Upload pretrain_output/qwen2.5-7b-instruct-bio/bio_all/save1epoch/checkpoint-1300/trainer_state.json with huggingface_hub
Browse files
pretrain_output/qwen2.5-7b-instruct-bio/bio_all/save1epoch/checkpoint-1300/trainer_state.json
ADDED
|
@@ -0,0 +1,2878 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1300,
|
| 3 |
+
"best_metric": 1.49000871,
|
| 4 |
+
"best_model_checkpoint": "/oss/wangyujia/BIO/pretrain_output/qwen2.5-7b-instruct-bio/bio_all/v1-20250608-152604/checkpoint-1300",
|
| 5 |
+
"epoch": 0.974878140232471,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1300,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0007499062617172854,
|
| 14 |
+
"grad_norm": 2.599110635105592,
|
| 15 |
+
"learning_rate": 7.462686567164179e-07,
|
| 16 |
+
"loss": 1.9313178062438965,
|
| 17 |
+
"memory(GiB)": 33.98,
|
| 18 |
+
"step": 1,
|
| 19 |
+
"token_acc": 0.4522775684089792,
|
| 20 |
+
"train_speed(iter/s)": 0.034623
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.0037495313085864268,
|
| 24 |
+
"grad_norm": 1.6476215857066085,
|
| 25 |
+
"learning_rate": 3.7313432835820893e-06,
|
| 26 |
+
"loss": 1.9246041774749756,
|
| 27 |
+
"memory(GiB)": 47.91,
|
| 28 |
+
"step": 5,
|
| 29 |
+
"token_acc": 0.5217140695642776,
|
| 30 |
+
"train_speed(iter/s)": 0.062333
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0074990626171728535,
|
| 34 |
+
"grad_norm": 3.931828202739245,
|
| 35 |
+
"learning_rate": 7.4626865671641785e-06,
|
| 36 |
+
"loss": 1.876804733276367,
|
| 37 |
+
"memory(GiB)": 47.91,
|
| 38 |
+
"step": 10,
|
| 39 |
+
"token_acc": 0.5210085682516842,
|
| 40 |
+
"train_speed(iter/s)": 0.069282
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.01124859392575928,
|
| 44 |
+
"grad_norm": 1.453673062114526,
|
| 45 |
+
"learning_rate": 1.119402985074627e-05,
|
| 46 |
+
"loss": 1.885706329345703,
|
| 47 |
+
"memory(GiB)": 47.91,
|
| 48 |
+
"step": 15,
|
| 49 |
+
"token_acc": 0.5256180791849419,
|
| 50 |
+
"train_speed(iter/s)": 0.07201
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.014998125234345707,
|
| 54 |
+
"grad_norm": 1.6299819743267712,
|
| 55 |
+
"learning_rate": 1.4925373134328357e-05,
|
| 56 |
+
"loss": 1.860133743286133,
|
| 57 |
+
"memory(GiB)": 47.91,
|
| 58 |
+
"step": 20,
|
| 59 |
+
"token_acc": 0.5312506455322284,
|
| 60 |
+
"train_speed(iter/s)": 0.073457
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.018747656542932135,
|
| 64 |
+
"grad_norm": 1.263967703308006,
|
| 65 |
+
"learning_rate": 1.865671641791045e-05,
|
| 66 |
+
"loss": 1.85156192779541,
|
| 67 |
+
"memory(GiB)": 47.91,
|
| 68 |
+
"step": 25,
|
| 69 |
+
"token_acc": 0.5241231585236131,
|
| 70 |
+
"train_speed(iter/s)": 0.074348
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.02249718785151856,
|
| 74 |
+
"grad_norm": 1.8821671870075913,
|
| 75 |
+
"learning_rate": 2.238805970149254e-05,
|
| 76 |
+
"loss": 1.830419921875,
|
| 77 |
+
"memory(GiB)": 47.91,
|
| 78 |
+
"step": 30,
|
| 79 |
+
"token_acc": 0.5093614741502065,
|
| 80 |
+
"train_speed(iter/s)": 0.074966
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.026246719160104987,
|
| 84 |
+
"grad_norm": 1.2331649551706934,
|
| 85 |
+
"learning_rate": 2.6119402985074626e-05,
|
| 86 |
+
"loss": 1.806067657470703,
|
| 87 |
+
"memory(GiB)": 47.91,
|
| 88 |
+
"step": 35,
|
| 89 |
+
"token_acc": 0.5335501408312203,
|
| 90 |
+
"train_speed(iter/s)": 0.075495
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.029996250468691414,
|
| 94 |
+
"grad_norm": 1.2813822954405418,
|
| 95 |
+
"learning_rate": 2.9850746268656714e-05,
|
| 96 |
+
"loss": 1.8119426727294923,
|
| 97 |
+
"memory(GiB)": 47.91,
|
| 98 |
+
"step": 40,
|
| 99 |
+
"token_acc": 0.5351251451381303,
|
| 100 |
+
