,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 1.997068566235746e-05,
14
+ "loss": 0.6565,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 1.994137132471492e-05,
20
+ "loss": 0.6664,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 1.9912056987072378e-05,
26
+ "loss": 0.6734,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.02,
31
+ "learning_rate": 1.9882742649429837e-05,
32
+ "loss": 0.6685,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 1.9853428311787297e-05,
38
+ "loss": 0.6654,
39
+ "step": 250
40
+ },
41
+ {
42
+ "epoch": 0.03,
43
+ "learning_rate": 1.9824113974144757e-05,
44
+ "loss": 0.6722,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 1.9794799636502217e-05,
50
+ "loss": 0.6712,
51
+ "step": 350
52
+ },
53
+ {
54
+ "epoch": 0.04,
55
+ "learning_rate": 1.9765485298859673e-05,
56
+ "loss": 0.6575,
57
+ "step": 400
58
+ },
59
+ {
60
+ "epoch": 0.04,
61
+ "learning_rate": 1.9736170961217133e-05,
62
+ "loss": 0.6598,
63
+ "step": 450
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 1.9706856623574592e-05,
68
+ "loss": 0.6666,
69
+ "step": 500
70
+ },
71
+ {
72
+ "epoch": 0.05,
73
+ "learning_rate": 1.9677542285932052e-05,
74
+ "loss": 0.6681,
75
+ "step": 550
76
+ },
77
+ {
78
+ "epoch": 0.05,
79
+ "learning_rate": 1.9648227948289512e-05,
80
+ "loss": 0.6732,
81
+ "step": 600
82
+ },
83
+ {
84
+ "epoch": 0.06,
85
+ "learning_rate": 1.961891361064697e-05,
86
+ "loss": 0.6622,
87
+ "step": 650
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "learning_rate": 1.9589599273004428e-05,
92
+ "loss": 0.6641,
93
+ "step": 700
94
+ },
95
+ {
96
+ "epoch": 0.07,
97
+ "learning_rate": 1.9560284935361888e-05,
98
+ "loss": 0.6581,
99
+ "step": 750
100
+ },
101
+ {
102
+ "epoch": 0.07,
103
+ "learning_rate": 1.9530970597719344e-05,
104
+ "loss": 0.6655,
105
+ "step": 800
106
+ },
107
+ {
108
+ "epoch": 0.07,
109
+ "learning_rate": 1.9501656260076804e-05,
110
+ "loss": 0.6667,
111
+ "step": 850
112
+ },
113
+ {
114
+ "epoch": 0.08,
115
+ "learning_rate": 1.9472341922434264e-05,
116
+ "loss": 0.6598,
117
+ "step": 900
118
+ },
119
+ {
120
+ "epoch": 0.08,
121
+ "learning_rate": 1.9443027584791724e-05,
122
+ "loss": 0.6617,
123
+ "step": 950
124
+ },
125
+ {
126
+ "epoch": 0.09,
127
+ "learning_rate": 1.9413713247149183e-05,
128
+ "loss": 0.6552,
129
+ "step": 1000
130
+ },
131
+ {
132
+ "epoch": 0.09,
133
+ "learning_rate": 1.938439890950664e-05,
134
+ "loss": 0.6613,
135
+ "step": 1050
136
+ },
137
+ {
138
+ "epoch": 0.1,
139
+ "learning_rate": 1.93550845718641e-05,
140
+ "loss": 0.6742,
141
+ "step": 1100
142
+ },
143
+ {
144
+ "epoch": 0.1,
145
+ "learning_rate": 1.932577023422156e-05,
146
+ "loss": 0.665,
147
+ "step": 1150
148
+ },
149
+ {
150
+ "epoch": 0.11,
151
+ "learning_rate": 1.929645589657902e-05,
152
+ "loss": 0.6695,
153
+ "step": 1200
154
+ },
155
+ {
156
+ "epoch": 0.11,
157
+ "learning_rate": 1.926714155893648e-05,
158
+ "loss": 0.6686,
159
+ "step": 1250
160
+ },
161
+ {
162
+ "epoch": 0.11,
163
+ "learning_rate": 1.9237827221293935e-05,
164
+ "loss": 0.6554,
165
+ "step": 1300
166
+ },
167
+ {
168
+ "epoch": 0.12,
169
+ "learning_rate": 1.9208512883651395e-05,
170
+ "loss": 0.6538,
171
+ "step": 1350
172
+ },
173
+ {
174
+ "epoch": 0.12,
175
+ "learning_rate": 1.9179198546008855e-05,
176
+ "loss": 0.6735,
177
+ "step": 1400
178
+ },
179
+ {
180
+ "epoch": 0.13,
181
+ "learning_rate": 1.9149884208366314e-05,
182
+ "loss": 0.6655,
183
+ "step": 1450
184
+ },
185
+ {
186
+ "epoch": 0.13,
187
+ "learning_rate": 1.9120569870723774e-05,
188
+ "loss": 0.6482,
189
+ "step": 1500
190
+ },
191
+ {
192
+ "epoch": 0.14,
193
+ "learning_rate": 1.909125553308123e-05,
194
+ "loss": 0.649,
195
+ "step": 1550
196
+ },
197
+ {
198
+ "epoch": 0.14,
199
+ "learning_rate": 1.906194119543869e-05,
200
+ "loss": 0.6638,
201
+ "step": 1600
202
+ },
203
+ {
204
+ "epoch": 0.15,
205
+ "learning_rate": 1.903262685779615e-05,
206
+ "loss": 0.6603,
207
+ "step": 1650
208
+ },
209
+ {
210
+ "epoch": 0.15,
211
+ "learning_rate": 1.900331252015361e-05,
212
+ "loss": 0.6682,
213
+ "step": 1700
214
+ },
215
+ {
216
+ "epoch": 0.15,
217
+ "learning_rate": 1.897399818251107e-05,
218
+ "loss": 0.6692,
219
+ "step": 1750
220
+ },
221
+ {
222
+ "epoch": 0.16,
223
+ "learning_rate": 1.894468384486853e-05,
224
+ "loss": 0.6579,
225
+ "step": 1800
226
+ },
227
+ {
228
+ "epoch": 0.16,
229
+ "learning_rate": 1.8915369507225986e-05,
230
+ "loss": 0.6659,
231
+ "step": 1850
232
+ },
233
+ {
234
+ "epoch": 0.17,
235
+ "learning_rate": 1.8886055169583445e-05,
236
+ "loss": 0.6806,
237
+ "step": 1900
238
+ },
239
+ {
240
+ "epoch": 0.17,
241
+ "learning_rate": 1.8856740831940902e-05,
242
+ "loss": 0.666,
243
+ "step": 1950
244
+ },
245
+ {
246
+ "epoch": 0.18,
247
+ "learning_rate": 1.882742649429836e-05,
248
+ "loss": 0.6621,
249
+ "step": 2000
250
+ },
251
+ {
252
+ "epoch": 0.18,
253
+ "learning_rate": 1.879811215665582e-05,
254
+ "loss": 0.6603,
255
+ "step": 2050
256
+ },
257
+ {
258
+ "epoch": 0.18,
259
+ "learning_rate": 1.876879781901328e-05,
260
+ "loss": 0.6656,
261
+ "step": 2100
262
+ },
263
+ {
264
+ "epoch": 0.19,
265
+ "learning_rate": 1.873948348137074e-05,
266
+ "loss": 0.6615,
267
+ "step": 2150
268
+ },
269
+ {
270
+ "epoch": 0.19,
271
+ "learning_rate": 1.8710169143728197e-05,
272
+ "loss": 0.6671,
273
+ "step": 2200
274
+ },
275
+ {
276
+ "epoch": 0.2,
277
+ "learning_rate": 1.8680854806085657e-05,
278
+ "loss": 0.6575,
279
+ "step": 2250
280
+ },
281
+ {
282
+ "epoch": 0.2,
283
+ "learning_rate": 1.8651540468443117e-05,
284
+ "loss": 0.667,
285
+ "step": 2300
286
+ },
287
+ {
288
+ "epoch": 0.21,
289
+ "learning_rate": 1.8622226130800576e-05,
290
+ "loss": 0.6575,
291
+ "step": 2350
292
+ },
293
+ {
294
+ "epoch": 0.21,
295
+ "learning_rate": 1.8592911793158036e-05,
296
+ "loss": 0.6594,
297
+ "step": 2400
298
+ },
299
+ {
300
+ "epoch": 0.22,
301
+ "learning_rate": 1.8563597455515493e-05,
302
+ "loss": 0.6467,
303
+ "step": 2450
304
+ },
305
+ {
306
+ "epoch": 0.22,
307
+ "learning_rate": 1.8534283117872952e-05,
308
+ "loss": 0.6678,
309
+ "step": 2500
310
+ },
311
+ {
312
+ "epoch": 0.22,
313
+ "learning_rate": 1.8504968780230412e-05,
314
+ "loss": 0.6696,
315
+ "step": 2550
316
+ },
317
+ {
318
+ "epoch": 0.23,
319
+ "learning_rate": 1.8475654442587872e-05,
320
+ "loss": 0.6681,
321
+ "step": 2600
322
+ },
323
+ {
324
+ "epoch": 0.23,
325
+ "learning_rate": 1.844634010494533e-05,
326
+ "loss": 0.6723,
327
+ "step": 2650
328
+ },
329
+ {
330
+ "epoch": 0.24,
331
+ "learning_rate": 1.8417025767302788e-05,
332
+ "loss": 0.6635,
333
+ "step": 2700
334
+ },
335
+ {
336
+ "epoch": 0.24,
337
+ "learning_rate": 1.8387711429660248e-05,
338
+ "loss": 0.6624,
339
+ "step": 2750
340
+ },
341
+ {
342
+ "epoch": 0.25,
343
+ "learning_rate": 1.8358397092017708e-05,
344
+ "loss": 0.6668,
345
+ "step": 2800
346
+ },
347
+ {
348
+ "epoch": 0.25,
349
+ "learning_rate": 1.8329082754375167e-05,
350
+ "loss": 0.6621,
351
+ "step": 2850
352
+ },
353
+ {
354
+ "epoch": 0.26,
355
+ "learning_rate": 1.8299768416732627e-05,
356
+ "loss": 0.66,
357
+ "step": 2900
358
+ },
359
+ {
360
+ "epoch": 0.26,
361
+ "learning_rate": 1.8270454079090087e-05,
362
+ "loss": 0.6606,
363
+ "step": 2950
364
+ },
365
+ {
366
+ "epoch": 0.26,
367
+ "learning_rate": 1.8241139741447543e-05,
368
+ "loss": 0.6741,
369
+ "step": 3000
370
+ },
371
+ {
372
+ "epoch": 0.27,
373
+ "learning_rate": 1.8211825403805003e-05,
374
+ "loss": 0.6697,
375
+ "step": 3050
376
+ },
377
+ {
378
+ "epoch": 0.27,
379
+ "learning_rate": 1.8182511066162463e-05,
380
+ "loss": 0.6632,
381
+ "step": 3100
382
+ },
383
+ {
384
+ "epoch": 0.28,
385
+ "learning_rate": 1.815319672851992e-05,
386
+ "loss": 0.6694,
387
+ "step": 3150
388
+ },
389
+ {
390
+ "epoch": 0.28,
391
+ "learning_rate": 1.812388239087738e-05,
392
+ "loss": 0.6587,
393
+ "step": 3200
394
+ },
395
+ {
396
+ "epoch": 0.29,
397
+ "learning_rate": 1.809456805323484e-05,
398
+ "loss": 0.6608,
399
+ "step": 3250
400
+ },
401
+ {
402
+ "epoch": 0.29,
403
+ "learning_rate": 1.80652537155923e-05,
404
+ "loss": 0.6656,
405
+ "step": 3300
406
+ },
407
+ {
408
+ "epoch": 0.29,
409
+ "learning_rate": 1.8035939377949755e-05,
410
+ "loss": 0.6596,
411
+ "step": 3350
412
+ },
413
+ {
414
+ "epoch": 0.3,
415
+ "learning_rate": 1.8006625040307214e-05,
416
+ "loss": 0.6651,
417
+ "step": 3400
418
+ },
419
+ {
420
+ "epoch": 0.3,
421
+ "learning_rate": 1.7977310702664674e-05,
422
+ "loss": 0.6664,
423
+ "step": 3450
424
+ },
425
+ {
426
+ "epoch": 0.31,
427
+ "learning_rate": 1.7947996365022134e-05,
428
+ "loss": 0.6643,
429
+ "step": 3500
430
+ },
431
+ {
432
+ "epoch": 0.31,
433
+ "learning_rate": 1.7918682027379594e-05,
434
+ "loss": 0.6545,
435
+ "step": 3550
436
+ },
437
+ {
438
+ "epoch": 0.32,
439
+ "learning_rate": 1.788936768973705e-05,
440
+ "loss": 0.6672,
441
+ "step": 3600
442
+ },
443
+ {
444
+ "epoch": 0.32,
445
+ "learning_rate": 1.786005335209451e-05,
446
+ "loss": 0.6681,
447
+ "step": 3650
448
+ },
449
+ {
450
+ "epoch": 0.33,
451
+ "learning_rate": 1.783073901445197e-05,
452
+ "loss": 0.6452,
453
+ "step": 3700
454
+ },
455
+ {
456
+ "epoch": 0.33,
457
+ "learning_rate": 1.780142467680943e-05,
458
+ "loss": 0.6667,
459
+ "step": 3750
460
+ },
461
+ {
462
+ "epoch": 0.33,
463
+ "learning_rate": 1.777211033916689e-05,
464
+ "loss": 0.6571,
465
+ "step": 3800
466
+ },
467
+ {
468
+ "epoch": 0.34,
469
+ "learning_rate": 1.7742796001524346e-05,
470
+ "loss": 0.6647,
471
+ "step": 3850
472
+ },
473
+ {
474
+ "epoch": 0.34,
475
+ "learning_rate": 1.7713481663881805e-05,
476
+ "loss": 0.6718,
477
+ "step": 3900
478
+ },
479
+ {
480
+ "epoch": 0.35,
481
+ "learning_rate": 1.7684167326239265e-05,
482
+ "loss": 0.6684,
483
+ "step": 3950
484
+ },
485
+ {
486
+ "epoch": 0.35,
487
+ "learning_rate": 1.7654852988596725e-05,
488
+ "loss": 0.663,
489
+ "step": 4000
490
+ },
491
+ {
492
+ "epoch": 0.36,
493
+ "learning_rate": 1.7625538650954185e-05,
494
+ "loss": 0.6542,
495
+ "step": 4050
496
+ },
497
+ {
498
+ "epoch": 0.36,
499
+ "learning_rate": 1.7596224313311644e-05,
500
+ "loss": 0.6645,
501
+ "step": 4100
502
+ },
503
+ {
504
+ "epoch": 0.36,
505
+ "learning_rate": 1.75669099756691e-05,
506
+ "loss": 0.6728,
507
+ "step": 4150
508
+ },
509
+ {
510
+ "epoch": 0.37,
511
+ "learning_rate": 1.753759563802656e-05,
512
+ "loss": 0.6696,
513
+ "step": 4200
514
+ },
515
+ {
516
+ "epoch": 0.37,
517
+ "learning_rate": 1.750828130038402e-05,
518
+ "loss": 0.6729,
519
+ "step": 4250
520
+ },
521
+ {
522
+ "epoch": 0.38,
523
+ "learning_rate": 1.747896696274148e-05,
524
+ "loss": 0.6484,
525
+ "step": 4300
526
+ },
527
+ {
528
+ "epoch": 0.38,
529
+ "learning_rate": 1.744965262509894e-05,
530
+ "loss": 0.659,
531
+ "step": 4350
532
+ },
533
+ {
534
+ "epoch": 0.39,
535
+ "learning_rate": 1.7420338287456396e-05,
536
+ "loss": 0.6593,
537
+ "step": 4400
538
+ },
539
+ {
540
+ "epoch": 0.39,
541
+ "learning_rate": 1.7391023949813856e-05,
542
+ "loss": 0.6626,
543
+ "step": 4450
544
+ },
545
+ {
546
+ "epoch": 0.4,
547
+ "learning_rate": 1.7361709612171312e-05,
548
+ "loss": 0.6598,
549
+ "step": 4500
550
+ },
551
+ {
552
+ "epoch": 0.4,
553
+ "learning_rate": 1.7332395274528772e-05,
554
+ "loss": 0.671,
555
+ "step": 4550
556
+ },
557
+ {
558
+ "epoch": 0.4,
559
+ "learning_rate": 1.7303080936886232e-05,