"train_speed(iter/s)": 0.075801
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03374578177727784,
|
| 104 |
+
"grad_norm": 1.146021109197427,
|
| 105 |
+
"learning_rate": 3.358208955223881e-05,
|
| 106 |
+
"loss": 1.7983106613159179,
|
| 107 |
+
"memory(GiB)": 47.91,
|
| 108 |
+
"step": 45,
|
| 109 |
+
"token_acc": 0.5089715774966911,
|
| 110 |
+
"train_speed(iter/s)": 0.076109
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.03749531308586427,
|
| 114 |
+
"grad_norm": 1.5698258174226443,
|
| 115 |
+
"learning_rate": 3.73134328358209e-05,
|
| 116 |
+
"loss": 1.8109302520751953,
|
| 117 |
+
"memory(GiB)": 47.91,
|
| 118 |
+
"step": 50,
|
| 119 |
+
"token_acc": 0.5358898788504374,
|
| 120 |
+
"train_speed(iter/s)": 0.076245
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.03749531308586427,
|
| 124 |
+
"eval_loss": 1.8122611045837402,
|
| 125 |
+
"eval_runtime": 42.4804,
|
| 126 |
+
"eval_samples_per_second": 60.828,
|
| 127 |
+
"eval_steps_per_second": 1.271,
|
| 128 |
+
"eval_token_acc": 0.5341195686265504,
|
| 129 |
+
"step": 50
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0412448443944507,
|
| 133 |
+
"grad_norm": 1.4262062071026742,
|
| 134 |
+
"learning_rate": 4.104477611940299e-05,
|
| 135 |
+
"loss": 1.8131763458251953,
|
| 136 |
+
"memory(GiB)": 55.02,
|
| 137 |
+
"step": 55,
|
| 138 |
+
"token_acc": 0.5220410448634508,
|
| 139 |
+
"train_speed(iter/s)": 0.067836
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.04499437570303712,
|
| 143 |
+
"grad_norm": 1.1215549627054104,
|
| 144 |
+
"learning_rate": 4.477611940298508e-05,
|
| 145 |
+
"loss": 1.809659194946289,
|
| 146 |
+
"memory(GiB)": 55.02,
|
| 147 |
+
"step": 60,
|
| 148 |
+
"token_acc": 0.5253696303037497,
|
| 149 |
+
"train_speed(iter/s)": 0.06858
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.048743907011623545,
|
| 153 |
+
"grad_norm": 1.7375128000683862,
|
| 154 |
+
"learning_rate": 4.850746268656717e-05,
|
| 155 |
+
"loss": 1.7928394317626952,
|
| 156 |
+
"memory(GiB)": 55.02,
|
| 157 |
+
"step": 65,
|
| 158 |
+
"token_acc": 0.5300976924860853,
|
| 159 |
+
"train_speed(iter/s)": 0.069253
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.05249343832020997,
|
| 163 |
+
"grad_norm": 1.423557726174795,
|
| 164 |
+
"learning_rate": 5.223880597014925e-05,
|
| 165 |
+
"loss": 1.7990863800048829,
|
| 166 |
+
"memory(GiB)": 55.02,
|
| 167 |
+
"step": 70,
|
| 168 |
+
"token_acc": 0.5473741492942418,
|
| 169 |
+
"train_speed(iter/s)": 0.069789
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.0562429696287964,
|
| 173 |
+
"grad_norm": 1.6209768045771702,
|
| 174 |
+
"learning_rate": 5.5970149253731344e-05,
|
| 175 |
+
"loss": 1.777482795715332,
|
| 176 |
+
"memory(GiB)": 55.02,
|
| 177 |
+
"step": 75,
|
| 178 |
+
"token_acc": 0.5445959451741862,
|
| 179 |
+
"train_speed(iter/s)": 0.070304
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.05999250093738283,
|
| 183 |
+
"grad_norm": 1.6996413227697587,
|
| 184 |
+
"learning_rate": 5.970149253731343e-05,
|
| 185 |
+
"loss": 1.7891302108764648,
|
| 186 |
+
"memory(GiB)": 55.02,
|
| 187 |
+
"step": 80,
|
| 188 |
+
"token_acc": 0.5425542501177547,
|
| 189 |
+
"train_speed(iter/s)": 0.07076
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.06374203224596925,
|
| 193 |
+
"grad_norm": 1.4408857516105606,
|
| 194 |
+
"learning_rate": 6.343283582089553e-05,
|
| 195 |
+
"loss": 1.8073835372924805,
|
| 196 |
+
"memory(GiB)": 55.02,
|
| 197 |
+
"step": 85,
|
| 198 |
+
"token_acc": 0.5275390258923073,
|
| 199 |
+
"train_speed(iter/s)": 0.07119
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.06749156355455568,
|
| 203 |
+
"grad_norm": 2.407811348410657,
|
| 204 |
+
"learning_rate": 6.716417910447762e-05,
|
| 205 |
+
"loss": 1.8137977600097657,
|
| 206 |
+
"memory(GiB)": 55.02,
|
| 207 |
+
"step": 90,
|
| 208 |
+
"token_acc": 0.5441885377611141,
|
| 209 |
+
"train_speed(iter/s)": 0.071575
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.0712410948631421,
|
| 213 |
+
"grad_norm": 1.5095105702162952,
|
| 214 |
+
"learning_rate": 7.089552238805971e-05,
|
| 215 |
+
"loss": 1.8175329208374023,
|
| 216 |
+
"memory(GiB)": 55.02,
|
| 217 |
+
"step": 95,
|
| 218 |
+
"token_acc": 0.5229251227468275,
|
| 219 |
+
"train_speed(iter/s)": 0.071905
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07499062617172854,
|
| 223 |
+
"grad_norm": 1.0574913783987563,
|
| 224 |
+
"learning_rate": 7.46268656716418e-05,
|
| 225 |
+
"loss": 1.808732795715332,
|
| 226 |
+
"memory(GiB)": 55.02,
|
| 227 |
+
"step": 100,
|
| 228 |
+
"token_acc": 0.5635726256778811,
|
| 229 |
+
"train_speed(iter/s)": 0.072206
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.07499062617172854,
|
| 233 |
+
"eval_loss": 1.8130474090576172,
|
| 234 |
+
"eval_runtime": 42.5277,
|
| 235 |
+
"eval_samples_per_second": 60.76,
|
| 236 |
+
"eval_steps_per_second": 1.27,
|
| 237 |
+
"eval_token_acc": 0.5359635692175534,
|
| 238 |
+
"step": 100
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.07874015748031496,
|
| 242 |
+
"grad_norm": 1.5241300762246714,
|
| 243 |
+
"learning_rate": 7.835820895522389e-05,
|
| 244 |
+
"loss": 1.8181468963623046,
|
| 245 |
+
"memory(GiB)": 55.02,
|
| 246 |
+
"step": 105,
|
| 247 |
+
"token_acc": 0.5268113227691311,
|
| 248 |
+
"train_speed(iter/s)": 0.068156
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0824896887889014,
|
| 252 |
+
"grad_norm": 1.3782073862037376,
|
| 253 |
+
"learning_rate": 8.208955223880597e-05,
|
| 254 |
+
"loss": 1.7923551559448243,
|
| 255 |
+
"memory(GiB)": 55.02,
|
| 256 |
+
"step": 110,
|
| 257 |
+
"token_acc": 0.5239069242557339,
|
| 258 |
+
"train_speed(iter/s)": 0.068562
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.08623922009748781,
|
| 262 |
+
"grad_norm": 1.3709547635473227,
|
| 263 |
+
"learning_rate": 8.582089552238807e-05,
|
| 264 |
+
"loss": 1.805144691467285,
|
| 265 |
+
"memory(GiB)": 55.02,
|
| 266 |
+
"step": 115,
|
| 267 |
+
"token_acc": 0.5338689998719254,
|
| 268 |
+
"train_speed(iter/s)": 0.068944
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.08998875140607424,
|
| 272 |
+
"grad_norm": 1.5697704322666293,
|
| 273 |
+
"learning_rate": 8.955223880597016e-05,
|
| 274 |
+
"loss": 1.8119869232177734,
|
| 275 |
+
"memory(GiB)": 55.02,
|
| 276 |
+
"step": 120,
|
| 277 |
+
"token_acc": 0.5432564966474297,
|
| 278 |
+
"train_speed(iter/s)": 0.069307
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.09373828271466067,
|
| 282 |
+
"grad_norm": 1.4352205441144528,
|
| 283 |
+
"learning_rate": 9.328358208955224e-05,
|
| 284 |
+
"loss": 1.8144481658935547,
|
| 285 |
+
"memory(GiB)": 55.02,
|
| 286 |
+
"step": 125,
|
| 287 |
+
"token_acc": 0.5343267984500858,
|
| 288 |
+
"train_speed(iter/s)": 0.069625
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.09748781402324709,
|
| 292 |
+
"grad_norm": 1.0772098175930582,
|
| 293 |
+
"learning_rate": 9.701492537313434e-05,
|
| 294 |
+
"loss": 1.8078815460205078,
|
| 295 |
+
"memory(GiB)": 55.02,
|
| 296 |
+
"step": 130,
|
| 297 |
+
"token_acc": 0.5310152421530283,
|
| 298 |
+
"train_speed(iter/s)": 0.06993
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.10123734533183353,
|
| 302 |
+
"grad_norm": 1.2112324603884481,
|
| 303 |
+
"learning_rate": 9.999996151315634e-05,
|
| 304 |
+
"loss": 1.8040082931518555,
|
| 305 |
+
"memory(GiB)": 55.02,
|
| 306 |
+
"step": 135,
|
| 307 |
+
"token_acc": 0.562970995219292,
|
| 308 |
+
"train_speed(iter/s)": 0.070196