560
+ "loss": 0.6554,
561
+ "step": 4600
562
+ },
563
+ {
564
+ "epoch": 0.41,
565
+ "learning_rate": 1.727376659924369e-05,
566
+ "loss": 0.6521,
567
+ "step": 4650
568
+ },
569
+ {
570
+ "epoch": 0.41,
571
+ "learning_rate": 1.724445226160115e-05,
572
+ "loss": 0.6645,
573
+ "step": 4700
574
+ },
575
+ {
576
+ "epoch": 0.42,
577
+ "learning_rate": 1.7215137923958608e-05,
578
+ "loss": 0.6576,
579
+ "step": 4750
580
+ },
581
+ {
582
+ "epoch": 0.42,
583
+ "learning_rate": 1.7185823586316067e-05,
584
+ "loss": 0.6465,
585
+ "step": 4800
586
+ },
587
+ {
588
+ "epoch": 0.43,
589
+ "learning_rate": 1.7156509248673527e-05,
590
+ "loss": 0.661,
591
+ "step": 4850
592
+ },
593
+ {
594
+ "epoch": 0.43,
595
+ "learning_rate": 1.7127194911030987e-05,
596
+ "loss": 0.6577,
597
+ "step": 4900
598
+ },
599
+ {
600
+ "epoch": 0.44,
601
+ "learning_rate": 1.7097880573388447e-05,
602
+ "loss": 0.6652,
603
+ "step": 4950
604
+ },
605
+ {
606
+ "epoch": 0.44,
607
+ "learning_rate": 1.7068566235745903e-05,
608
+ "loss": 0.6425,
609
+ "step": 5000
610
+ },
611
+ {
612
+ "epoch": 0.44,
613
+ "learning_rate": 1.7039251898103363e-05,
614
+ "loss": 0.6687,
615
+ "step": 5050
616
+ },
617
+ {
618
+ "epoch": 0.45,
619
+ "learning_rate": 1.7009937560460823e-05,
620
+ "loss": 0.6721,
621
+ "step": 5100
622
+ },
623
+ {
624
+ "epoch": 0.45,
625
+ "learning_rate": 1.6980623222818282e-05,
626
+ "loss": 0.6592,
627
+ "step": 5150
628
+ },
629
+ {
630
+ "epoch": 0.46,
631
+ "learning_rate": 1.6951308885175742e-05,
632
+ "loss": 0.6593,
633
+ "step": 5200
634
+ },
635
+ {
636
+ "epoch": 0.46,
637
+ "learning_rate": 1.6921994547533202e-05,
638
+ "loss": 0.6496,
639
+ "step": 5250
640
+ },
641
+ {
642
+ "epoch": 0.47,
643
+ "learning_rate": 1.6892680209890658e-05,
644
+ "loss": 0.6719,
645
+ "step": 5300
646
+ },
647
+ {
648
+ "epoch": 0.47,
649
+ "learning_rate": 1.6863365872248118e-05,
650
+ "loss": 0.6709,
651
+ "step": 5350
652
+ },
653
+ {
654
+ "epoch": 0.47,
655
+ "learning_rate": 1.6834051534605578e-05,
656
+ "loss": 0.6631,
657
+ "step": 5400
658
+ },
659
+ {
660
+ "epoch": 0.48,
661
+ "learning_rate": 1.6804737196963038e-05,
662
+ "loss": 0.6539,
663
+ "step": 5450
664
+ },
665
+ {
666
+ "epoch": 0.48,
667
+ "learning_rate": 1.6775422859320497e-05,
668
+ "loss": 0.6597,
669
+ "step": 5500
670
+ },
671
+ {
672
+ "epoch": 0.49,
673
+ "learning_rate": 1.6746108521677954e-05,
674
+ "loss": 0.6609,
675
+ "step": 5550
676
+ },
677
+ {
678
+ "epoch": 0.49,
679
+ "learning_rate": 1.6716794184035413e-05,
680
+ "loss": 0.6658,
681
+ "step": 5600
682
+ },
683
+ {
684
+ "epoch": 0.5,
685
+ "learning_rate": 1.668747984639287e-05,
686
+ "loss": 0.6678,
687
+ "step": 5650
688
+ },
689
+ {
690
+ "epoch": 0.5,
691
+ "learning_rate": 1.665816550875033e-05,
692
+ "loss": 0.6564,
693
+ "step": 5700
694
+ },
695
+ {
696
+ "epoch": 0.51,
697
+ "learning_rate": 1.662885117110779e-05,
698
+ "loss": 0.6589,
699
+ "step": 5750
700
+ },
701
+ {
702
+ "epoch": 0.51,
703
+ "learning_rate": 1.659953683346525e-05,
704
+ "loss": 0.6583,
705
+ "step": 5800
706
+ },
707
+ {
708
+ "epoch": 0.51,
709
+ "learning_rate": 1.657022249582271e-05,
710
+ "loss": 0.6675,
711
+ "step": 5850
712
+ },
713
+ {
714
+ "epoch": 0.52,
715
+ "learning_rate": 1.6540908158180165e-05,
716
+ "loss": 0.6667,
717
+ "step": 5900
718
+ },
719
+ {
720
+ "epoch": 0.52,
721
+ "learning_rate": 1.6511593820537625e-05,
722
+ "loss": 0.667,
723
+ "step": 5950
724
+ },
725
+ {
726
+ "epoch": 0.53,
727
+ "learning_rate": 1.6482279482895085e-05,
728
+ "loss": 0.6757,
729
+ "step": 6000
730
+ },
731
+ {
732
+ "epoch": 0.53,
733
+ "learning_rate": 1.6452965145252545e-05,
734
+ "loss": 0.6724,
735
+ "step": 6050
736
+ },
737
+ {
738
+ "epoch": 0.54,
739
+ "learning_rate": 1.6423650807610004e-05,
740
+ "loss": 0.6649,
741
+ "step": 6100
742
+ },
743
+ {
744
+ "epoch": 0.54,
745
+ "learning_rate": 1.639433646996746e-05,
746
+ "loss": 0.6588,
747
+ "step": 6150
748
+ },
749
+ {
750
+ "epoch": 0.55,
751
+ "learning_rate": 1.636502213232492e-05,
752
+ "loss": 0.665,
753
+ "step": 6200
754
+ },
755
+ {
756
+ "epoch": 0.55,
757
+ "learning_rate": 1.633570779468238e-05,
758
+ "loss": 0.6516,
759
+ "step": 6250
760
+ },
761
+ {
762
+ "epoch": 0.55,
763
+ "learning_rate": 1.630639345703984e-05,
764
+ "loss": 0.6655,
765
+ "step": 6300
766
+ },
767
+ {
768
+ "epoch": 0.56,
769
+ "learning_rate": 1.62770791193973e-05,
770
+ "loss": 0.653,
771
+ "step": 6350
772
+ },
773
+ {
774
+ "epoch": 0.56,
775
+ "learning_rate": 1.624776478175476e-05,
776
+ "loss": 0.6822,
777
+ "step": 6400
778
+ },
779
+ {
780
+ "epoch": 0.57,
781
+ "learning_rate": 1.6218450444112216e-05,
782
+ "loss": 0.6699,
783
+ "step": 6450
784
+ },
785
+ {
786
+ "epoch": 0.57,
787
+ "learning_rate": 1.6189136106469676e-05,
788
+ "loss": 0.6612,
789
+ "step": 6500
790
+ },
791
+ {
792
+ "epoch": 0.58,
793
+ "learning_rate": 1.6159821768827135e-05,
794
+ "loss": 0.6661,
795
+ "step": 6550
796
+ },
797
+ {
798
+ "epoch": 0.58,
799
+ "learning_rate": 1.6130507431184595e-05,
800
+ "loss": 0.6657,
801
+ "step": 6600
802
+ },
803
+ {
804
+ "epoch": 0.58,
805
+ "learning_rate": 1.6101193093542055e-05,
806
+ "loss": 0.6605,
807
+ "step": 6650
808
+ },
809
+ {
810
+ "epoch": 0.59,
811
+ "learning_rate": 1.607187875589951e-05,
812
+ "loss": 0.655,
813
+ "step": 6700
814
+ },
815
+ {
816
+ "epoch": 0.59,
817
+ "learning_rate": 1.604256441825697e-05,
818
+ "loss": 0.6557,
819
+ "step": 6750
820
+ },
821
+ {
822
+ "epoch": 0.6,
823
+ "learning_rate": 1.601325008061443e-05,
824
+ "loss": 0.6729,
825
+ "step": 6800
826
+ },
827
+ {
828
+ "epoch": 0.6,
829
+ "learning_rate": 1.5983935742971887e-05,
830
+ "loss": 0.6627,
831
+ "step": 6850
832
+ },
833
+ {
834
+ "epoch": 0.61,
835
+ "learning_rate": 1.5954621405329347e-05,
836
+ "loss": 0.6621,
837
+ "step": 6900
838
+ },
839
+ {
840
+ "epoch": 0.61,
841
+ "learning_rate": 1.5925307067686807e-05,
842
+ "loss": 0.659,
843
+ "step": 6950
844
+ },
845
+ {
846
+ "epoch": 0.62,
847
+ "learning_rate": 1.5895992730044266e-05,
848
+ "loss": 0.6553,
849
+ "step": 7000
850
+ },
851
+ {
852
+ "epoch": 0.62,
853
+ "learning_rate": 1.5866678392401723e-05,
854
+ "loss": 0.6664,
855
+ "step": 7050
856
+ },
857
+ {
858
+ "epoch": 0.62,
859
+ "learning_rate": 1.5837364054759183e-05,
860
+ "loss": 0.6701,
861
+ "step": 7100
862
+ },
863
+ {
864
+ "epoch": 0.63,
865
+ "learning_rate": 1.5808049717116642e-05,
866
+ "loss": 0.6698,
867
+ "step": 7150
868
+ },
869
+ {
870
+ "epoch": 0.63,
871
+ "learning_rate": 1.5778735379474102e-05,
872
+ "loss": 0.6463,
873
+ "step": 7200
874
+ },
875
+ {
876
+ "epoch": 0.64,
877
+ "learning_rate": 1.5749421041831562e-05,
878
+ "loss": 0.6728,
879
+ "step": 7250
880
+ },
881
+ {
882
+ "epoch": 0.64,
883
+ "learning_rate": 1.5720106704189018e-05,
884
+ "loss": 0.6783,
885
+ "step": 7300
886
+ },
887
+ {
888
+ "epoch": 0.65,
889
+ "learning_rate": 1.5690792366546478e-05,
890
+ "loss": 0.6692,
891
+ "step": 7350
892
+ },
893
+ {
894
+ "epoch": 0.65,
895
+ "learning_rate": 1.5661478028903938e-05,
896
+ "loss": 0.6626,
897
+ "step": 7400
898
+ },
899
+ {
900
+ "epoch": 0.66,
901
+ "learning_rate": 1.5632163691261397e-05,
902
+ "loss": 0.6696,
903
+ "step": 7450
904
+ },
905
+ {
906
+ "epoch": 0.66,
907
+ "learning_rate": 1.5602849353618857e-05,
908
+ "loss": 0.6465,
909
+ "step": 7500
910
+ },
911
+ {
912
+ "epoch": 0.66,
913
+ "learning_rate": 1.5573535015976317e-05,
914
+ "loss": 0.6698,
915
+ "step": 7550
916
+ },
917
+ {
918
+ "epoch": 0.67,
919
+ "learning_rate": 1.5544220678333773e-05,
920
+ "loss": 0.673,
921
+ "step": 7600
922
+ },
923
+ {
924
+ "epoch": 0.67,
925
+ "learning_rate": 1.5514906340691233e-05,
926
+ "loss": 0.6675,
927
+ "step": 7650
928
+ },
929
+ {
930
+ "epoch": 0.68,
931
+ "learning_rate": 1.5485592003048693e-05,
932
+ "loss": 0.6627,
933
+ "step": 7700
934
+ },
935
+ {
936
+ "epoch": 0.68,
937
+ "learning_rate": 1.5456277665406153e-05,
938
+ "loss": 0.6587,
939
+ "step": 7750
940
+ },
941
+ {
942
+ "epoch": 0.69,
943
+ "learning_rate": 1.5426963327763612e-05,
944
+ "loss": 0.6545,
945
+ "step": 7800
946
+ },
947
+ {
948
+ "epoch": 0.69,
949
+ "learning_rate": 1.539764899012107e-05,
950
+ "loss": 0.6624,
951
+ "step": 7850
952
+ },
953
+ {
954
+ "epoch": 0.69,
955
+ "learning_rate": 1.536833465247853e-05,
956
+ "loss": 0.6588,
957
+ "step": 7900
958
+ },
959
+ {
960
+ "epoch": 0.7,
961
+ "learning_rate": 1.5339020314835988e-05,
962
+ "loss": 0.6594,
963
+ "step": 7950
964
+ },
965
+ {
966
+ "epoch": 0.7,
967
+ "learning_rate": 1.5309705977193448e-05,
968
+ "loss": 0.6539,
969
+ "step": 8000
970
+ },
971
+ {
972
+ "epoch": 0.71,
973
+ "learning_rate": 1.5280391639550908e-05,
974
+ "loss": 0.6616,
975
+ "step": 8050
976
+ },
977
+ {
978
+ "epoch": 0.71,
979
+ "learning_rate": 1.5251077301908364e-05,
980
+ "loss": 0.6565,
981
+ "step": 8100
982
+ },
983
+ {
984
+ "epoch": 0.72,
985
+ "learning_rate": 1.5221762964265822e-05,
986
+ "loss": 0.6759,
987
+ "step": 8150
988
+ },
989
+ {
990
+ "epoch": 0.72,
991
+ "learning_rate": 1.5192448626623282e-05,
992
+ "loss": 0.6628,
993
+ "step": 8200
994
+ },
995
+ {
996
+ "epoch": 0.73,
997
+ "learning_rate": 1.516313428898074e-05,
998
+ "loss": 0.6671,
999
+ "step": 8250
1000
+ },
1001
+ {
1002
+ "epoch": 0.73,
1003
+ "learning_rate": 1.51338199513382e-05,
1004
+ "loss": 0.6536,
1005
+ "step": 8300
1006
+ },
1007
+ {
1008
+ "epoch": 0.73,
1009
+ "learning_rate": 1.510450561369566e-05,
1010
+ "loss": 0.664,
1011
+ "step": 8350
1012
+ },
1013
+ {
1014
+ "epoch": 0.74,
1015
+ "learning_rate": 1.5075191276053118e-05,
1016
+ "loss": 0.6635,
1017
+ "step": 8400
1018
+ },
1019
+ {
1020
+ "epoch": 0.74,
1021
+ "learning_rate": 1.5045876938410577e-05,
1022
+ "loss": 0.6456,
1023
+ "step": 8450
1024
+ },
1025
+ {
1026
+ "epoch": 0.75,
1027
+ "learning_rate": 1.5016562600768037e-05,
1028
+ "loss": 0.6613,
1029
+ "step": 8500
1030
+ },
1031
+ {
1032
+ "epoch": 0.75,
1033
+ "learning_rate": 1.4987248263125495e-05,
1034
+ "loss": 0.6608,
1035
+ "step": 8550
1036
+ },
1037
+ {
1038
+ "epoch": 0.76,
1039
+ "learning_rate": 1.4957933925482955e-05,
1040
+ "loss": 0.6624,
1041
+ "step": 8600
1042
+ },
1043
+ {
1044
+ "epoch": 0.76,
1045
+ "learning_rate": 1.4928619587840413e-05,
1046
+ "loss": 0.6669,
1047
+ "step": 8650
1048
+ },
1049
+ {
1050
+ "epoch": 0.77,
1051
+ "learning_rate": 1.4899305250197873e-05,
1052
+ "loss": 0.6596,
1053
+ "step": 8700
1054
+ },
1055
+ {
1056
+ "epoch": 0.77,
1057
+ "learning_rate": 1.4869990912555333e-05,
1058
+ "loss": 0.6627,
1059
+ "step": 8750
1060
+ },
1061
+ {
1062
+ "epoch": 0.77,
1063
+ "learning_rate": 1.484067657491279e-05,
1064
+ "loss": 0.6717,
1065
+ "step": 8800
1066
+ },
1067
+ {
1068
+ "epoch": 0.78,
1069
+ "learning_rate": 1.481136223727025e-05,
1070
+ "loss": 0.6602,
1071
+ "step": 8850
1072
+ },
1073
+ {
1074
+ "epoch": 0.78,
1075
+ "learning_rate": 1.478204789962771e-05,
1076
+ "loss": 0.6529,
1077
+ "step": 8900
1078
+ },
1079
+ {
1080
+ "epoch": 0.79,
1081
+ "learning_rate": 1.4752733561985168e-05,
1082
+ "loss": 0.6541,
1083
+ "step": 8950
1084
+ },
1085
+ {
1086
+ "epoch": 0.79,
1087
+ "learning_rate": 1.4723419224342628e-05,
1088
+ "loss": 0.664,
1089
+ "step": 9000
1090
+ },
1091
+ {
1092
+ "epoch": 0.8,
1093
+ "learning_rate": 1.4694104886700086e-05,
1094
+ "loss": 0.6528,