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.10498687664041995,
|
| 312 |
+
"grad_norm": 1.0656904759090435,
|
| 313 |
+
"learning_rate": 9.999861447984952e-05,
|
| 314 |
+
"loss": 1.8146705627441406,
|
| 315 |
+
"memory(GiB)": 55.02,
|
| 316 |
+
"step": 140,
|
| 317 |
+
"token_acc": 0.5615102804730743,
|
| 318 |
+
"train_speed(iter/s)": 0.070439
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.10873640794900638,
|
| 322 |
+
"grad_norm": 2.0685657658744496,
|
| 323 |
+
"learning_rate": 9.999534316360888e-05,
|
| 324 |
+
"loss": 1.8131877899169921,
|
| 325 |
+
"memory(GiB)": 55.02,
|
| 326 |
+
"step": 145,
|
| 327 |
+
"token_acc": 0.5601988044865337,
|
| 328 |
+
"train_speed(iter/s)": 0.070652
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.1124859392575928,
|
| 332 |
+
"grad_norm": 1.128936311892965,
|
| 333 |
+
"learning_rate": 9.99901476903367e-05,
|
| 334 |
+
"loss": 1.8154624938964843,
|
| 335 |
+
"memory(GiB)": 55.02,
|
| 336 |
+
"step": 150,
|
| 337 |
+
"token_acc": 0.5398418512306165,
|
| 338 |
+
"train_speed(iter/s)": 0.070881
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1124859392575928,
|
| 342 |
+
"eval_loss": 1.8072588443756104,
|
| 343 |
+
"eval_runtime": 42.4992,
|
| 344 |
+
"eval_samples_per_second": 60.801,
|
| 345 |
+
"eval_steps_per_second": 1.271,
|
| 346 |
+
"eval_token_acc": 0.5414565622774629,
|
| 347 |
+
"step": 150
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.11623547056617922,
|
| 351 |
+
"grad_norm": 1.3097337031305452,
|
| 352 |
+
"learning_rate": 9.998302825998971e-05,
|
| 353 |
+
"loss": 1.835479736328125,
|
| 354 |
+
"memory(GiB)": 55.02,
|
| 355 |
+
"step": 155,
|
| 356 |
+
"token_acc": 0.5200763092872446,
|
| 357 |
+
"train_speed(iter/s)": 0.068272
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.11998500187476566,
|
| 361 |
+
"grad_norm": 1.3791667223824373,
|
| 362 |
+
"learning_rate": 9.997398514657146e-05,
|
| 363 |
+
"loss": 1.7949041366577148,
|
| 364 |
+
"memory(GiB)": 55.02,
|
| 365 |
+
"step": 160,
|
| 366 |
+
"token_acc": 0.5402708481824662,
|
| 367 |
+
"train_speed(iter/s)": 0.068538
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.12373453318335208,
|
| 371 |
+
"grad_norm": 1.3445038043799733,
|
| 372 |
+
"learning_rate": 9.99630186981218e-05,
|
| 373 |
+
"loss": 1.7906240463256835,
|
| 374 |
+
"memory(GiB)": 55.02,
|
| 375 |
+
"step": 165,
|
| 376 |
+
"token_acc": 0.5636040993739283,
|
| 377 |
+
"train_speed(iter/s)": 0.068806
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.1274840644919385,
|
| 381 |
+
"grad_norm": 1.052591418120588,
|
| 382 |
+
"learning_rate": 9.99501293367034e-05,
|
| 383 |
+
"loss": 1.7821390151977539,
|
| 384 |
+
"memory(GiB)": 55.02,
|
| 385 |
+
"step": 170,
|
| 386 |
+
"token_acc": 0.5525334677793846,
|
| 387 |
+
"train_speed(iter/s)": 0.069054
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13123359580052493,
|
| 391 |
+
"grad_norm": 1.7209094321145573,
|
| 392 |
+
"learning_rate": 9.993531755838559e-05,
|
| 393 |
+
"loss": 1.779231643676758,
|
| 394 |
+
"memory(GiB)": 55.02,
|
| 395 |
+
"step": 175,
|
| 396 |
+
"token_acc": 0.555622240004742,
|
| 397 |
+
"train_speed(iter/s)": 0.069294
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.13498312710911137,
|
| 401 |
+
"grad_norm": 1.3020374131788652,
|
| 402 |
+
"learning_rate": 9.991858393322518e-05,
|
| 403 |
+
"loss": 1.7615657806396485,
|
| 404 |
+
"memory(GiB)": 55.02,
|
| 405 |
+
"step": 180,
|
| 406 |
+
"token_acc": 0.5600944200380303,
|
| 407 |
+
"train_speed(iter/s)": 0.069521
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.13873265841769777,
|
| 411 |
+
"grad_norm": 0.8409092209708593,
|
| 412 |
+
"learning_rate": 9.989992910524462e-05,
|
| 413 |
+
"loss": 1.7596569061279297,
|
| 414 |
+
"memory(GiB)": 55.02,
|
| 415 |
+
"step": 185,
|
| 416 |
+
"token_acc": 0.547273661123296,
|
| 417 |
+
"train_speed(iter/s)": 0.069737
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.1424821897262842,
|
| 421 |
+
"grad_norm": 0.9036013980450643,
|
| 422 |
+
"learning_rate": 9.987935379240715e-05,
|
| 423 |
+
"loss": 1.7873863220214843,
|
| 424 |
+
"memory(GiB)": 55.02,
|
| 425 |
+
"step": 190,
|
| 426 |
+
"token_acc": 0.5532748673298422,
|
| 427 |
+
"train_speed(iter/s)": 0.069945
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.14623172103487064,
|
| 431 |
+
"grad_norm": 0.9391195058494871,
|
| 432 |
+
"learning_rate": 9.985685878658917e-05,
|
| 433 |
+
"loss": 1.7729263305664062,
|
| 434 |
+
"memory(GiB)": 55.02,
|
| 435 |
+
"step": 195,
|
| 436 |
+
"token_acc": 0.5603855027679431,
|
| 437 |
+
"train_speed(iter/s)": 0.070138
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.14998125234345708,
|
| 441 |
+
"grad_norm": 1.156160947315859,
|
| 442 |
+
"learning_rate": 9.98324449535498e-05,
|
| 443 |
+
"loss": 1.7558784484863281,
|
| 444 |
+
"memory(GiB)": 55.02,
|
| 445 |
+
"step": 200,
|
| 446 |
+
"token_acc": 0.574762321861274,
|
| 447 |
+
"train_speed(iter/s)": 0.07032
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.14998125234345708,
|
| 451 |
+
"eval_loss": 1.7644097805023193,
|
| 452 |
+
"eval_runtime": 42.6196,
|
| 453 |
+
"eval_samples_per_second": 60.629,
|
| 454 |
+
"eval_steps_per_second": 1.267,
|
| 455 |
+
"eval_token_acc": 0.5549946829655976,
|
| 456 |
+
"step": 200
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.15373078365204348,
|
| 460 |
+
"grad_norm": 0.8530822052579083,
|
| 461 |
+
"learning_rate": 9.98061132328975e-05,
|
| 462 |
+
"loss": 1.7540859222412108,
|
| 463 |
+
"memory(GiB)": 55.02,
|
| 464 |
+
"step": 205,
|
| 465 |
+
"token_acc": 0.5509568615941848,
|
| 466 |
+
"train_speed(iter/s)": 0.068046
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.15748031496062992,
|
| 470 |
+
"grad_norm": 0.9405436575808871,
|
| 471 |
+
"learning_rate": 9.977786463805399e-05,
|
| 472 |
+
"loss": 1.748789405822754,
|
| 473 |
+
"memory(GiB)": 55.02,
|
| 474 |
+
"step": 210,
|
| 475 |
+
"token_acc": 0.5745406513681887,
|
| 476 |
+
"train_speed(iter/s)": 0.068248
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.16122984626921635,
|
| 480 |
+
"grad_norm": 0.8916504651164545,
|
| 481 |
+
"learning_rate": 9.974770025621516e-05,
|
| 482 |
+
"loss": 1.7355411529541016,
|
| 483 |
+
"memory(GiB)": 55.02,
|
| 484 |
+
"step": 215,
|
| 485 |
+
"token_acc": 0.54879880877993,
|
| 486 |
+
"train_speed(iter/s)": 0.068448
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.1649793775778028,
|
| 490 |
+
"grad_norm": 0.6167439672546616,
|
| 491 |
+
"learning_rate": 9.97156212483093e-05,
|
| 492 |
+
"loss": 1.7479259490966796,
|
| 493 |
+
"memory(GiB)": 55.02,
|
| 494 |
+
"step": 220,
|
| 495 |
+
"token_acc": 0.5428539106579747,
|
| 496 |
+
"train_speed(iter/s)": 0.068652
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.1687289088863892,
|
| 500 |
+
"grad_norm": 0.9497065226845872,
|
| 501 |
+
"learning_rate": 9.968162884895236e-05,
|
| 502 |
+
"loss": 1.7566562652587892,
|
| 503 |
+
"memory(GiB)": 55.02,
|
| 504 |
+
"step": 225,
|
| 505 |
+
"token_acc": 0.5665545348686765,
|
| 506 |
+
"train_speed(iter/s)": 0.068836
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.17247844019497563,
|
| 510 |
+
"grad_norm": 0.9520763907053652,
|
| 511 |
+
"learning_rate": 9.964572436640045e-05,
|
| 512 |
+
"loss": 1.7468915939331056,
|
| 513 |
+
"memory(GiB)": 55.02,
|
| 514 |
+
"step": 230,
|
| 515 |
+
"token_acc": 0.5804798709412955,
|
| 516 |
+
"train_speed(iter/s)": 0.069016