1095
+ "step": 9050
1096
+ },
1097
+ {
1098
+ "epoch": 0.8,
1099
+ "learning_rate": 1.4664790549057546e-05,
1100
+ "loss": 0.6644,
1101
+ "step": 9100
1102
+ },
1103
+ {
1104
+ "epoch": 0.8,
1105
+ "learning_rate": 1.4635476211415006e-05,
1106
+ "loss": 0.6732,
1107
+ "step": 9150
1108
+ },
1109
+ {
1110
+ "epoch": 0.81,
1111
+ "learning_rate": 1.4606161873772464e-05,
1112
+ "loss": 0.6744,
1113
+ "step": 9200
1114
+ },
1115
+ {
1116
+ "epoch": 0.81,
1117
+ "learning_rate": 1.4576847536129923e-05,
1118
+ "loss": 0.668,
1119
+ "step": 9250
1120
+ },
1121
+ {
1122
+ "epoch": 0.82,
1123
+ "learning_rate": 1.4547533198487383e-05,
1124
+ "loss": 0.6531,
1125
+ "step": 9300
1126
+ },
1127
+ {
1128
+ "epoch": 0.82,
1129
+ "learning_rate": 1.451821886084484e-05,
1130
+ "loss": 0.6587,
1131
+ "step": 9350
1132
+ },
1133
+ {
1134
+ "epoch": 0.83,
1135
+ "learning_rate": 1.4488904523202298e-05,
1136
+ "loss": 0.6643,
1137
+ "step": 9400
1138
+ },
1139
+ {
1140
+ "epoch": 0.83,
1141
+ "learning_rate": 1.4459590185559757e-05,
1142
+ "loss": 0.6519,
1143
+ "step": 9450
1144
+ },
1145
+ {
1146
+ "epoch": 0.84,
1147
+ "learning_rate": 1.4430275847917217e-05,
1148
+ "loss": 0.6626,
1149
+ "step": 9500
1150
+ },
1151
+ {
1152
+ "epoch": 0.84,
1153
+ "learning_rate": 1.4400961510274675e-05,
1154
+ "loss": 0.6578,
1155
+ "step": 9550
1156
+ },
1157
+ {
1158
+ "epoch": 0.84,
1159
+ "learning_rate": 1.4371647172632135e-05,
1160
+ "loss": 0.6553,
1161
+ "step": 9600
1162
+ },
1163
+ {
1164
+ "epoch": 0.85,
1165
+ "learning_rate": 1.4342332834989595e-05,
1166
+ "loss": 0.6521,
1167
+ "step": 9650
1168
+ },
1169
+ {
1170
+ "epoch": 0.85,
1171
+ "learning_rate": 1.4313018497347053e-05,
1172
+ "loss": 0.6591,
1173
+ "step": 9700
1174
+ },
1175
+ {
1176
+ "epoch": 0.86,
1177
+ "learning_rate": 1.4283704159704513e-05,
1178
+ "loss": 0.6471,
1179
+ "step": 9750
1180
+ },
1181
+ {
1182
+ "epoch": 0.86,
1183
+ "learning_rate": 1.425438982206197e-05,
1184
+ "loss": 0.6605,
1185
+ "step": 9800
1186
+ },
1187
+ {
1188
+ "epoch": 0.87,
1189
+ "learning_rate": 1.422507548441943e-05,
1190
+ "loss": 0.6623,
1191
+ "step": 9850
1192
+ },
1193
+ {
1194
+ "epoch": 0.87,
1195
+ "learning_rate": 1.419576114677689e-05,
1196
+ "loss": 0.6608,
1197
+ "step": 9900
1198
+ },
1199
+ {
1200
+ "epoch": 0.88,
1201
+ "learning_rate": 1.4166446809134348e-05,
1202
+ "loss": 0.6682,
1203
+ "step": 9950
1204
+ },
1205
+ {
1206
+ "epoch": 0.88,
1207
+ "learning_rate": 1.4137132471491808e-05,
1208
+ "loss": 0.6588,
1209
+ "step": 10000
1210
+ },
1211
+ {
1212
+ "epoch": 0.88,
1213
+ "eval_accuracy": 0.6318327974276527,
1214
+ "eval_combined_score": 0.3159163987138264,
1215
+ "eval_f1": 0.0,
1216
+ "eval_loss": 0.6581798195838928,
1217
+ "eval_runtime": 154.7649,
1218
+ "eval_samples_per_second": 261.235,
1219
+ "eval_steps_per_second": 16.328,
1220
+ "step": 10000
1221
+ },
1222
+ {
1223
+ "epoch": 0.88,
1224
+ "learning_rate": 1.4107818133849268e-05,
1225
+ "loss": 0.6584,
1226
+ "step": 10050
1227
+ },
1228
+ {
1229
+ "epoch": 0.89,
1230
+ "learning_rate": 1.4078503796206726e-05,
1231
+ "loss": 0.6695,
1232
+ "step": 10100
1233
+ },
1234
+ {
1235
+ "epoch": 0.89,
1236
+ "learning_rate": 1.4049189458564186e-05,
1237
+ "loss": 0.6528,
1238
+ "step": 10150
1239
+ },
1240
+ {
1241
+ "epoch": 0.9,
1242
+ "learning_rate": 1.4019875120921644e-05,
1243
+ "loss": 0.6585,
1244
+ "step": 10200
1245
+ },
1246
+ {
1247
+ "epoch": 0.9,
1248
+ "learning_rate": 1.3990560783279103e-05,
1249
+ "loss": 0.6625,
1250
+ "step": 10250
1251
+ },
1252
+ {
1253
+ "epoch": 0.91,
1254
+ "learning_rate": 1.3961246445636563e-05,
1255
+ "loss": 0.6705,
1256
+ "step": 10300
1257
+ },
1258
+ {
1259
+ "epoch": 0.91,
1260
+ "learning_rate": 1.3931932107994021e-05,
1261
+ "loss": 0.642,
1262
+ "step": 10350
1263
+ },
1264
+ {
1265
+ "epoch": 0.91,
1266
+ "learning_rate": 1.3902617770351481e-05,
1267
+ "loss": 0.6584,
1268
+ "step": 10400
1269
+ },
1270
+ {
1271
+ "epoch": 0.92,
1272
+ "learning_rate": 1.387330343270894e-05,
1273
+ "loss": 0.6375,
1274
+ "step": 10450
1275
+ },
1276
+ {
1277
+ "epoch": 0.92,
1278
+ "learning_rate": 1.3843989095066399e-05,
1279
+ "loss": 0.6638,
1280
+ "step": 10500
1281
+ },
1282
+ {
1283
+ "epoch": 0.93,
1284
+ "learning_rate": 1.3814674757423859e-05,
1285
+ "loss": 0.648,
1286
+ "step": 10550
1287
+ },
1288
+ {
1289
+ "epoch": 0.93,
1290
+ "learning_rate": 1.3785360419781315e-05,
1291
+ "loss": 0.6609,
1292
+ "step": 10600
1293
+ },
1294
+ {
1295
+ "epoch": 0.94,
1296
+ "learning_rate": 1.3756046082138775e-05,
1297
+ "loss": 0.6485,
1298
+ "step": 10650
1299
+ },
1300
+ {
1301
+ "epoch": 0.94,
1302
+ "learning_rate": 1.3726731744496233e-05,
1303
+ "loss": 0.664,
1304
+ "step": 10700
1305
+ },
1306
+ {
1307
+ "epoch": 0.95,
1308
+ "learning_rate": 1.3697417406853692e-05,
1309
+ "loss": 0.6593,
1310
+ "step": 10750
1311
+ },
1312
+ {
1313
+ "epoch": 0.95,
1314
+ "learning_rate": 1.3668103069211152e-05,
1315
+ "loss": 0.661,
1316
+ "step": 10800
1317
+ },
1318
+ {
1319
+ "epoch": 0.95,
1320
+ "learning_rate": 1.363878873156861e-05,
1321
+ "loss": 0.6565,
1322
+ "step": 10850
1323
+ },
1324
+ {
1325
+ "epoch": 0.96,
1326
+ "learning_rate": 1.360947439392607e-05,
1327
+ "loss": 0.6567,
1328
+ "step": 10900
1329
+ },
1330
+ {
1331
+ "epoch": 0.96,
1332
+ "learning_rate": 1.3580160056283528e-05,
1333
+ "loss": 0.655,
1334
+ "step": 10950
1335
+ },
1336
+ {
1337
+ "epoch": 0.97,
1338
+ "learning_rate": 1.3550845718640988e-05,
1339
+ "loss": 0.6593,
1340
+ "step": 11000
1341
+ },
1342
+ {
1343
+ "epoch": 0.97,
1344
+ "learning_rate": 1.3521531380998448e-05,
1345
+ "loss": 0.6629,
1346
+ "step": 11050
1347
+ },
1348
+ {
1349
+ "epoch": 0.98,
1350
+ "learning_rate": 1.3492217043355906e-05,
1351
+ "loss": 0.6647,
1352
+ "step": 11100
1353
+ },
1354
+ {
1355
+ "epoch": 0.98,
1356
+ "learning_rate": 1.3462902705713365e-05,
1357
+ "loss": 0.6592,
1358
+ "step": 11150
1359
+ },
1360
+ {
1361
+ "epoch": 0.98,
1362
+ "learning_rate": 1.3433588368070825e-05,
1363
+ "loss": 0.6646,
1364
+ "step": 11200
1365
+ },
1366
+ {
1367
+ "epoch": 0.99,
1368
+ "learning_rate": 1.3404274030428283e-05,
1369
+ "loss": 0.6648,
1370
+ "step": 11250
1371
+ },
1372
+ {
1373
+ "epoch": 0.99,
1374
+ "learning_rate": 1.3374959692785743e-05,
1375
+ "loss": 0.6466,
1376
+ "step": 11300
1377
+ },
1378
+ {
1379
+ "epoch": 1.0,
1380
+ "learning_rate": 1.3345645355143201e-05,
1381
+ "loss": 0.6496,
1382
+ "step": 11350
1383
+ },
1384
+ {
1385
+ "epoch": 1.0,
1386
+ "learning_rate": 1.3316331017500661e-05,
1387
+ "loss": 0.6659,
1388
+ "step": 11400
1389
+ },
1390
+ {
1391
+ "epoch": 1.01,
1392
+ "learning_rate": 1.328701667985812e-05,
1393
+ "loss": 0.6649,
1394
+ "step": 11450
1395
+ },
1396
+ {
1397
+ "epoch": 1.01,
1398
+ "learning_rate": 1.3257702342215579e-05,
1399
+ "loss": 0.6775,
1400
+ "step": 11500
1401
+ },
1402
+ {
1403
+ "epoch": 1.02,
1404
+ "learning_rate": 1.3228388004573038e-05,
1405
+ "loss": 0.6582,
1406
+ "step": 11550
1407
+ },
1408
+ {
1409
+ "epoch": 1.02,
1410
+ "learning_rate": 1.3199073666930498e-05,
1411
+ "loss": 0.6506,
1412
+ "step": 11600
1413
+ },
1414
+ {
1415
+ "epoch": 1.02,
1416
+ "learning_rate": 1.3169759329287956e-05,
1417
+ "loss": 0.6573,
1418
+ "step": 11650
1419
+ },
1420
+ {
1421
+ "epoch": 1.03,
1422
+ "learning_rate": 1.3140444991645416e-05,
1423
+ "loss": 0.655,
1424
+ "step": 11700
1425
+ },
1426
+ {
1427
+ "epoch": 1.03,
1428
+ "learning_rate": 1.3111130654002874e-05,
1429
+ "loss": 0.6535,
1430
+ "step": 11750
1431
+ },
1432
+ {
1433
+ "epoch": 1.04,
1434
+ "learning_rate": 1.3081816316360334e-05,
1435
+ "loss": 0.6593,
1436
+ "step": 11800
1437
+ },
1438
+ {
1439
+ "epoch": 1.04,
1440
+ "learning_rate": 1.305250197871779e-05,
1441
+ "loss": 0.6647,
1442
+ "step": 11850
1443
+ },
1444
+ {
1445
+ "epoch": 1.05,
1446
+ "learning_rate": 1.302318764107525e-05,
1447
+ "loss": 0.6683,
1448
+ "step": 11900
1449
+ },
1450
+ {
1451
+ "epoch": 1.05,
1452
+ "learning_rate": 1.299387330343271e-05,
1453
+ "loss": 0.6558,
1454
+ "step": 11950
1455
+ },
1456
+ {
1457
+ "epoch": 1.06,
1458
+ "learning_rate": 1.2964558965790168e-05,
1459
+ "loss": 0.6513,
1460
+ "step": 12000
1461
+ },
1462
+ {
1463
+ "epoch": 1.06,
1464
+ "learning_rate": 1.2935244628147628e-05,
1465
+ "loss": 0.66,
1466
+ "step": 12050
1467
+ },
1468
+ {
1469
+ "epoch": 1.06,
1470
+ "learning_rate": 1.2905930290505086e-05,
1471
+ "loss": 0.6613,
1472
+ "step": 12100
1473
+ },
1474
+ {
1475
+ "epoch": 1.07,
1476
+ "learning_rate": 1.2876615952862545e-05,
1477
+ "loss": 0.6647,
1478
+ "step": 12150
1479
+ },
1480
+ {
1481
+ "epoch": 1.07,
1482
+ "learning_rate": 1.2847301615220005e-05,
1483
+ "loss": 0.6489,
1484
+ "step": 12200
1485
+ },
1486
+ {
1487
+ "epoch": 1.08,
1488
+ "learning_rate": 1.2817987277577463e-05,
1489
+ "loss": 0.6533,
1490
+ "step": 12250
1491
+ },
1492
+ {
1493
+ "epoch": 1.08,
1494
+ "learning_rate": 1.2788672939934923e-05,
1495
+ "loss": 0.6707,
1496
+ "step": 12300
1497
+ },
1498
+ {
1499
+ "epoch": 1.09,
1500
+ "learning_rate": 1.2759358602292383e-05,
1501
+ "loss": 0.6565,
1502
+ "step": 12350
1503
+ },
1504
+ {
1505
+ "epoch": 1.09,
1506
+ "learning_rate": 1.2730044264649841e-05,
1507
+ "loss": 0.6638,
1508
+ "step": 12400
1509
+ },
1510
+ {
1511
+ "epoch": 1.09,
1512
+ "learning_rate": 1.27007299270073e-05,
1513
+ "loss": 0.6689,
1514
+ "step": 12450
1515
+ },
1516
+ {
1517
+ "epoch": 1.1,
1518
+ "learning_rate": 1.2671415589364759e-05,
1519
+ "loss": 0.6603,
1520
+ "step": 12500
1521
+ },
1522
+ {
1523
+ "epoch": 1.1,
1524
+ "learning_rate": 1.2642101251722218e-05,
1525
+ "loss": 0.6533,
1526
+ "step": 12550
1527
+ },
1528
+ {
1529
+ "epoch": 1.11,
1530
+ "learning_rate": 1.2612786914079678e-05,
1531
+ "loss": 0.6615,
1532
+ "step": 12600
1533
+ },
1534
+ {
1535
+ "epoch": 1.11,
1536
+ "learning_rate": 1.2583472576437136e-05,
1537
+ "loss": 0.6625,
1538
+ "step": 12650
1539
+ },
1540
+ {
1541
+ "epoch": 1.12,
1542
+ "learning_rate": 1.2554158238794596e-05,
1543
+ "loss": 0.6495,
1544
+ "step": 12700
1545
+ },
1546
+ {
1547
+ "epoch": 1.12,
1548
+ "learning_rate": 1.2524843901152056e-05,
1549
+ "loss": 0.6431,
1550
+ "step": 12750
1551
+ },
1552
+ {
1553
+ "epoch": 1.13,
1554
+ "learning_rate": 1.2495529563509514e-05,
1555
+ "loss": 0.6611,
1556
+ "step": 12800
1557
+ },
1558
+ {
1559
+ "epoch": 1.13,
1560
+ "learning_rate": 1.2466215225866974e-05,
1561
+ "loss": 0.6559,
1562
+ "step": 12850
1563
+ },
1564
+ {
1565
+ "epoch": 1.13,
1566
+ "learning_rate": 1.2436900888224432e-05,
1567
+ "loss": 0.6475,
1568
+ "step": 12900
1569
+ },
1570
+ {
1571
+ "epoch": 1.14,
1572
+ "learning_rate": 1.2407586550581891e-05,
1573
+ "loss": 0.6557,
1574
+ "step": 12950
1575
+ },
1576
+ {
1577
+ "epoch": 1.14,
1578
+ "learning_rate": 1.2378272212939351e-05,
1579
+ "loss": 0.6571,
1580
+ "step": 13000
1581
+ },
1582
+ {
1583
+ "epoch": 1.15,
1584
+ "learning_rate": 1.234895787529681e-05,
1585
+ "loss": 0.6547,
1586
+ "step": 13050
1587
+ },
1588
+ {
1589
+ "epoch": 1.15,
1590
+ "learning_rate": 1.2319643537654267e-05,
1591
+ "loss": 0.6577,
1592
+ "step": 13100
1593
+ },
1594
+ {
1595
+ "epoch": 1.16,
1596
+ "learning_rate": 1.2290329200011725e-05,
1597
+ "loss": 0.653,
1598
+ "step": 13150
1599
+ },
1600
+ {
1601
+ "epoch": 1.16,
1602
+ "learning_rate": 1.2261014862369185e-05,