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.17622797150356206,
|
| 520 |
+
"grad_norm": 0.879699483363928,
|
| 521 |
+
"learning_rate": 9.960790918249956e-05,
|
| 522 |
+
"loss": 1.7292707443237305,
|
| 523 |
+
"memory(GiB)": 55.02,
|
| 524 |
+
"step": 235,
|
| 525 |
+
"token_acc": 0.546071485432256,
|
| 526 |
+
"train_speed(iter/s)": 0.06918
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.17997750281214847,
|
| 530 |
+
"grad_norm": 0.812353538688655,
|
| 531 |
+
"learning_rate": 9.956818475263227e-05,
|
| 532 |
+
"loss": 1.742987823486328,
|
| 533 |
+
"memory(GiB)": 55.02,
|
| 534 |
+
"step": 240,
|
| 535 |
+
"token_acc": 0.5293990201491442,
|
| 536 |
+
"train_speed(iter/s)": 0.069354
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.1837270341207349,
|
| 540 |
+
"grad_norm": 1.2390489337415873,
|
| 541 |
+
"learning_rate": 9.952655260566177e-05,
|
| 542 |
+
"loss": 1.7355167388916015,
|
| 543 |
+
"memory(GiB)": 55.02,
|
| 544 |
+
"step": 245,
|
| 545 |
+
"token_acc": 0.5857872169461704,
|
| 546 |
+
"train_speed(iter/s)": 0.06951
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.18747656542932134,
|
| 550 |
+
"grad_norm": 1.2017374168858055,
|
| 551 |
+
"learning_rate": 9.948301434387308e-05,
|
| 552 |
+
"loss": 1.707486343383789,
|
| 553 |
+
"memory(GiB)": 55.02,
|
| 554 |
+
"step": 250,
|
| 555 |
+
"token_acc": 0.5416734483506944,
|
| 556 |
+
"train_speed(iter/s)": 0.069669
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.18747656542932134,
|
| 560 |
+
"eval_loss": 1.726110816001892,
|
| 561 |
+
"eval_runtime": 42.7656,
|
| 562 |
+
"eval_samples_per_second": 60.422,
|
| 563 |
+
"eval_steps_per_second": 1.263,
|
| 564 |
+
"eval_token_acc": 0.5661786487906093,
|
| 565 |
+
"step": 250
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.19122609673790777,
|
| 569 |
+
"grad_norm": 0.9535729080905381,
|
| 570 |
+
"learning_rate": 9.943757164291131e-05,
|
| 571 |
+
"loss": 1.7093420028686523,
|
| 572 |
+
"memory(GiB)": 56.25,
|
| 573 |
+
"step": 255,
|
| 574 |
+
"token_acc": 0.5646496045754068,
|
| 575 |
+
"train_speed(iter/s)": 0.068117
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.19497562804649418,
|
| 579 |
+
"grad_norm": 0.8083691640470303,
|
| 580 |
+
"learning_rate": 9.939022625171723e-05,
|
| 581 |
+
"loss": 1.7073307037353516,
|
| 582 |
+
"memory(GiB)": 56.25,
|
| 583 |
+
"step": 260,
|
| 584 |
+
"token_acc": 0.590283475213148,
|
| 585 |
+
"train_speed(iter/s)": 0.068292
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.19872515935508062,
|
| 589 |
+
"grad_norm": 1.044808052165735,
|
| 590 |
+
"learning_rate": 9.934097999245984e-05,
|
| 591 |
+
"loss": 1.7080917358398438,
|
| 592 |
+
"memory(GiB)": 56.25,
|
| 593 |
+
"step": 265,
|
| 594 |
+
"token_acc": 0.5553280666197902,
|
| 595 |
+
"train_speed(iter/s)": 0.068466
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.20247469066366705,
|
| 599 |
+
"grad_norm": 0.6908513906497431,
|
| 600 |
+
"learning_rate": 9.928983476046643e-05,
|
| 601 |
+
"loss": 1.7027660369873048,
|
| 602 |
+
"memory(GiB)": 56.25,
|
| 603 |
+
"step": 270,
|
| 604 |
+
"token_acc": 0.5782281084573244,
|
| 605 |
+
"train_speed(iter/s)": 0.06862
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.20622422197225346,
|
| 609 |
+
"grad_norm": 0.9757034569407818,
|
| 610 |
+
"learning_rate": 9.923679252414947e-05,
|
| 611 |
+
"loss": 1.706964111328125,
|
| 612 |
+
"memory(GiB)": 56.25,
|
| 613 |
+
"step": 275,
|
| 614 |
+
"token_acc": 0.5773073735969484,
|
| 615 |
+
"train_speed(iter/s)": 0.068778
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.2099737532808399,
|
| 619 |
+
"grad_norm": 0.8130401940526767,
|
| 620 |
+
"learning_rate": 9.918185532493094e-05,
|
| 621 |
+
"loss": 1.6927738189697266,