1603
+ "loss": 0.6619,
1604
+ "step": 13200
1605
+ },
1606
+ {
1607
+ "epoch": 1.17,
1608
+ "learning_rate": 1.2231700524726643e-05,
1609
+ "loss": 0.6625,
1610
+ "step": 13250
1611
+ },
1612
+ {
1613
+ "epoch": 1.17,
1614
+ "learning_rate": 1.2202386187084103e-05,
1615
+ "loss": 0.6582,
1616
+ "step": 13300
1617
+ },
1618
+ {
1619
+ "epoch": 1.17,
1620
+ "learning_rate": 1.2173071849441563e-05,
1621
+ "loss": 0.6511,
1622
+ "step": 13350
1623
+ },
1624
+ {
1625
+ "epoch": 1.18,
1626
+ "learning_rate": 1.214375751179902e-05,
1627
+ "loss": 0.6612,
1628
+ "step": 13400
1629
+ },
1630
+ {
1631
+ "epoch": 1.18,
1632
+ "learning_rate": 1.211444317415648e-05,
1633
+ "loss": 0.6592,
1634
+ "step": 13450
1635
+ },
1636
+ {
1637
+ "epoch": 1.19,
1638
+ "learning_rate": 1.208512883651394e-05,
1639
+ "loss": 0.6526,
1640
+ "step": 13500
1641
+ },
1642
+ {
1643
+ "epoch": 1.19,
1644
+ "learning_rate": 1.2055814498871398e-05,
1645
+ "loss": 0.6591,
1646
+ "step": 13550
1647
+ },
1648
+ {
1649
+ "epoch": 1.2,
1650
+ "learning_rate": 1.2026500161228858e-05,
1651
+ "loss": 0.6709,
1652
+ "step": 13600
1653
+ },
1654
+ {
1655
+ "epoch": 1.2,
1656
+ "learning_rate": 1.1997185823586316e-05,
1657
+ "loss": 0.6564,
1658
+ "step": 13650
1659
+ },
1660
+ {
1661
+ "epoch": 1.2,
1662
+ "learning_rate": 1.1967871485943776e-05,
1663
+ "loss": 0.656,
1664
+ "step": 13700
1665
+ },
1666
+ {
1667
+ "epoch": 1.21,
1668
+ "learning_rate": 1.1938557148301236e-05,
1669
+ "loss": 0.668,
1670
+ "step": 13750
1671
+ },
1672
+ {
1673
+ "epoch": 1.21,
1674
+ "learning_rate": 1.1909242810658694e-05,
1675
+ "loss": 0.6636,
1676
+ "step": 13800
1677
+ },
1678
+ {
1679
+ "epoch": 1.22,
1680
+ "learning_rate": 1.1879928473016154e-05,
1681
+ "loss": 0.6634,
1682
+ "step": 13850
1683
+ },
1684
+ {
1685
+ "epoch": 1.22,
1686
+ "learning_rate": 1.1850614135373613e-05,
1687
+ "loss": 0.6562,
1688
+ "step": 13900
1689
+ },
1690
+ {
1691
+ "epoch": 1.23,
1692
+ "learning_rate": 1.1821299797731071e-05,
1693
+ "loss": 0.6518,
1694
+ "step": 13950
1695
+ },
1696
+ {
1697
+ "epoch": 1.23,
1698
+ "learning_rate": 1.1791985460088531e-05,
1699
+ "loss": 0.6706,
1700
+ "step": 14000
1701
+ },
1702
+ {
1703
+ "epoch": 1.24,
1704
+ "learning_rate": 1.176267112244599e-05,
1705
+ "loss": 0.6531,
1706
+ "step": 14050
1707
+ },
1708
+ {
1709
+ "epoch": 1.24,
1710
+ "learning_rate": 1.1733356784803449e-05,
1711
+ "loss": 0.6622,
1712
+ "step": 14100
1713
+ },
1714
+ {
1715
+ "epoch": 1.24,
1716
+ "learning_rate": 1.1704042447160909e-05,
1717
+ "loss": 0.6389,
1718
+ "step": 14150
1719
+ },
1720
+ {
1721
+ "epoch": 1.25,
1722
+ "learning_rate": 1.1674728109518367e-05,
1723
+ "loss": 0.6642,
1724
+ "step": 14200
1725
+ },
1726
+ {
1727
+ "epoch": 1.25,
1728
+ "learning_rate": 1.1645413771875827e-05,
1729
+ "loss": 0.6527,
1730
+ "step": 14250
1731
+ },
1732
+ {
1733
+ "epoch": 1.26,
1734
+ "learning_rate": 1.1616099434233283e-05,
1735
+ "loss": 0.6406,
1736
+ "step": 14300
1737
+ },
1738
+ {
1739
+ "epoch": 1.26,
1740
+ "learning_rate": 1.1586785096590743e-05,
1741
+ "loss": 0.6627,
1742
+ "step": 14350
1743
+ },
1744
+ {
1745
+ "epoch": 1.27,
1746
+ "learning_rate": 1.15574707589482e-05,
1747
+ "loss": 0.6546,
1748
+ "step": 14400
1749
+ },
1750
+ {
1751
+ "epoch": 1.27,
1752
+ "learning_rate": 1.152815642130566e-05,
1753
+ "loss": 0.6597,
1754
+ "step": 14450
1755
+ },
1756
+ {
1757
+ "epoch": 1.28,
1758
+ "learning_rate": 1.149884208366312e-05,
1759
+ "loss": 0.6525,
1760
+ "step": 14500
1761
+ },
1762
+ {
1763
+ "epoch": 1.28,
1764
+ "learning_rate": 1.1469527746020578e-05,
1765
+ "loss": 0.6565,
1766
+ "step": 14550
1767
+ },
1768
+ {
1769
+ "epoch": 1.28,
1770
+ "learning_rate": 1.1440213408378038e-05,
1771
+ "loss": 0.6645,
1772
+ "step": 14600
1773
+ },
1774
+ {
1775
+ "epoch": 1.29,
1776
+ "learning_rate": 1.1410899070735498e-05,
1777
+ "loss": 0.6632,
1778
+ "step": 14650
1779
+ },
1780
+ {
1781
+ "epoch": 1.29,
1782
+ "learning_rate": 1.1381584733092956e-05,
1783
+ "loss": 0.6624,
1784
+ "step": 14700
1785
+ },
1786
+ {
1787
+ "epoch": 1.3,
1788
+ "learning_rate": 1.1352270395450416e-05,
1789
+ "loss": 0.6661,
1790
+ "step": 14750
1791
+ },
1792
+ {
1793
+ "epoch": 1.3,
1794
+ "learning_rate": 1.1322956057807874e-05,
1795
+ "loss": 0.6545,
1796
+ "step": 14800
1797
+ },
1798
+ {
1799
+ "epoch": 1.31,
1800
+ "learning_rate": 1.1293641720165334e-05,
1801
+ "loss": 0.6601,
1802
+ "step": 14850
1803
+ },
1804
+ {
1805
+ "epoch": 1.31,
1806
+ "learning_rate": 1.1264327382522793e-05,
1807
+ "loss": 0.6514,
1808
+ "step": 14900
1809
+ },
1810
+ {
1811
+ "epoch": 1.31,
1812
+ "learning_rate": 1.1235013044880251e-05,
1813
+ "loss": 0.6669,
1814
+ "step": 14950
1815
+ },
1816
+ {
1817
+ "epoch": 1.32,
1818
+ "learning_rate": 1.1205698707237711e-05,
1819
+ "loss": 0.6476,
1820
+ "step": 15000
1821
+ },
1822
+ {
1823
+ "epoch": 1.32,
1824
+ "learning_rate": 1.1176384369595171e-05,
1825
+ "loss": 0.6591,
1826
+ "step": 15050
1827
+ },
1828
+ {
1829
+ "epoch": 1.33,
1830
+ "learning_rate": 1.1147070031952629e-05,
1831
+ "loss": 0.6662,
1832
+ "step": 15100
1833
+ },
1834
+ {
1835
+ "epoch": 1.33,
1836
+ "learning_rate": 1.1117755694310089e-05,
1837
+ "loss": 0.6478,
1838
+ "step": 15150
1839
+ },
1840
+ {
1841
+ "epoch": 1.34,
1842
+ "learning_rate": 1.1088441356667547e-05,
1843
+ "loss": 0.6467,
1844
+ "step": 15200
1845
+ },
1846
+ {
1847
+ "epoch": 1.34,
1848
+ "learning_rate": 1.1059127019025007e-05,
1849
+ "loss": 0.6491,
1850
+ "step": 15250
1851
+ },
1852
+ {
1853
+ "epoch": 1.35,
1854
+ "learning_rate": 1.1029812681382466e-05,
1855
+ "loss": 0.6664,
1856
+ "step": 15300
1857
+ },
1858
+ {
1859
+ "epoch": 1.35,
1860
+ "learning_rate": 1.1000498343739924e-05,
1861
+ "loss": 0.6544,
1862
+ "step": 15350
1863
+ },
1864
+ {
1865
+ "epoch": 1.35,
1866
+ "learning_rate": 1.0971184006097384e-05,
1867
+ "loss": 0.6565,
1868
+ "step": 15400
1869
+ },
1870
+ {
1871
+ "epoch": 1.36,
1872
+ "learning_rate": 1.0941869668454844e-05,
1873
+ "loss": 0.6679,
1874
+ "step": 15450
1875
+ },
1876
+ {
1877
+ "epoch": 1.36,
1878
+ "learning_rate": 1.0912555330812302e-05,
1879
+ "loss": 0.6584,
1880
+ "step": 15500
1881
+ },
1882
+ {
1883
+ "epoch": 1.37,
1884
+ "learning_rate": 1.0883240993169758e-05,
1885
+ "loss": 0.647,
1886
+ "step": 15550
1887
+ },
1888
+ {
1889
+ "epoch": 1.37,
1890
+ "learning_rate": 1.0853926655527218e-05,
1891
+ "loss": 0.6614,
1892
+ "step": 15600
1893
+ },
1894
+ {
1895
+ "epoch": 1.38,
1896
+ "learning_rate": 1.0824612317884678e-05,
1897
+ "loss": 0.6668,
1898
+ "step": 15650
1899
+ },
1900
+ {
1901
+ "epoch": 1.38,
1902
+ "learning_rate": 1.0795297980242136e-05,
1903
+ "loss": 0.6578,
1904
+ "step": 15700
1905
+ },
1906
+ {
1907
+ "epoch": 1.39,
1908
+ "learning_rate": 1.0765983642599596e-05,
1909
+ "loss": 0.662,
1910
+ "step": 15750
1911
+ },
1912
+ {
1913
+ "epoch": 1.39,
1914
+ "learning_rate": 1.0736669304957055e-05,
1915
+ "loss": 0.6527,
1916
+ "step": 15800
1917
+ },
1918
+ {
1919
+ "epoch": 1.39,
1920
+ "learning_rate": 1.0707354967314513e-05,
1921
+ "loss": 0.6361,
1922
+ "step": 15850
1923
+ },
1924
+ {
1925
+ "epoch": 1.4,
1926
+ "learning_rate": 1.0678040629671973e-05,
1927
+ "loss": 0.6421,
1928
+ "step": 15900
1929
+ },
1930
+ {
1931
+ "epoch": 1.4,
1932
+ "learning_rate": 1.0648726292029431e-05,
1933
+ "loss": 0.6567,
1934
+ "step": 15950
1935
+ },
1936
+ {
1937
+ "epoch": 1.41,
1938
+ "learning_rate": 1.0619411954386891e-05,
1939
+ "loss": 0.6486,
1940
+ "step": 16000
1941
+ },
1942
+ {
1943
+ "epoch": 1.41,
1944
+ "learning_rate": 1.059009761674435e-05,
1945
+ "loss": 0.6613,
1946
+ "step": 16050
1947
+ },
1948
+ {
1949
+ "epoch": 1.42,
1950
+ "learning_rate": 1.0560783279101809e-05,
1951
+ "loss": 0.6592,
1952
+ "step": 16100
1953
+ },
1954
+ {
1955
+ "epoch": 1.42,
1956
+ "learning_rate": 1.0531468941459269e-05,
1957
+ "loss": 0.643,
1958
+ "step": 16150
1959
+ },
1960
+ {
1961
+ "epoch": 1.42,
1962
+ "learning_rate": 1.0502154603816728e-05,
1963
+ "loss": 0.6584,
1964
+ "step": 16200
1965
+ },
1966
+ {
1967
+ "epoch": 1.43,
1968
+ "learning_rate": 1.0472840266174186e-05,
1969
+ "loss": 0.6517,
1970
+ "step": 16250
1971
+ },
1972
+ {
1973
+ "epoch": 1.43,
1974
+ "learning_rate": 1.0443525928531646e-05,
1975
+ "loss": 0.6584,
1976
+ "step": 16300
1977
+ },
1978
+ {
1979
+ "epoch": 1.44,
1980
+ "learning_rate": 1.0414211590889104e-05,
1981
+ "loss": 0.6457,
1982
+ "step": 16350
1983
+ },
1984
+ {
1985
+ "epoch": 1.44,
1986
+ "learning_rate": 1.0384897253246564e-05,
1987
+ "loss": 0.6484,
1988
+ "step": 16400
1989
+ },
1990
+ {
1991
+ "epoch": 1.45,
1992
+ "learning_rate": 1.0355582915604024e-05,
1993
+ "loss": 0.636,
1994
+ "step": 16450
1995
+ },
1996
+ {
1997
+ "epoch": 1.45,
1998
+ "learning_rate": 1.0326268577961482e-05,
1999
+ "loss": 0.6669,
2000
+ "step": 16500
2001
+ },
2002
+ {
2003
+ "epoch": 1.46,
2004
+ "learning_rate": 1.0296954240318942e-05,
2005
+ "loss": 0.6554,
2006
+ "step": 16550
2007
+ },
2008
+ {
2009
+ "epoch": 1.46,
2010
+ "learning_rate": 1.0267639902676401e-05,
2011
+ "loss": 0.6501,
2012
+ "step": 16600
2013
+ },
2014
+ {
2015
+ "epoch": 1.46,
2016
+ "learning_rate": 1.023832556503386e-05,
2017
+ "loss": 0.6582,
2018
+ "step": 16650
2019
+ },
2020
+ {
2021
+ "epoch": 1.47,
2022
+ "learning_rate": 1.020901122739132e-05,
2023
+ "loss": 0.6514,
2024
+ "step": 16700
2025
+ },
2026
+ {
2027
+ "epoch": 1.47,
2028
+ "learning_rate": 1.0179696889748777e-05,
2029
+ "loss": 0.6624,
2030
+ "step": 16750
2031
+ },
2032
+ {
2033
+ "epoch": 1.48,
2034
+ "learning_rate": 1.0150382552106235e-05,
2035
+ "loss": 0.6544,
2036
+ "step": 16800
2037
+ },
2038
+ {
2039
+ "epoch": 1.48,
2040
+ "learning_rate": 1.0121068214463693e-05,
2041
+ "loss": 0.6585,
2042
+ "step": 16850
2043
+ },
2044
+ {
2045
+ "epoch": 1.49,
2046
+ "learning_rate": 1.0091753876821153e-05,
2047
+ "loss": 0.6659,
2048
+ "step": 16900
2049
+ },
2050
+ {
2051
+ "epoch": 1.49,
2052
+ "learning_rate": 1.0062439539178613e-05,
2053
+ "loss": 0.6606,
2054
+ "step": 16950
2055
+ },
2056
+ {
2057
+ "epoch": 1.5,
2058
+ "learning_rate": 1.0033125201536071e-05,
2059
+ "loss": 0.6446,
2060
+ "step": 17000
2061
+ },
2062
+ {
2063
+ "epoch": 1.5,
2064
+ "learning_rate": 1.000381086389353e-05,
2065
+ "loss": 0.6634,
2066
+ "step": 17050
2067
+ },
2068
+ {
2069
+ "epoch": 1.5,
2070
+ "learning_rate": 9.974496526250989e-06,
2071
+ "loss": 0.6592,
2072
+ "step": 17100
2073
+ },
2074
+ {
2075
+ "epoch": 1.51,
2076
+ "learning_rate": 9.945182188608449e-06,
2077
+ "loss": 0.6473,
2078
+ "step": 17150
2079
+ },
2080
+ {
2081
+ "epoch": 1.51,
2082
+ "learning_rate": 9.915867850965908e-06,
2083
+ "loss": 0.6536,
2084
+ "step": 17200
2085
+ },
2086
+ {
2087
+ "epoch": 1.52,
2088
+ "learning_rate": 9.886553513323366e-06,
2089
+ "loss": 0.6589,
2090
+ "step": 17250
2091
+ },
2092
+ {
2093
+ "epoch": 1.52,
2094
+ "learning_rate": 9.857239175680826e-06,
2095
+ "loss": 0.6491,
2096
+ "step": 17300
2097
+ },
2098
+ {
2099
+ "epoch": 1.53,
2100
+ "learning_rate": 9.827924838038286e-06,
2101
+ "loss": 0.6588,
2102
+ "step": 17350
2103
+ },
2104
+ {
2105
+ "epoch": 1.53,
2106
+ "learning_rate": 9.798610500395744e-06,
2107
+ "loss": 0.6457,
2108
+ "step": 17400
2109
+ },
2110
+ {
2111
+ "epoch": 1.53,
2112
+ "learning_rate": 9.769296162753204e-06,
2113
+ "loss": 0.6514,
2114
+ "step": 17450
2115
+ },
2116
+ {
2117
+ "epoch": 1.54,
2118
+ "learning_rate": 9.739981825110662e-06,
2119
+ "loss": 0.6549,
2120
+ "step": 17500
2121
+ },
2122
+ {
2123
+ "epoch": 1.54,
2124
+ "learning_rate": 9.710667487468122e-06,
2125
+ "loss": 0.6493,
2126
+ "step": 17550
2127
+ },
2128
+ {
2129
+ "epoch": 1.55,
2130
+ "learning_rate": 9.681353149825581e-06,
2131
+ "loss": 0.6495,
2132
+ "step": 17600
2133
+ },
2134
+ {
2135
+ "epoch": 1.55,
2136
+ "learning_rate": 9.65203881218304e-06,
2137
+ "loss": 0.6509,
2138
+ "step": 17650
2139
+ },
2140
+ {
2141
+ "epoch": 1.56,
2142
+ "learning_rate": 9.6227244745405e-06,
2143
+ "loss": 0.6679,
2144
+ "step": 17700
2145
+ },
2146
+ {
2147
+ "epoch": 1.56,
2148
+ "learning_rate": 9.593410136897957e-06,
2149
+ "loss": 0.6598,
2150
+ "step": 17750
2151
+ },
2152
+ {
2153
+ "epoch": 1.57,
2154
+ "learning_rate": 9.564095799255417e-06,
2155
+ "loss": 0.6561,
2156
+ "step": 17800
2157
+ },
2158
+ {
2159
+ "epoch": 1.57,
2160
+ "learning_rate": 9.534781461612875e-06,
2161
+ "loss": 0.6435,
2162
+ "step": 17850
2163
+ },
2164
+ {
2165
+ "epoch": 1.57,
2166
+ "learning_rate": 9.505467123970335e-06,
2167
+ "loss": 0.6579,
2168
+ "step": 17900
2169
+ },
2170
+ {
2171
+ "epoch": 1.58,
2172
+ "learning_rate": 9.476152786327793e-06,
2173
+ "loss": 0.644,
2174
+ "step": 17950
2175
+ },
2176
+ {
2177
+ "epoch": 1.58,
2178
+ "learning_rate": 9.446838448685253e-06,
2179
+ "loss": 0.6605,
2180
+ "step": 18000
2181
+ },
2182
+ {
2183
+ "epoch": 1.59,
2184
+ "learning_rate": 9.417524111042712e-06,
2185
+ "loss": 0.6466,
2186
+ "step": 18050
2187
+ },
2188
+ {
2189
+ "epoch": 1.59,
2190
+ "learning_rate": 9.38820977340017e-06,
2191
+ "loss": 0.6609,
2192
+ "step": 18100
2193
+ },
2194
+ {
2195
+ "epoch": 1.6,
2196
+ "learning_rate": 9.35889543575763e-06,
2197
+ "loss": 0.6538,
2198
+ "step": 18150
2199
+ },
2200
+ {
2201
+ "epoch": 1.6,
2202
+ "learning_rate": 9.329581098115088e-06,
2203
+ "loss": 0.6377,
2204
+ "step": 18200
2205
+ },
2206
+ {
2207
+ "epoch": 1.6,
2208
+ "learning_rate": 9.300266760472548e-06,
2209
+ "loss": 0.6656,
2210
+ "step": 18250
2211
+ },
2212
+ {
2213
+ "epoch": 1.61,
2214
+ "learning_rate": 9.270952422830008e-06,
2215
+ "loss": 0.6541,
2216
+ "step": 18300
2217
+ },
2218
+ {
2219
+ "epoch": 1.61,
2220
+ "learning_rate": 9.241638085187466e-06,
2221
+ "loss": 0.6542,
2222
+ "step": 18350
2223
+ },
2224
+ {
2225
+ "epoch": 1.62,
2226
+ "learning_rate": 9.212323747544924e-06,
2227
+ "loss": 0.6551,
2228
+ "step": 18400
2229
+ },
2230
+ {
2231
+ "epoch": 1.62,
2232
+ "learning_rate": 9.183009409902384e-06,
2233
+ "loss": 0.6514,
2234
+ "step": 18450
2235
+ },
2236
+ {
2237
+ "epoch": 1.63,
2238
+ "learning_rate": 9.153695072259843e-06,
2239
+ "loss": 0.6468,
2240
+ "step": 18500
2241
+ },
2242
+ {
2243
+ "epoch": 1.63,
2244
+ "learning_rate": 9.124380734617302e-06,
2245
+ "loss": 0.6465,
2246
+ "step": 18550
2247
+ },
2248
+ {
2249
+ "epoch": 1.64,
2250
+ "learning_rate": 9.095066396974761e-06,
2251
+ "loss": 0.6431,
2252
+ "step": 18600
2253
+ },
2254
+ {
2255
+ "epoch": 1.64,
2256
+ "learning_rate": 9.06575205933222e-06,
2257
+ "loss": 0.6518,
2258
+ "step": 18650
2259
+ },
2260
+ {
2261
+ "epoch": 1.64,
2262
+ "learning_rate": 9.036437721689679e-06,
2263
+ "loss": 0.6615,
2264
+ "step": 18700
2265
+ },
2266
+ {
2267
+ "epoch": 1.65,
2268
+ "learning_rate": 9.007123384047139e-06,
2269
+ "loss": 0.6573,
2270
+ "step": 18750
2271
+ },
2272
+ {
2273
+ "epoch": 1.65,
2274
+ "learning_rate": 8.977809046404597e-06,
2275
+ "loss": 0.6639,
2276
+ "step": 18800
2277
+ },
2278
+ {
2279
+ "epoch": 1.66,
2280
+ "learning_rate": 8.948494708762057e-06,
2281
+ "loss": 0.6428,
2282
+ "step": 18850
2283
+ },
2284
+ {
2285
+ "epoch": 1.66,
2286
+ "learning_rate": 8.919180371119516e-06,
2287
+ "loss": 0.6528,
2288
+ "step": 18900
2289
+ },
2290
+ {
2291
+ "epoch": 1.67,
2292
+ "learning_rate": 8.889866033476975e-06,
2293
+ "loss": 0.6575,
2294
+ "step": 18950
2295
+ },
2296
+ {
2297
+ "epoch": 1.67,
2298
+ "learning_rate": 8.860551695834433e-06,
2299
+ "loss": 0.6603,
2300
+ "step": 19000
2301
+ },
2302
+ {
2303
+ "epoch": 1.68,
2304
+ "learning_rate": 8.831237358191892e-06,
2305
+ "loss": 0.6522,
2306
+ "step": 19050
2307
+ },
2308
+ {
2309
+ "epoch": 1.68,
2310
+ "learning_rate": 8.80192302054935e-06,
2311
+ "loss": 0.6492,
2312
+ "step": 19100
2313
+ },
2314
+ {
2315
+ "epoch": 1.68,
2316
+ "learning_rate": 8.77260868290681e-06,
2317
+ "loss": 0.6511,
2318
+ "step": 19150
2319
+ },
2320
+ {
2321
+ "epoch": 1.69,
2322
+ "learning_rate": 8.74329434526427e-06,
2323
+ "loss": 0.651,
2324
+ "step": 19200
2325
+ },
2326
+ {
2327
+ "epoch": 1.69,
2328
+ "learning_rate": 8.713980007621728e-06,
2329
+ "loss": 0.6479,
2330
+ "step": 19250
2331
+ },
2332
+ {
2333
+ "epoch": 1.7,
2334
+ "learning_rate": 8.684665669979188e-06,
2335
+ "loss": 0.6587,
2336
+ "step": 19300
2337
+ },
2338
+ {
2339
+ "epoch": 1.7,
2340
+ "learning_rate": 8.655351332336648e-06,
2341
+ "loss": 0.657,
2342
+ "step": 19350
2343
+ },
2344
+ {
2345
+ "epoch": 1.71,
2346
+ "learning_rate": 8.626036994694106e-06,
2347
+ "loss": 0.6549,
2348
+ "step": 19400
2349
+ },
2350
+ {
2351
+ "epoch": 1.71,
2352
+ "learning_rate": 8.596722657051565e-06,
2353
+ "loss": 0.6549,
2354
+ "step": 19450
2355
+ },
2356
+ {
2357
+ "epoch": 1.71,
2358
+ "learning_rate": 8.567408319409023e-06,
2359
+ "loss": 0.6523,
2360
+ "step": 19500
2361
+ },
2362
+ {
2363
+ "epoch": 1.72,
2364
+ "learning_rate": 8.538093981766483e-06,
2365
+ "loss": 0.6518,
2366
+ "step": 19550
2367
+ },
2368
+ {
2369
+ "epoch": 1.72,
2370
+ "learning_rate": 8.508779644123941e-06,
2371
+ "loss": 0.6416,
2372
+ "step": 19600
2373
+ },
2374
+ {
2375
+ "epoch": 1.73,
2376
+ "learning_rate": 8.479465306481401e-06,
2377
+ "loss": 0.6538,
2378
+ "step": 19650
2379
+ },
2380
+ {
2381
+ "epoch": 1.73,
2382
+ "learning_rate": 8.450150968838859e-06,
2383
+ "loss": 0.6588,
2384
+ "step": 19700
2385
+ },
2386
+ {
2387
+ "epoch": 1.74,
2388
+ "learning_rate": 8.420836631196319e-06,
2389
+ "loss": 0.6499,
2390
+ "step": 19750
2391
+ },
2392
+ {
2393
+ "epoch": 1.74,
2394
+ "learning_rate": 8.391522293553777e-06,
2395
+ "loss": 0.6438,
2396
+ "step": 19800
2397
+ },
2398
+ {
2399
+ "epoch": 1.75,
2400
+ "learning_rate": 8.362207955911237e-06,
2401
+ "loss": 0.653,
2402
+ "step": 19850
2403
+ },
2404
+ {
2405
+ "epoch": 1.75,
2406
+ "learning_rate": 8.332893618268696e-06,
2407
+ "loss": 0.6608,
2408
+ "step": 19900
2409
+ },
2410
+ {
2411
+ "epoch": 1.75,
2412
+ "learning_rate": 8.303579280626155e-06,
2413
+ "loss": 0.6552,
2414
+ "step": 19950
2415
+ },
2416
+ {
2417
+ "epoch": 1.76,
2418
+ "learning_rate": 8.274264942983614e-06,
2419
+ "loss": 0.6572,
2420
+ "step": 20000
2421
+ },
2422
+ {
2423
+ "epoch": 1.76,
2424
+ "eval_accuracy": 0.6318327974276527,
2425
+ "eval_combined_score": 0.3159163987138264,
2426
+ "eval_f1": 0.0,
2427
+ "eval_loss": 0.6582868695259094,
2428
+ "eval_runtime": 155.0376,
2429
+ "eval_samples_per_second": 260.776,
2430
+ "eval_steps_per_second": 16.299,
2431
+ "step": 20000
2432
+ },
2433
+ {
2434
+ "epoch": 1.76,
2435
+ "learning_rate": 8.244950605341074e-06,
2436
+ "loss": 0.6543,
2437
+ "step": 20050
2438
+ },
2439
+ {
2440
+ "epoch": 1.77,
2441
+ "learning_rate": 8.215636267698532e-06,
2442
+ "loss": 0.6574,
2443
+ "step": 20100
2444
+ },
2445
+ {
2446
+ "epoch": 1.77,
2447
+ "learning_rate": 8.186321930055992e-06,
2448
+ "loss": 0.6448,
2449
+ "step": 20150
2450
+ },
2451
+ {
2452
+ "epoch": 1.78,
2453
+ "learning_rate": 8.15700759241345e-06,
2454
+ "loss": 0.6612,
2455
+ "step": 20200
2456
+ },
2457
+ {
2458
+ "epoch": 1.78,
2459
+ "learning_rate": 8.127693254770908e-06,
2460
+ "loss": 0.6586,
2461
+ "step": 20250
2462
+ },
2463
+ {
2464
+ "epoch": 1.79,
2465
+ "learning_rate": 8.098378917128368e-06,
2466
+ "loss": 0.6421,
2467
+ "step": 20300
2468
+ },
2469
+ {
2470
+ "epoch": 1.79,
2471
+ "learning_rate": 8.069064579485828e-06,
2472
+ "loss": 0.6702,
2473
+ "step": 20350
2474
+ },
2475
+ {
2476
+ "epoch": 1.79,
2477
+ "learning_rate": 8.039750241843286e-06,
2478
+ "loss": 0.6408,
2479
+ "step": 20400
2480
+ },
2481
+ {
2482
+ "epoch": 1.8,
2483
+ "learning_rate": 8.010435904200745e-06,
2484
+ "loss": 0.6415,
2485
+ "step": 20450
2486
+ },
2487
+ {
2488
+ "epoch": 1.8,
2489
+ "learning_rate": 7.981121566558205e-06,
2490
+ "loss": 0.6601,
2491
+ "step": 20500
2492
+ },
2493
+ {
2494
+ "epoch": 1.81,
2495
+ "learning_rate": 7.951807228915663e-06,
2496
+ "loss": 0.6425,
2497
+ "step": 20550
2498
+ },
2499
+ {
2500
+ "epoch": 1.81,
2501
+ "learning_rate": 7.922492891273123e-06,
2502
+ "loss": 0.6564,
2503
+ "step": 20600
2504
+ },
2505
+ {
2506
+ "epoch": 1.82,
2507
+ "learning_rate": 7.893178553630581e-06,
2508
+ "loss": 0.6474,
2509
+ "step": 20650
2510
+ },
2511
+ {
2512
+ "epoch": 1.82,
2513
+ "learning_rate": 7.86386421598804e-06,
2514
+ "loss": 0.651,
2515
+ "step": 20700
2516
+ },
2517
+ {
2518
+ "epoch": 1.82,
2519
+ "learning_rate": 7.8345498783455e-06,
2520
+ "loss": 0.6417,
2521
+ "step": 20750
2522
+ },
2523
+ {
2524
+ "epoch": 1.83,
2525
+ "learning_rate": 7.805235540702959e-06,
2526
+ "loss": 0.655,
2527
+ "step": 20800
2528
+ },
2529
+ {
2530
+ "epoch": 1.83,
2531
+ "learning_rate": 7.775921203060417e-06,
2532
+ "loss": 0.6477,
2533
+ "step": 20850
2534
+ },
2535
+ {
2536
+ "epoch": 1.84,
2537
+ "learning_rate": 7.746606865417876e-06,
2538
+ "loss": 0.6385,
2539
+ "step": 20900
2540
+ },
2541
+ {
2542
+ "epoch": 1.84,
2543
+ "learning_rate": 7.717292527775334e-06,
2544
+ "loss": 0.6667,
2545
+ "step": 20950
2546
+ },
2547
+ {
2548
+ "epoch": 1.85,
2549
+ "learning_rate": 7.687978190132794e-06,
2550
+ "loss": 0.6496,
2551
+ "step": 21000
2552
+ },
2553
+ {
2554
+ "epoch": 1.85,
2555
+ "learning_rate": 7.658663852490254e-06,
2556
+ "loss": 0.6661,
2557
+ "step": 21050
2558
+ },
2559
+ {
2560
+ "epoch": 1.86,
2561
+ "learning_rate": 7.629349514847713e-06,
2562
+ "loss": 0.6536,
2563
+ "step": 21100
2564
+ },
2565
+ {
2566
+ "epoch": 1.86,
2567
+ "learning_rate": 7.600035177205172e-06,
2568
+ "loss": 0.6587,
2569
+ "step": 21150
2570
+ },
2571
+ {
2572
+ "epoch": 1.86,
2573
+ "learning_rate": 7.570720839562631e-06,
2574
+ "loss": 0.6531,
2575
+ "step": 21200
2576
+ },
2577
+ {
2578
+ "epoch": 1.87,
2579
+ "learning_rate": 7.54140650192009e-06,
2580
+ "loss": 0.6502,
2581
+ "step": 21250
2582
+ },
2583
+ {
2584
+ "epoch": 1.87,
2585
+ "learning_rate": 7.512092164277549e-06,
2586
+ "loss": 0.6511,
2587
+ "step": 21300
2588
+ },
2589
+ {
2590
+ "epoch": 1.88,
2591
+ "learning_rate": 7.482777826635008e-06,
2592
+ "loss": 0.6445,
2593
+ "step": 21350
2594
+ },
2595
+ {
2596
+ "epoch": 1.88,
2597
+ "learning_rate": 7.453463488992467e-06,
2598
+ "loss": 0.6538,
2599
+ "step": 21400
2600
+ },
2601
+ {
2602
+ "epoch": 1.89,
2603
+ "learning_rate": 7.424149151349925e-06,
2604
+ "loss": 0.6572,
2605
+ "step": 21450
2606
+ },
2607
+ {
2608
+ "epoch": 1.89,
2609
+ "learning_rate": 7.394834813707384e-06,
2610
+ "loss": 0.6518,
2611
+ "step": 21500
2612
+ },
2613
+ {
2614
+ "epoch": 1.9,
2615
+ "learning_rate": 7.365520476064844e-06,
2616
+ "loss": 0.6577,
2617
+ "step": 21550
2618
+ },
2619
+ {
2620
+ "epoch": 1.9,
2621
+ "learning_rate": 7.336206138422303e-06,
2622
+ "loss": 0.6383,
2623
+ "step": 21600
2624
+ },
2625
+ {