|
| 622 |
+
"memory(GiB)": 56.25,
|
| 623 |
+
"step": 280,
|
| 624 |
+
"token_acc": 0.564187791083405,
|
| 625 |
+
"train_speed(iter/s)": 0.068932
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.21372328458942633,
|
| 629 |
+
"grad_norm": 0.7101884022135254,
|
| 630 |
+
"learning_rate": 9.912502527716371e-05,
|
| 631 |
+
"loss": 1.7061391830444337,
|
| 632 |
+
"memory(GiB)": 56.25,
|
| 633 |
+
"step": 285,
|
| 634 |
+
"token_acc": 0.567270352694357,
|
| 635 |
+
"train_speed(iter/s)": 0.069077
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.21747281589801276,
|
| 639 |
+
"grad_norm": 1.0405770482370391,
|
| 640 |
+
"learning_rate": 9.906630456805024e-05,
|
| 641 |
+
"loss": 1.693251609802246,
|
| 642 |
+
"memory(GiB)": 56.25,
|
| 643 |
+
"step": 290,
|
| 644 |
+
"token_acc": 0.5830466566329822,
|
| 645 |
+
"train_speed(iter/s)": 0.06922
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.22122234720659917,
|
| 649 |
+
"grad_norm": 0.7173171146227341,
|
| 650 |
+
"learning_rate": 9.900569545755829e-05,
|
| 651 |
+
"loss": 1.734977912902832,
|
| 652 |
+
"memory(GiB)": 56.25,
|
| 653 |
+
"step": 295,
|
| 654 |
+
"token_acc": 0.5786780061551989,
|
| 655 |
+
"train_speed(iter/s)": 0.069349
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.2249718785151856,
|
| 659 |
+
"grad_norm": 0.8583527878554754,
|
| 660 |
+
"learning_rate": 9.894320027833404e-05,
|
| 661 |
+
"loss": 1.6970809936523437,
|
| 662 |
+
"memory(GiB)": 56.25,
|
| 663 |
+
"step": 300,
|
| 664 |
+
"token_acc": 0.5620832531039843,
|
| 665 |
+
"train_speed(iter/s)": 0.069485
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.2249718785151856,
|
| 669 |
+
"eval_loss": 1.6963896751403809,
|
| 670 |
+
"eval_runtime": 42.3215,
|
| 671 |
+
"eval_samples_per_second": 61.056,
|
| 672 |
+
"eval_steps_per_second": 1.276,
|
| 673 |
+
"eval_token_acc": 0.5736050898401073,
|
| 674 |
+
"step": 300
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.22872140982377204,
|
| 678 |
+
"grad_norm": 1.02219656563603,
|
| 679 |
+
"learning_rate": 9.887882143561232e-05,
|
| 680 |
+
"loss": 1.692136573791504,
|
| 681 |
+
"memory(GiB)": 56.25,
|
| 682 |
+
"step": 305,
|
| 683 |
+
"token_acc": 0.5753743839396092,
|
| 684 |
+
"train_speed(iter/s)": 0.068161
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.23247094113235844,
|
| 688 |
+
"grad_norm": 1.0265502084931866,
|
| 689 |
+
"learning_rate": 9.881256140712389e-05,
|
| 690 |
+
"loss": 1.681289291381836,
|
| 691 |
+
"memory(GiB)": 56.25,
|
| 692 |
+
"step": 310,
|
| 693 |
+
"token_acc": 0.5533130928258396,
|
| 694 |
+
"train_speed(iter/s)": 0.068309
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.23622047244094488,
|
| 698 |
+
"grad_norm": 0.6785024630526387,
|
| 699 |
+
"learning_rate": 9.87444227430003e-05,
|
| 700 |
+
"loss": 1.700368881225586,
|
| 701 |
+
"memory(GiB)": 56.25,
|
| 702 |
+
"step": 315,
|
| 703 |
+
"token_acc": 0.5594038867929421,
|
| 704 |
+
"train_speed(iter/s)": 0.068441
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.2399700037495313,
|
| 708 |
+
"grad_norm": 0.8186982112011071,
|
| 709 |
+
"learning_rate": 9.867440806567561e-05,
|
| 710 |
+
"loss": 1.651713752746582,
|
| 711 |
+
"memory(GiB)": 56.25,
|
| 712 |
+
"step": 320,
|
| 713 |
+
"token_acc": 0.5809736507806417,
|
| 714 |
+
"train_speed(iter/s)": 0.068583
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.24371953505811775,
|
| 718 |
+
"grad_norm": 0.7708218517714257,
|
| 719 |
+
"learning_rate": 9.860252006978539e-05,
|
| 720 |
+
"loss": 1.6446165084838866,
|
| 721 |
+
"memory(GiB)": 56.25,
|
| 722 |
+
"step": 325,
|
| 723 |
+
"token_acc": 0.5725964461860685,
|
| 724 |
+
"train_speed(iter/s)": 0.068723