2626
+ "epoch": 1.9,
2627
+ "learning_rate": 7.306891800779762e-06,
2628
+ "loss": 0.6482,
2629
+ "step": 21650
2630
+ },
2631
+ {
2632
+ "epoch": 1.91,
2633
+ "learning_rate": 7.277577463137221e-06,
2634
+ "loss": 0.6465,
2635
+ "step": 21700
2636
+ },
2637
+ {
2638
+ "epoch": 1.91,
2639
+ "learning_rate": 7.24826312549468e-06,
2640
+ "loss": 0.6478,
2641
+ "step": 21750
2642
+ },
2643
+ {
2644
+ "epoch": 1.92,
2645
+ "learning_rate": 7.218948787852139e-06,
2646
+ "loss": 0.6452,
2647
+ "step": 21800
2648
+ },
2649
+ {
2650
+ "epoch": 1.92,
2651
+ "learning_rate": 7.189634450209598e-06,
2652
+ "loss": 0.6519,
2653
+ "step": 21850
2654
+ },
2655
+ {
2656
+ "epoch": 1.93,
2657
+ "learning_rate": 7.160320112567057e-06,
2658
+ "loss": 0.6557,
2659
+ "step": 21900
2660
+ },
2661
+ {
2662
+ "epoch": 1.93,
2663
+ "learning_rate": 7.131005774924516e-06,
2664
+ "loss": 0.6512,
2665
+ "step": 21950
2666
+ },
2667
+ {
2668
+ "epoch": 1.93,
2669
+ "learning_rate": 7.101691437281976e-06,
2670
+ "loss": 0.6478,
2671
+ "step": 22000
2672
+ },
2673
+ {
2674
+ "epoch": 1.94,
2675
+ "learning_rate": 7.072377099639435e-06,
2676
+ "loss": 0.6691,
2677
+ "step": 22050
2678
+ },
2679
+ {
2680
+ "epoch": 1.94,
2681
+ "learning_rate": 7.043062761996893e-06,
2682
+ "loss": 0.6534,
2683
+ "step": 22100
2684
+ },
2685
+ {
2686
+ "epoch": 1.95,
2687
+ "learning_rate": 7.013748424354352e-06,
2688
+ "loss": 0.6493,
2689
+ "step": 22150
2690
+ },
2691
+ {
2692
+ "epoch": 1.95,
2693
+ "learning_rate": 6.984434086711811e-06,
2694
+ "loss": 0.6391,
2695
+ "step": 22200
2696
+ },
2697
+ {
2698
+ "epoch": 1.96,
2699
+ "learning_rate": 6.9551197490692704e-06,
2700
+ "loss": 0.6521,
2701
+ "step": 22250
2702
+ },
2703
+ {
2704
+ "epoch": 1.96,
2705
+ "learning_rate": 6.925805411426729e-06,
2706
+ "loss": 0.6484,
2707
+ "step": 22300
2708
+ },
2709
+ {
2710
+ "epoch": 1.97,
2711
+ "learning_rate": 6.896491073784188e-06,
2712
+ "loss": 0.6426,
2713
+ "step": 22350
2714
+ },
2715
+ {
2716
+ "epoch": 1.97,
2717
+ "learning_rate": 6.867176736141647e-06,
2718
+ "loss": 0.6586,
2719
+ "step": 22400
2720
+ },
2721
+ {
2722
+ "epoch": 1.97,
2723
+ "learning_rate": 6.837862398499107e-06,
2724
+ "loss": 0.6575,
2725
+ "step": 22450
2726
+ },
2727
+ {
2728
+ "epoch": 1.98,
2729
+ "learning_rate": 6.808548060856566e-06,
2730
+ "loss": 0.6676,
2731
+ "step": 22500
2732
+ },
2733
+ {
2734
+ "epoch": 1.98,
2735
+ "learning_rate": 6.779233723214025e-06,
2736
+ "loss": 0.648,
2737
+ "step": 22550
2738
+ },
2739
+ {
2740
+ "epoch": 1.99,
2741
+ "learning_rate": 6.749919385571484e-06,
2742
+ "loss": 0.6555,
2743
+ "step": 22600
2744
+ },
2745
+ {
2746
+ "epoch": 1.99,
2747
+ "learning_rate": 6.7206050479289434e-06,
2748
+ "loss": 0.6567,
2749
+ "step": 22650
2750
+ },
2751
+ {
2752
+ "epoch": 2.0,
2753
+ "learning_rate": 6.6912907102864015e-06,
2754
+ "loss": 0.6614,
2755
+ "step": 22700
2756
+ },
2757
+ {
2758
+ "epoch": 2.0,
2759
+ "learning_rate": 6.66197637264386e-06,
2760
+ "loss": 0.6609,
2761
+ "step": 22750
2762
+ },
2763
+ {
2764
+ "epoch": 2.01,
2765
+ "learning_rate": 6.632662035001319e-06,
2766
+ "loss": 0.6582,
2767
+ "step": 22800
2768
+ },
2769
+ {
2770
+ "epoch": 2.01,
2771
+ "learning_rate": 6.603347697358778e-06,
2772
+ "loss": 0.6648,
2773
+ "step": 22850
2774
+ },
2775
+ {
2776
+ "epoch": 2.01,
2777
+ "learning_rate": 6.574033359716237e-06,
2778
+ "loss": 0.6409,
2779
+ "step": 22900
2780
+ },
2781
+ {
2782
+ "epoch": 2.02,
2783
+ "learning_rate": 6.544719022073697e-06,
2784
+ "loss": 0.6501,
2785
+ "step": 22950
2786
+ },
2787
+ {
2788
+ "epoch": 2.02,
2789
+ "learning_rate": 6.515404684431156e-06,
2790
+ "loss": 0.6592,
2791
+ "step": 23000
2792
+ },
2793
+ {
2794
+ "epoch": 2.03,
2795
+ "learning_rate": 6.486090346788615e-06,
2796
+ "loss": 0.6492,
2797
+ "step": 23050
2798
+ },
2799
+ {
2800
+ "epoch": 2.03,
2801
+ "learning_rate": 6.456776009146074e-06,
2802
+ "loss": 0.6503,
2803
+ "step": 23100
2804
+ },
2805
+ {
2806
+ "epoch": 2.04,
2807
+ "learning_rate": 6.427461671503533e-06,
2808
+ "loss": 0.6569,
2809
+ "step": 23150
2810
+ },
2811
+ {
2812
+ "epoch": 2.04,
2813
+ "learning_rate": 6.398147333860992e-06,
2814
+ "loss": 0.6379,
2815
+ "step": 23200
2816
+ },
2817
+ {
2818
+ "epoch": 2.04,
2819
+ "learning_rate": 6.368832996218451e-06,
2820
+ "loss": 0.6514,
2821
+ "step": 23250
2822
+ },
2823
+ {
2824
+ "epoch": 2.05,
2825
+ "learning_rate": 6.33951865857591e-06,
2826
+ "loss": 0.6492,
2827
+ "step": 23300
2828
+ },
2829
+ {
2830
+ "epoch": 2.05,
2831
+ "learning_rate": 6.310204320933368e-06,
2832
+ "loss": 0.6393,
2833
+ "step": 23350
2834
+ },
2835
+ {
2836
+ "epoch": 2.06,
2837
+ "learning_rate": 6.280889983290828e-06,
2838
+ "loss": 0.6628,
2839
+ "step": 23400
2840
+ },
2841
+ {
2842
+ "epoch": 2.06,
2843
+ "learning_rate": 6.251575645648287e-06,
2844
+ "loss": 0.6436,
2845
+ "step": 23450
2846
+ },
2847
+ {
2848
+ "epoch": 2.07,
2849
+ "learning_rate": 6.222261308005746e-06,
2850
+ "loss": 0.6562,
2851
+ "step": 23500
2852
+ },
2853
+ {
2854
+ "epoch": 2.07,
2855
+ "learning_rate": 6.192946970363205e-06,
2856
+ "loss": 0.6552,
2857
+ "step": 23550
2858
+ },
2859
+ {
2860
+ "epoch": 2.08,
2861
+ "learning_rate": 6.1636326327206645e-06,
2862
+ "loss": 0.656,
2863
+ "step": 23600
2864
+ },
2865
+ {
2866
+ "epoch": 2.08,
2867
+ "learning_rate": 6.134318295078123e-06,
2868
+ "loss": 0.6592,
2869
+ "step": 23650
2870
+ },
2871
+ {
2872
+ "epoch": 2.08,
2873
+ "learning_rate": 6.105003957435582e-06,
2874
+ "loss": 0.6626,
2875
+ "step": 23700
2876
+ },
2877
+ {
2878
+ "epoch": 2.09,
2879
+ "learning_rate": 6.075689619793041e-06,
2880
+ "loss": 0.6568,
2881
+ "step": 23750
2882
+ },
2883
+ {
2884
+ "epoch": 2.09,
2885
+ "learning_rate": 6.046375282150501e-06,
2886
+ "loss": 0.6427,
2887
+ "step": 23800
2888
+ },
2889
+ {
2890
+ "epoch": 2.1,
2891
+ "learning_rate": 6.01706094450796e-06,
2892
+ "loss": 0.6667,
2893
+ "step": 23850
2894
+ },
2895
+ {
2896
+ "epoch": 2.1,
2897
+ "learning_rate": 5.987746606865419e-06,
2898
+ "loss": 0.642,
2899
+ "step": 23900
2900
+ },
2901
+ {
2902
+ "epoch": 2.11,
2903
+ "learning_rate": 5.958432269222877e-06,
2904
+ "loss": 0.6536,
2905
+ "step": 23950
2906
+ },
2907
+ {
2908
+ "epoch": 2.11,
2909
+ "learning_rate": 5.929117931580336e-06,
2910
+ "loss": 0.6507,
2911
+ "step": 24000
2912
+ },
2913
+ {
2914
+ "epoch": 2.12,
2915
+ "learning_rate": 5.8998035939377955e-06,
2916
+ "loss": 0.6507,
2917
+ "step": 24050
2918
+ },
2919
+ {
2920
+ "epoch": 2.12,
2921
+ "learning_rate": 5.8704892562952545e-06,
2922
+ "loss": 0.6446,
2923
+ "step": 24100
2924
+ },
2925
+ {
2926
+ "epoch": 2.12,
2927
+ "learning_rate": 5.841174918652713e-06,
2928
+ "loss": 0.6528,
2929
+ "step": 24150
2930
+ },
2931
+ {
2932
+ "epoch": 2.13,
2933
+ "learning_rate": 5.811860581010172e-06,
2934
+ "loss": 0.6476,
2935
+ "step": 24200
2936
+ },
2937
+ {
2938
+ "epoch": 2.13,
2939
+ "learning_rate": 5.782546243367631e-06,
2940
+ "loss": 0.6496,
2941
+ "step": 24250
2942
+ },
2943
+ {
2944
+ "epoch": 2.14,
2945
+ "learning_rate": 5.753231905725091e-06,
2946
+ "loss": 0.6495,
2947
+ "step": 24300
2948
+ },
2949
+ {
2950
+ "epoch": 2.14,
2951
+ "learning_rate": 5.72391756808255e-06,
2952
+ "loss": 0.6474,
2953
+ "step": 24350
2954
+ },
2955
+ {
2956
+ "epoch": 2.15,
2957
+ "learning_rate": 5.694603230440009e-06,
2958
+ "loss": 0.6506,
2959
+ "step": 24400
2960
+ },
2961
+ {
2962
+ "epoch": 2.15,
2963
+ "learning_rate": 5.665288892797468e-06,
2964
+ "loss": 0.6491,
2965
+ "step": 24450
2966
+ },
2967
+ {
2968
+ "epoch": 2.15,
2969
+ "learning_rate": 5.6359745551549274e-06,
2970
+ "loss": 0.6568,
2971
+ "step": 24500
2972
+ },
2973
+ {
2974
+ "epoch": 2.16,
2975
+ "learning_rate": 5.606660217512386e-06,
2976
+ "loss": 0.6673,
2977
+ "step": 24550
2978
+ },
2979
+ {
2980
+ "epoch": 2.16,
2981
+ "learning_rate": 5.5773458798698444e-06,
2982
+ "loss": 0.6491,
2983
+ "step": 24600
2984
+ },
2985
+ {
2986
+ "epoch": 2.17,
2987
+ "learning_rate": 5.548031542227303e-06,
2988
+ "loss": 0.6384,
2989
+ "step": 24650
2990
+ },
2991
+ {
2992
+ "epoch": 2.17,
2993
+ "learning_rate": 5.518717204584762e-06,
2994
+ "loss": 0.6501,
2995
+ "step": 24700
2996
+ },
2997
+ {
2998
+ "epoch": 2.18,
2999
+ "learning_rate": 5.489402866942222e-06,
3000
+ "loss": 0.6571,
3001
+ "step": 24750
3002
+ },
3003
+ {
3004
+ "epoch": 2.18,
3005
+ "learning_rate": 5.460088529299681e-06,
3006
+ "loss": 0.6574,
3007
+ "step": 24800
3008
+ },
3009
+ {
3010
+ "epoch": 2.19,
3011
+ "learning_rate": 5.43077419165714e-06,
3012
+ "loss": 0.6487,
3013
+ "step": 24850
3014
+ },
3015
+ {
3016
+ "epoch": 2.19,
3017
+ "learning_rate": 5.401459854014599e-06,
3018
+ "loss": 0.6436,
3019
+ "step": 24900
3020
+ },
3021
+ {
3022
+ "epoch": 2.19,
3023
+ "learning_rate": 5.3721455163720585e-06,
3024
+ "loss": 0.6573,
3025
+ "step": 24950
3026
+ },
3027
+ {
3028
+ "epoch": 2.2,
3029
+ "learning_rate": 5.3428311787295174e-06,
3030
+ "loss": 0.6554,
3031
+ "step": 25000
3032
+ },
3033
+ {
3034
+ "epoch": 2.2,
3035
+ "learning_rate": 5.313516841086976e-06,
3036
+ "loss": 0.6528,
3037
+ "step": 25050
3038
+ },
3039
+ {
3040
+ "epoch": 2.21,
3041
+ "learning_rate": 5.284202503444435e-06,
3042
+ "loss": 0.6569,
3043
+ "step": 25100
3044
+ },
3045
+ {
3046
+ "epoch": 2.21,
3047
+ "learning_rate": 5.254888165801895e-06,
3048
+ "loss": 0.659,
3049
+ "step": 25150
3050
+ },
3051
+ {
3052
+ "epoch": 2.22,
3053
+ "learning_rate": 5.225573828159353e-06,
3054
+ "loss": 0.6512,
3055
+ "step": 25200
3056
+ },
3057
+ {
3058
+ "epoch": 2.22,
3059
+ "learning_rate": 5.196259490516812e-06,
3060
+ "loss": 0.6489,
3061
+ "step": 25250
3062
+ },
3063
+ {
3064
+ "epoch": 2.22,
3065
+ "learning_rate": 5.166945152874271e-06,
3066
+ "loss": 0.6534,
3067
+ "step": 25300
3068
+ },
3069
+ {
3070
+ "epoch": 2.23,
3071
+ "learning_rate": 5.13763081523173e-06,
3072
+ "loss": 0.6535,
3073
+ "step": 25350
3074
+ },
3075
+ {
3076
+ "epoch": 2.23,
3077
+ "learning_rate": 5.108316477589189e-06,
3078
+ "loss": 0.6458,
3079
+ "step": 25400
3080
+ },
3081
+ {
3082
+ "epoch": 2.24,
3083
+ "learning_rate": 5.0790021399466485e-06,
3084
+ "loss": 0.6484,
3085
+ "step": 25450
3086
+ },
3087
+ {
3088
+ "epoch": 2.24,
3089
+ "learning_rate": 5.049687802304107e-06,
3090
+ "loss": 0.6537,
3091
+ "step": 25500
3092
+ },
3093
+ {
3094
+ "epoch": 2.25,
3095
+ "learning_rate": 5.020373464661566e-06,
3096
+ "loss": 0.6486,
3097
+ "step": 25550
3098
+ },
3099
+ {
3100
+ "epoch": 2.25,
3101
+ "learning_rate": 4.991059127019025e-06,
3102
+ "loss": 0.6472,
3103
+ "step": 25600
3104
+ },
3105
+ {
3106
+ "epoch": 2.26,
3107
+ "learning_rate": 4.961744789376484e-06,
3108
+ "loss": 0.6374,
3109
+ "step": 25650
3110
+ },
3111
+ {
3112
+ "epoch": 2.26,
3113
+ "learning_rate": 4.932430451733943e-06,
3114
+ "loss": 0.6463,
3115
+ "step": 25700
3116
+ },
3117
+ {
3118
+ "epoch": 2.26,
3119
+ "learning_rate": 4.903116114091403e-06,
3120
+ "loss": 0.6695,
3121
+ "step": 25750
3122
+ },
3123
+ {
3124
+ "epoch": 2.27,
3125
+ "learning_rate": 4.873801776448862e-06,
3126
+ "loss": 0.6293,
3127
+ "step": 25800
3128
+ },
3129
+ {
3130
+ "epoch": 2.27,
3131
+ "learning_rate": 4.844487438806321e-06,
3132
+ "loss": 0.6488,
3133
+ "step": 25850
3134
+ },
3135
+ {
3136
+ "epoch": 2.28,
3137
+ "learning_rate": 4.8151731011637796e-06,
3138
+ "loss": 0.6463,
3139
+ "step": 25900
3140
+ },
3141
+ {
3142
+ "epoch": 2.28,
3143
+ "learning_rate": 4.785858763521239e-06,
3144
+ "loss": 0.663,
3145
+ "step": 25950
3146
+ },
3147
+ {
3148
+ "epoch": 2.29,
3149
+ "learning_rate": 4.756544425878697e-06,
3150
+ "loss": 0.6452,
3151
+ "step": 26000
3152
+ },
3153
+ {
3154
+ "epoch": 2.29,
3155
+ "learning_rate": 4.727230088236156e-06,
3156
+ "loss": 0.64,
3157
+ "step": 26050
3158
+ },
3159
+ {
3160
+ "epoch": 2.3,
3161
+ "learning_rate": 4.697915750593616e-06,
3162
+ "loss": 0.6511,
3163
+ "step": 26100
3164
+ },
3165
+ {
3166
+ "epoch": 2.3,
3167
+ "learning_rate": 4.668601412951075e-06,
3168
+ "loss": 0.6578,
3169
+ "step": 26150
3170
+ },
3171
+ {
3172
+ "epoch": 2.3,
3173
+ "learning_rate": 4.639287075308534e-06,
3174
+ "loss": 0.6417,
3175
+ "step": 26200
3176
+ },
3177
+ {
3178
+ "epoch": 2.31,
3179
+ "learning_rate": 4.609972737665993e-06,
3180
+ "loss": 0.6569,
3181
+ "step": 26250
3182
+ },
3183
+ {
3184
+ "epoch": 2.31,
3185
+ "learning_rate": 4.580658400023452e-06,
3186
+ "loss": 0.6441,
3187
+ "step": 26300
3188
+ },
3189
+ {
3190
+ "epoch": 2.32,
3191
+ "learning_rate": 4.551344062380911e-06,
3192
+ "loss": 0.6515,
3193
+ "step": 26350
3194
+ },
3195
+ {
3196
+ "epoch": 2.32,
3197
+ "learning_rate": 4.5220297247383695e-06,
3198
+ "loss": 0.6431,
3199
+ "step": 26400
3200
+ },
3201
+ {
3202
+ "epoch": 2.33,
3203
+ "learning_rate": 4.492715387095829e-06,
3204
+ "loss": 0.6634,
3205
+ "step": 26450
3206
+ },
3207
+ {
3208
+ "epoch": 2.33,
3209
+ "learning_rate": 4.463401049453288e-06,
3210
+ "loss": 0.6495,
3211
+ "step": 26500
3212
+ },
3213
+ {
3214
+ "epoch": 2.33,
3215
+ "learning_rate": 4.434086711810747e-06,
3216
+ "loss": 0.648,
3217
+ "step": 26550
3218
+ },
3219
+ {
3220
+ "epoch": 2.34,
3221
+ "learning_rate": 4.404772374168206e-06,
3222
+ "loss": 0.641,
3223
+ "step": 26600
3224
+ },
3225
+ {
3226
+ "epoch": 2.34,
3227
+ "learning_rate": 4.375458036525665e-06,
3228
+ "loss": 0.6597,
3229
+ "step": 26650
3230
+ },
3231
+ {
3232
+ "epoch": 2.35,
3233
+ "learning_rate": 4.346143698883124e-06,
3234
+ "loss": 0.6576,
3235
+ "step": 26700
3236
+ },
3237
+ {
3238
+ "epoch": 2.35,
3239
+ "learning_rate": 4.316829361240583e-06,
3240
+ "loss": 0.6558,
3241
+ "step": 26750
3242
+ },
3243
+ {
3244
+ "epoch": 2.36,
3245
+ "learning_rate": 4.2875150235980425e-06,
3246
+ "loss": 0.6388,
3247
+ "step": 26800
3248
+ },
3249
+ {
3250
+ "epoch": 2.36,
3251
+ "learning_rate": 4.2582006859555014e-06,
3252
+ "loss": 0.6517,
3253
+ "step": 26850
3254
+ },
3255
+ {
3256
+ "epoch": 2.37,
3257
+ "learning_rate": 4.22888634831296e-06,
3258
+ "loss": 0.6514,
3259
+ "step": 26900
3260
+ },
3261
+ {
3262
+ "epoch": 2.37,
3263
+ "learning_rate": 4.199572010670419e-06,
3264
+ "loss": 0.6525,
3265
+ "step": 26950
3266
+ },
3267
+ {
3268
+ "epoch": 2.37,
3269
+ "learning_rate": 4.170257673027878e-06,
3270
+ "loss": 0.6546,
3271
+ "step": 27000
3272
+ },
3273
+ {
3274
+ "epoch": 2.38,
3275
+ "learning_rate": 4.140943335385337e-06,
3276
+ "loss": 0.6523,
3277
+ "step": 27050
3278
+ },
3279
+ {
3280
+ "epoch": 2.38,
3281
+ "learning_rate": 4.111628997742797e-06,
3282
+ "loss": 0.655,
3283
+ "step": 27100
3284
+ },
3285
+ {
3286
+ "epoch": 2.39,
3287
+ "learning_rate": 4.082314660100256e-06,
3288
+ "loss": 0.6412,
3289
+ "step": 27150
3290
+ },
3291
+ {
3292
+ "epoch": 2.39,
3293
+ "learning_rate": 4.053000322457714e-06,
3294
+ "loss": 0.64,
3295
+ "step": 27200
3296
+ },
3297
+ {
3298
+ "epoch": 2.4,
3299
+ "learning_rate": 4.023685984815174e-06,
3300
+ "loss": 0.6408,
3301
+ "step": 27250
3302
+ },
3303
+ {
3304
+ "epoch": 2.4,
3305
+ "learning_rate": 3.9943716471726325e-06,
3306
+ "loss": 0.6375,
3307
+ "step": 27300
3308
+ },
3309
+ {
3310
+ "epoch": 2.41,
3311
+ "learning_rate": 3.965057309530091e-06,
3312
+ "loss": 0.6475,
3313
+ "step": 27350
3314
+ },
3315
+ {
3316
+ "epoch": 2.41,
3317
+ "learning_rate": 3.93574297188755e-06,
3318
+ "loss": 0.6514,
3319
+ "step": 27400
3320
+ },
3321
+ {
3322
+ "epoch": 2.41,
3323
+ "learning_rate": 3.90642863424501e-06,
3324
+ "loss": 0.6484,
3325
+ "step": 27450
3326
+ },
3327
+ {
3328
+ "epoch": 2.42,
3329
+ "learning_rate": 3.877114296602469e-06,
3330
+ "loss": 0.6548,
3331
+ "step": 27500
3332
+ },
3333
+ {
3334
+ "epoch": 2.42,
3335
+ "learning_rate": 3.847799958959927e-06,
3336
+ "loss": 0.6537,
3337
+ "step": 27550
3338
+ },
3339
+ {
3340
+ "epoch": 2.43,
3341
+ "learning_rate": 3.818485621317387e-06,
3342
+ "loss": 0.6362,
3343
+ "step": 27600
3344
+ },
3345
+ {
3346
+ "epoch": 2.43,
3347
+ "learning_rate": 3.7891712836748457e-06,
3348
+ "loss": 0.6425,
3349
+ "step": 27650
3350
+ },
3351
+ {
3352
+ "epoch": 2.44,
3353
+ "learning_rate": 3.7598569460323047e-06,
3354
+ "loss": 0.6609,
3355
+ "step": 27700
3356
+ },
3357
+ {
3358
+ "epoch": 2.44,
3359
+ "learning_rate": 3.730542608389764e-06,
3360
+ "loss": 0.6624,
3361
+ "step": 27750
3362
+ },
3363
+ {
3364
+ "epoch": 2.44,
3365
+ "learning_rate": 3.701228270747223e-06,
3366
+ "loss": 0.6638,
3367
+ "step": 27800
3368
+ },
3369
+ {
3370
+ "epoch": 2.45,
3371
+ "learning_rate": 3.6719139331046814e-06,
3372
+ "loss": 0.6628,
3373
+ "step": 27850
3374
+ },
3375
+ {
3376
+ "epoch": 2.45,
3377
+ "learning_rate": 3.6425995954621407e-06,
3378
+ "loss": 0.651,
3379
+ "step": 27900
3380
+ },
3381
+ {
3382
+ "epoch": 2.46,
3383
+ "learning_rate": 3.6132852578195996e-06,
3384
+ "loss": 0.6535,
3385
+ "step": 27950
3386
+ },
3387
+ {
3388
+ "epoch": 2.46,
3389
+ "learning_rate": 3.583970920177059e-06,
3390
+ "loss": 0.6632,
3391
+ "step": 28000
3392
+ },
3393
+ {
3394
+ "epoch": 2.47,
3395
+ "learning_rate": 3.554656582534518e-06,
3396
+ "loss": 0.6555,
3397
+ "step": 28050
3398
+ },
3399
+ {
3400
+ "epoch": 2.47,
3401
+ "learning_rate": 3.5253422448919772e-06,
3402
+ "loss": 0.6377,
3403
+ "step": 28100
3404
+ },
3405
+ {
3406
+ "epoch": 2.48,
3407
+ "learning_rate": 3.4960279072494357e-06,
3408
+ "loss": 0.6488,
3409
+ "step": 28150
3410
+ },
3411
+ {
3412
+ "epoch": 2.48,
3413
+ "learning_rate": 3.466713569606895e-06,
3414
+ "loss": 0.6524,
3415
+ "step": 28200
3416
+ },
3417
+ {
3418
+ "epoch": 2.48,
3419
+ "learning_rate": 3.437399231964354e-06,
3420
+ "loss": 0.6371,
3421
+ "step": 28250
3422
+ },
3423
+ {
3424
+ "epoch": 2.49,
3425
+ "learning_rate": 3.4080848943218133e-06,
3426
+ "loss": 0.6469,
3427
+ "step": 28300
3428
+ },
3429
+ {
3430
+ "epoch": 2.49,
3431
+ "learning_rate": 3.3787705566792722e-06,
3432
+ "loss": 0.654,
3433
+ "step": 28350
3434
+ },
3435
+ {
3436
+ "epoch": 2.5,
3437
+ "learning_rate": 3.3494562190367316e-06,
3438
+ "loss": 0.6456,
3439
+ "step": 28400
3440
+ },
3441
+ {
3442
+ "epoch": 2.5,
3443
+ "learning_rate": 3.32014188139419e-06,
3444
+ "loss": 0.6559,
3445
+ "step": 28450
3446
+ },
3447
+ {
3448
+ "epoch": 2.51,
3449
+ "learning_rate": 3.290827543751649e-06,
3450
+ "loss": 0.659,
3451
+ "step": 28500
3452
+ },
3453
+ {
3454
+ "epoch": 2.51,
3455
+ "learning_rate": 3.2615132061091083e-06,
3456
+ "loss": 0.6482,
3457
+ "step": 28550
3458
+ },
3459
+ {
3460
+ "epoch": 2.52,
3461
+ "learning_rate": 3.232198868466567e-06,
3462
+ "loss": 0.6506,
3463
+ "step": 28600
3464
+ },
3465
+ {
3466
+ "epoch": 2.52,
3467
+ "learning_rate": 3.2028845308240265e-06,
3468
+ "loss": 0.6486,
3469
+ "step": 28650
3470
+ },
3471
+ {
3472
+ "epoch": 2.52,
3473
+ "learning_rate": 3.1735701931814855e-06,
3474
+ "loss": 0.6547,
3475
+ "step": 28700
3476
+ },
3477
+ {
3478
+ "epoch": 2.53,
3479
+ "learning_rate": 3.144255855538944e-06,
3480
+ "loss": 0.6495,
3481
+ "step": 28750
3482
+ },
3483
+ {
3484
+ "epoch": 2.53,
3485
+ "learning_rate": 3.1149415178964033e-06,
3486
+ "loss": 0.6516,
3487
+ "step": 28800
3488
+ },
3489
+ {
3490
+ "epoch": 2.54,
3491
+ "learning_rate": 3.085627180253862e-06,
3492
+ "loss": 0.647,
3493
+ "step": 28850
3494
+ },
3495
+ {
3496
+ "epoch": 2.54,
3497
+ "learning_rate": 3.0563128426113215e-06,
3498
+ "loss": 0.662,
3499
+ "step": 28900
3500
+ },
3501
+ {
3502
+ "epoch": 2.55,
3503
+ "learning_rate": 3.0269985049687804e-06,
3504
+ "loss": 0.6513,
3505
+ "step": 28950
3506
+ },
3507
+ {
3508
+ "epoch": 2.55,
3509
+ "learning_rate": 2.9976841673262398e-06,
3510
+ "loss": 0.6439,
3511
+ "step": 29000
3512
+ },
3513
+ {
3514
+ "epoch": 2.55,
3515
+ "learning_rate": 2.9683698296836987e-06,
3516
+ "loss": 0.6529,
3517
+ "step": 29050
3518
+ },
3519
+ {
3520
+ "epoch": 2.56,
3521
+ "learning_rate": 2.939055492041157e-06,
3522
+ "loss": 0.6498,
3523
+ "step": 29100
3524
+ },
3525
+ {
3526
+ "epoch": 2.56,
3527
+ "learning_rate": 2.9097411543986165e-06,
3528
+ "loss": 0.6432,
3529
+ "step": 29150
3530
+ },
3531
+ {
3532
+ "epoch": 2.57,
3533
+ "learning_rate": 2.8804268167560754e-06,
3534
+ "loss": 0.6514,
3535
+ "step": 29200
3536
+ },
3537
+ {
3538
+ "epoch": 2.57,
3539
+ "learning_rate": 2.8511124791135348e-06,
3540
+ "loss": 0.6469,
3541
+ "step": 29250
3542
+ },
3543
+ {
3544
+ "epoch": 2.58,
3545
+ "learning_rate": 2.8217981414709937e-06,
3546
+ "loss": 0.6464,
3547
+ "step": 29300
3548
+ },
3549
+ {
3550
+ "epoch": 2.58,
3551
+ "learning_rate": 2.792483803828453e-06,
3552
+ "loss": 0.6373,
3553
+ "step": 29350
3554
+ },
3555
+ {
3556
+ "epoch": 2.59,
3557
+ "learning_rate": 2.7631694661859115e-06,
3558
+ "loss": 0.6384,
3559
+ "step": 29400
3560
+ },
3561
+ {
3562
+ "epoch": 2.59,
3563
+ "learning_rate": 2.733855128543371e-06,
3564
+ "loss": 0.6635,
3565
+ "step": 29450
3566
+ },
3567
+ {
3568
+ "epoch": 2.59,
3569
+ "learning_rate": 2.7045407909008298e-06,
3570
+ "loss": 0.6488,
3571
+ "step": 29500
3572
+ },
3573
+ {
3574
+ "epoch": 2.6,
3575
+ "learning_rate": 2.675226453258289e-06,
3576
+ "loss": 0.655,
3577
+ "step": 29550
3578
+ },
3579
+ {
3580
+ "epoch": 2.6,
3581
+ "learning_rate": 2.645912115615748e-06,
3582
+ "loss": 0.6482,
3583
+ "step": 29600
3584
+ },
3585
+ {
3586
+ "epoch": 2.61,
3587
+ "learning_rate": 2.616597777973207e-06,
3588
+ "loss": 0.6406,
3589
+ "step": 29650
3590
+ },
3591
+ {
3592
+ "epoch": 2.61,
3593
+ "learning_rate": 2.587283440330666e-06,
3594
+ "loss": 0.6474,
3595
+ "step": 29700
3596
+ },
3597
+ {
3598
+ "epoch": 2.62,
3599
+ "learning_rate": 2.5579691026881247e-06,
3600
+ "loss": 0.6339,
3601
+ "step": 29750
3602
+ },
3603
+ {
3604
+ "epoch": 2.62,
3605
+ "learning_rate": 2.528654765045584e-06,
3606
+ "loss": 0.66,
3607
+ "step": 29800
3608
+ },
3609
+ {
3610
+ "epoch": 2.63,
3611
+ "learning_rate": 2.499340427403043e-06,
3612
+ "loss": 0.6501,
3613
+ "step": 29850
3614
+ },
3615
+ {
3616
+ "epoch": 2.63,
3617
+ "learning_rate": 2.4700260897605023e-06,
3618
+ "loss": 0.6454,
3619
+ "step": 29900
3620
+ },
3621
+ {
3622
+ "epoch": 2.63,
3623
+ "learning_rate": 2.440711752117961e-06,
3624
+ "loss": 0.6556,
3625
+ "step": 29950
3626
+ },
3627
+ {
3628
+ "epoch": 2.64,
3629
+ "learning_rate": 2.41139741447542e-06,
3630
+ "loss": 0.6578,
3631
+ "step": 30000
3632
+ },
3633
+ {
3634
+ "epoch": 2.64,
3635
+ "eval_accuracy": 0.6318327974276527,
3636
+ "eval_combined_score": 0.3159163987138264,
3637
+ "eval_f1": 0.0,
3638
+ "eval_loss": 0.6771143078804016,
3639
+ "eval_runtime": 154.5998,
3640
+ "eval_samples_per_second": 261.514,
3641
+ "eval_steps_per_second": 16.345,
3642
+ "step": 30000
3643
+ },
3644
+ {
3645
+ "epoch": 2.64,
3646
+ "learning_rate": 2.382083076832879e-06,
3647
+ "loss": 0.6529,
3648
+ "step": 30050
3649
+ },
3650
+ {
3651
+ "epoch": 2.65,
3652
+ "learning_rate": 2.352768739190338e-06,
3653
+ "loss": 0.6414,
3654
+ "step": 30100
3655
+ },
3656
+ {
3657
+ "epoch": 2.65,
3658
+ "learning_rate": 2.3234544015477973e-06,
3659
+ "loss": 0.6505,
3660
+ "step": 30150
3661
+ },
3662
+ {
3663
+ "epoch": 2.66,
3664
+ "learning_rate": 2.2941400639052562e-06,
3665
+ "loss": 0.6564,
3666
+ "step": 30200
3667
+ },
3668
+ {
3669
+ "epoch": 2.66,
3670
+ "learning_rate": 2.264825726262715e-06,
3671
+ "loss": 0.641,
3672
+ "step": 30250
3673
+ },
3674
+ {
3675
+ "epoch": 2.66,
3676
+ "learning_rate": 2.2355113886201745e-06,
3677
+ "loss": 0.6493,
3678
+ "step": 30300
3679
+ },
3680
+ {
3681
+ "epoch": 2.67,
3682
+ "learning_rate": 2.2061970509776334e-06,
3683
+ "loss": 0.6501,
3684
+ "step": 30350
3685
+ },
3686
+ {
3687
+ "epoch": 2.67,
3688
+ "learning_rate": 2.1768827133350923e-06,
3689
+ "loss": 0.6528,
3690
+ "step": 30400
3691
+ },
3692
+ {
3693
+ "epoch": 2.68,
3694
+ "learning_rate": 2.1475683756925512e-06,
3695
+ "loss": 0.6602,
3696
+ "step": 30450
3697
+ },
3698
+ {
3699
+ "epoch": 2.68,
3700
+ "learning_rate": 2.1182540380500106e-06,
3701
+ "loss": 0.6385,
3702
+ "step": 30500
3703
+ },
3704
+ {
3705
+ "epoch": 2.69,
3706
+ "learning_rate": 2.0889397004074695e-06,
3707
+ "loss": 0.6488,
3708
+ "step": 30550
3709
+ },
3710
+ {
3711
+ "epoch": 2.69,
3712
+ "learning_rate": 2.0596253627649284e-06,
3713
+ "loss": 0.6565,
3714
+ "step": 30600
3715
+ },
3716
+ {
3717
+ "epoch": 2.7,
3718
+ "learning_rate": 2.0303110251223877e-06,
3719
+ "loss": 0.6587,
3720
+ "step": 30650
3721
+ },
3722
+ {
3723
+ "epoch": 2.7,
3724
+ "learning_rate": 2.0009966874798466e-06,
3725
+ "loss": 0.6438,
3726
+ "step": 30700
3727
+ },
3728
+ {
3729
+ "epoch": 2.7,
3730
+ "learning_rate": 1.9716823498373055e-06,
3731
+ "loss": 0.6497,
3732
+ "step": 30750
3733
+ },
3734
+ {
3735
+ "epoch": 2.71,
3736
+ "learning_rate": 1.9423680121947645e-06,
3737
+ "loss": 0.6606,
3738
+ "step": 30800
3739
+ },
3740
+ {
3741
+ "epoch": 2.71,
3742
+ "learning_rate": 1.9130536745522234e-06,
3743
+ "loss": 0.6403,
3744
+ "step": 30850
3745
+ },
3746
+ {
3747
+ "epoch": 2.72,
3748
+ "learning_rate": 1.8837393369096827e-06,
3749
+ "loss": 0.6597,
3750
+ "step": 30900
3751
+ },
3752
+ {
3753
+ "epoch": 2.72,
3754
+ "learning_rate": 1.8544249992671418e-06,
3755
+ "loss": 0.6514,
3756
+ "step": 30950
3757
+ },
3758
+ {
3759
+ "epoch": 2.73,
3760
+ "learning_rate": 1.825110661624601e-06,
3761
+ "loss": 0.6405,
3762
+ "step": 31000
3763
+ },
3764
+ {
3765
+ "epoch": 2.73,
3766
+ "learning_rate": 1.7957963239820597e-06,
3767
+ "loss": 0.6558,
3768
+ "step": 31050
3769
+ },
3770
+ {
3771
+ "epoch": 2.74,
3772
+ "learning_rate": 1.7664819863395188e-06,
3773
+ "loss": 0.6435,
3774
+ "step": 31100
3775
+ },
3776
+ {
3777
+ "epoch": 2.74,
3778
+ "learning_rate": 1.737167648696978e-06,
3779
+ "loss": 0.6437,
3780
+ "step": 31150
3781
+ },
3782
+ {
3783
+ "epoch": 2.74,
3784
+ "learning_rate": 1.7078533110544368e-06,
3785
+ "loss": 0.6556,
3786
+ "step": 31200
3787
+ },
3788
+ {
3789
+ "epoch": 2.75,
3790
+ "learning_rate": 1.678538973411896e-06,
3791
+ "loss": 0.642,
3792
+ "step": 31250
3793
+ },
3794
+ {
3795
+ "epoch": 2.75,
3796
+ "learning_rate": 1.649224635769355e-06,
3797
+ "loss": 0.652,
3798
+ "step": 31300
3799
+ },
3800
+ {
3801
+ "epoch": 2.76,
3802
+ "learning_rate": 1.619910298126814e-06,
3803
+ "loss": 0.6504,
3804
+ "step": 31350
3805
+ },
3806
+ {
3807
+ "epoch": 2.76,
3808
+ "learning_rate": 1.590595960484273e-06,
3809
+ "loss": 0.6408,
3810
+ "step": 31400
3811
+ },
3812
+ {
3813
+ "epoch": 2.77,
3814
+ "learning_rate": 1.561281622841732e-06,
3815
+ "loss": 0.6511,
3816
+ "step": 31450
3817
+ },
3818
+ {
3819
+ "epoch": 2.77,
3820
+ "learning_rate": 1.531967285199191e-06,
3821
+ "loss": 0.6518,
3822
+ "step": 31500
3823
+ },
3824
+ {
3825
+ "epoch": 2.77,
3826
+ "learning_rate": 1.50265294755665e-06,
3827
+ "loss": 0.6513,
3828
+ "step": 31550
3829
+ },
3830
+ {
3831
+ "epoch": 2.78,
3832
+ "learning_rate": 1.4733386099141092e-06,
3833
+ "loss": 0.6518,
3834
+ "step": 31600
3835
+ },
3836
+ {
3837
+ "epoch": 2.78,
3838
+ "learning_rate": 1.444024272271568e-06,
3839
+ "loss": 0.6392,
3840
+ "step": 31650
3841
+ },
3842
+ {
3843
+ "epoch": 2.79,
3844
+ "learning_rate": 1.4147099346290272e-06,
3845
+ "loss": 0.6515,
3846
+ "step": 31700
3847
+ },
3848
+ {
3849
+ "epoch": 2.79,
3850
+ "learning_rate": 1.3853955969864864e-06,
3851
+ "loss": 0.6379,
3852
+ "step": 31750
3853
+ },
3854
+ {
3855
+ "epoch": 2.8,
3856
+ "learning_rate": 1.356081259343945e-06,
3857
+ "loss": 0.6533,
3858
+ "step": 31800
3859
+ },
3860
+ {
3861
+ "epoch": 2.8,
3862
+ "learning_rate": 1.3267669217014042e-06,
3863
+ "loss": 0.6431,
3864
+ "step": 31850
3865
+ },
3866
+ {
3867
+ "epoch": 2.81,
3868
+ "learning_rate": 1.2974525840588633e-06,
3869
+ "loss": 0.6521,
3870
+ "step": 31900
3871
+ },
3872
+ {
3873
+ "epoch": 2.81,
3874
+ "learning_rate": 1.2681382464163222e-06,
3875
+ "loss": 0.6455,
3876
+ "step": 31950
3877
+ },
3878
+ {
3879
+ "epoch": 2.81,
3880
+ "learning_rate": 1.2388239087737813e-06,
3881
+ "loss": 0.6486,
3882
+ "step": 32000
3883
+ },
3884
+ {
3885
+ "epoch": 2.82,
3886
+ "learning_rate": 1.2095095711312405e-06,
3887
+ "loss": 0.639,
3888
+ "step": 32050
3889
+ },
3890
+ {
3891
+ "epoch": 2.82,
3892
+ "learning_rate": 1.1801952334886994e-06,
3893
+ "loss": 0.6445,
3894
+ "step": 32100
3895
+ },
3896
+ {
3897
+ "epoch": 2.83,
3898
+ "learning_rate": 1.1508808958461585e-06,
3899
+ "loss": 0.6465,
3900
+ "step": 32150
3901
+ },
3902
+ {
3903
+ "epoch": 2.83,
3904
+ "learning_rate": 1.1215665582036176e-06,
3905
+ "loss": 0.6556,
3906
+ "step": 32200
3907
+ },
3908
+ {
3909
+ "epoch": 2.84,
3910
+ "learning_rate": 1.0922522205610765e-06,
3911
+ "loss": 0.6417,
3912
+ "step": 32250
3913
+ },
3914
+ {
3915
+ "epoch": 2.84,
3916
+ "learning_rate": 1.0629378829185355e-06,
3917
+ "loss": 0.6525,
3918
+ "step": 32300
3919
+ },
3920
+ {
3921
+ "epoch": 2.84,
3922
+ "learning_rate": 1.0336235452759946e-06,
3923
+ "loss": 0.6523,
3924
+ "step": 32350
3925
+ },
3926
+ {
3927
+ "epoch": 2.85,
3928
+ "learning_rate": 1.0043092076334537e-06,
3929
+ "loss": 0.6509,
3930
+ "step": 32400
3931
+ },
3932
+ {
3933
+ "epoch": 2.85,
3934
+ "learning_rate": 9.749948699909126e-07,
3935
+ "loss": 0.6473,
3936
+ "step": 32450
3937
+ },
3938
+ {
3939
+ "epoch": 2.86,
3940
+ "learning_rate": 9.456805323483717e-07,
3941
+ "loss": 0.6571,
3942
+ "step": 32500
3943
+ },
3944
+ {
3945
+ "epoch": 2.86,
3946
+ "learning_rate": 9.163661947058307e-07,
3947
+ "loss": 0.6418,
3948
+ "step": 32550
3949
+ },
3950
+ {
3951
+ "epoch": 2.87,
3952
+ "learning_rate": 8.870518570632897e-07,
3953
+ "loss": 0.6282,
3954
+ "step": 32600
3955
+ },
3956
+ {
3957
+ "epoch": 2.87,
3958
+ "learning_rate": 8.577375194207488e-07,
3959
+ "loss": 0.6324,
3960
+ "step": 32650
3961
+ },
3962
+ {
3963
+ "epoch": 2.88,
3964
+ "learning_rate": 8.284231817782078e-07,
3965
+ "loss": 0.655,
3966
+ "step": 32700
3967
+ },
3968
+ {
3969
+ "epoch": 2.88,
3970
+ "learning_rate": 7.991088441356669e-07,
3971
+ "loss": 0.6515,
3972
+ "step": 32750
3973
+ },
3974
+ {
3975
+ "epoch": 2.88,
3976
+ "learning_rate": 7.697945064931259e-07,
3977
+ "loss": 0.6394,
3978
+ "step": 32800
3979
+ },
3980
+ {
3981
+ "epoch": 2.89,
3982
+ "learning_rate": 7.404801688505849e-07,
3983
+ "loss": 0.6503,
3984
+ "step": 32850
3985
+ },
3986
+ {
3987
+ "epoch": 2.89,
3988
+ "learning_rate": 7.11165831208044e-07,
3989
+ "loss": 0.6684,
3990
+ "step": 32900
3991
+ },
3992
+ {
3993
+ "epoch": 2.9,
3994
+ "learning_rate": 6.818514935655029e-07,
3995
+ "loss": 0.6305,
3996
+ "step": 32950
3997
+ },
3998
+ {
3999
+ "epoch": 2.9,
4000
+ "learning_rate": 6.525371559229619e-07,
4001
+ "loss": 0.6438,
4002
+ "step": 33000
4003
+ },
4004
+ {
4005
+ "epoch": 2.91,
4006
+ "learning_rate": 6.23222818280421e-07,
4007
+ "loss": 0.6553,
4008
+ "step": 33050
4009
+ },
4010
+ {
4011
+ "epoch": 2.91,
4012
+ "learning_rate": 5.939084806378801e-07,
4013
+ "loss": 0.6603,
4014
+ "step": 33100
4015
+ },
4016
+ {
4017
+ "epoch": 2.92,
4018
+ "learning_rate": 5.645941429953391e-07,
4019
+ "loss": 0.6361,
4020
+ "step": 33150
4021
+ },
4022
+ {
4023
+ "epoch": 2.92,
4024
+ "learning_rate": 5.352798053527981e-07,
4025
+ "loss": 0.6313,
4026
+ "step": 33200
4027
+ },
4028
+ {
4029
+ "epoch": 2.92,
4030
+ "learning_rate": 5.059654677102571e-07,
4031
+ "loss": 0.6525,
4032
+ "step": 33250
4033
+ },
4034
+ {
4035
+ "epoch": 2.93,
4036
+ "learning_rate": 4.7665113006771615e-07,
4037
+ "loss": 0.6403,
4038
+ "step": 33300
4039
+ },
4040
+ {
4041
+ "epoch": 2.93,
4042
+ "learning_rate": 4.473367924251752e-07,
4043
+ "loss": 0.6374,
4044
+ "step": 33350
4045
+ },
4046
+ {
4047
+ "epoch": 2.94,
4048
+ "learning_rate": 4.180224547826342e-07,
4049
+ "loss": 0.649,
4050
+ "step": 33400
4051
+ },
4052
+ {
4053
+ "epoch": 2.94,
4054
+ "learning_rate": 3.8870811714009326e-07,
4055
+ "loss": 0.6467,
4056
+ "step": 33450
4057
+ },
4058
+ {
4059
+ "epoch": 2.95,
4060
+ "learning_rate": 3.5939377949755233e-07,
4061
+ "loss": 0.66,
4062
+ "step": 33500
4063
+ },
4064
+ {
4065
+ "epoch": 2.95,
4066
+ "learning_rate": 3.300794418550113e-07,
4067
+ "loss": 0.659,
4068
+ "step": 33550
4069
+ },
4070
+ {
4071
+ "epoch": 2.95,
4072
+ "learning_rate": 3.007651042124703e-07,
4073
+ "loss": 0.6488,
4074
+ "step": 33600
4075
+ },
4076
+ {
4077
+ "epoch": 2.96,
4078
+ "learning_rate": 2.714507665699294e-07,
4079
+ "loss": 0.6368,
4080
+ "step": 33650
4081
+ },
4082
+ {
4083
+ "epoch": 2.96,
4084
+ "learning_rate": 2.421364289273884e-07,
4085
+ "loss": 0.6401,
4086
+ "step": 33700
4087
+ },
4088
+ {
4089
+ "epoch": 2.97,
4090
+ "learning_rate": 2.1282209128484745e-07,
4091
+ "loss": 0.645,
4092
+ "step": 33750
4093
+ },
4094
+ {
4095
+ "epoch": 2.97,
4096
+ "learning_rate": 1.8350775364230647e-07,
4097
+ "loss": 0.6528,
4098
+ "step": 33800
4099
+ },
4100
+ {
4101
+ "epoch": 2.98,
4102
+ "learning_rate": 1.5419341599976551e-07,
4103
+ "loss": 0.6588,
4104
+ "step": 33850
4105
+ },
4106
+ {
4107
+ "epoch": 2.98,
4108
+ "learning_rate": 1.2487907835722453e-07,
4109
+ "loss": 0.6447,
4110
+ "step": 33900
4111
+ },
4112
+ {
4113
+ "epoch": 2.99,
4114
+ "learning_rate": 9.556474071468355e-08,
4115
+ "loss": 0.6362,
4116
+ "step": 33950
4117
+ },
4118
+ {
4119
+ "epoch": 2.99,
4120
+ "learning_rate": 6.62504030721426e-08,
4121
+ "loss": 0.6499,
4122
+ "step": 34000
4123
+ },
4124
+ {
4125
+ "epoch": 2.99,
4126
+ "learning_rate": 3.693606542960162e-08,
4127
+ "loss": 0.65,
4128
+ "step": 34050
4129
+ },
4130
+ {
4131
+ "epoch": 3.0,
4132
+ "learning_rate": 7.621727787060652e-09,
4133
+ "loss": 0.659,
4134
+ "step": 34100
4135
+ },
4136
+ {
4137
+ "epoch": 3.0,
4138
+ "step": 34113,
4139
+ "total_flos": 1.6921743221185622e+17,
4140
+ "train_loss": 0.6555979044355706,
4141
+ "train_runtime": 14432.7407,
4142
+ "train_samples_per_second": 75.629,
4143
+ "train_steps_per_second": 2.364
4144
+ }
4145
+ ],
4146
+ "logging_steps": 50,
4147
+ "max_steps": 34113,
4148
+ "num_input_tokens_seen": 0,
4149
+ "num_train_epochs": 3,
4150
+ "save_steps": 10000,
4151
+ "total_flos": 1.6921743221185622e+17,
4152
+ "train_batch_size": 32,
4153
+ "trial_name": null,
4154
+ "trial_params": null
4155
+ }