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.24746906636670415,
|
| 728 |
+
"grad_norm": 0.7939878893188687,
|
| 729 |
+
"learning_rate": 9.852876152206324e-05,
|
| 730 |
+
"loss": 1.6771530151367187,
|
| 731 |
+
"memory(GiB)": 56.25,
|
| 732 |
+
"step": 330,
|
| 733 |
+
"token_acc": 0.6100589964409382,
|
| 734 |
+
"train_speed(iter/s)": 0.068845
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.2512185976752906,
|
| 738 |
+
"grad_norm": 0.8766685163862739,
|
| 739 |
+
"learning_rate": 9.84531352612341e-05,
|
| 740 |
+
"loss": 1.6633163452148438,
|
| 741 |
+
"memory(GiB)": 56.25,
|
| 742 |
+
"step": 335,
|
| 743 |
+
"token_acc": 0.5594746182921707,
|
| 744 |
+
"train_speed(iter/s)": 0.068967
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.254968128983877,
|
| 748 |
+
"grad_norm": 0.790969923079702,
|
| 749 |
+
"learning_rate": 9.837564419790505e-05,
|
| 750 |
+
"loss": 1.6586078643798827,
|
| 751 |
+
"memory(GiB)": 56.25,
|
| 752 |
+
"step": 340,
|
| 753 |
+
"token_acc": 0.5988987217305801,
|
| 754 |
+
"train_speed(iter/s)": 0.06909
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.25871766029246346,
|
| 758 |
+
"grad_norm": 0.6770725447273741,
|
| 759 |
+
"learning_rate": 9.829629131445342e-05,
|
| 760 |
+
"loss": 1.668022918701172,
|
| 761 |
+
"memory(GiB)": 56.25,
|
| 762 |
+
"step": 345,
|
| 763 |
+
"token_acc": 0.5707568963174344,
|
| 764 |
+
"train_speed(iter/s)": 0.069203
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.26246719160104987,
|
| 768 |
+
"grad_norm": 0.7799933153419535,
|
| 769 |
+
"learning_rate": 9.821507966491177e-05,
|
| 770 |
+
"loss": 1.6570024490356445,
|
| 771 |
+
"memory(GiB)": 56.25,
|
| 772 |
+
"step": 350,
|
| 773 |
+
"token_acc": 0.5849244653182151,
|
| 774 |
+
"train_speed(iter/s)": 0.069316
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.26246719160104987,
|
| 778 |
+
"eval_loss": 1.6741433143615723,
|
| 779 |
+
"eval_runtime": 42.6309,
|
| 780 |
+
"eval_samples_per_second": 60.613,
|
| 781 |
+
"eval_steps_per_second": 1.267,
|
| 782 |
+
"eval_token_acc": 0.5806211005050056,
|
| 783 |
+
"step": 350
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.26621672290963627,
|
| 787 |
+
"grad_norm": 0.6513714817338814,
|
| 788 |
+
"learning_rate": 9.813201237485055e-05,
|
| 789 |
+
"loss": 1.6708358764648437,
|
| 790 |
+
"memory(GiB)": 56.25,
|
| 791 |
+
"step": 355,
|
| 792 |
+
"token_acc": 0.5786368647267129,
|
| 793 |
+
"train_speed(iter/s)": 0.068202
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.26996625421822273,
|
| 797 |
+
"grad_norm": 0.7272397663811043,
|
| 798 |
+
"learning_rate": 9.804709264125771e-05,
|
| 799 |
+
"loss": 1.6509124755859375,
|
| 800 |
+
"memory(GiB)": 56.25,
|
| 801 |
+
"step": 360,
|
| 802 |
+
"token_acc": 0.566716319672983,
|
| 803 |
+
"train_speed(iter/s)": 0.068328
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.27371578552680914,
|
| 807 |
+
"grad_norm": 0.581610737755514,
|
| 808 |
+
"learning_rate": 9.79603237324157e-05,
|
| 809 |
+
"loss": 1.6543540954589844,
|
| 810 |
+
"memory(GiB)": 56.25,
|
| 811 |
+
"step": 365,
|
| 812 |
+
"token_acc": 0.5954949494949495,
|
| 813 |
+
"train_speed(iter/s)": 0.068449
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.27746531683539555,
|
| 817 |
+
"grad_norm": 0.8524334413256146,
|
| 818 |
+
"learning_rate": 9.787170898777571e-05,
|
| 819 |
+
"loss": 1.6785432815551757,
|
| 820 |
+
"memory(GiB)": 56.25,
|
| 821 |
+
"step": 370,
|
| 822 |
+
"token_acc": 0.5928388613555106,
|
| 823 |
+
"train_speed(iter/s)": 0.068569
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.281214848143982,
|
| 827 |
+
"grad_norm": 0.6680096057004244,
|
| 828 |
+
"learning_rate": 9.7781251817829e-05,
|
| 829 |
+
"loss": 1.6574161529541016,
|
| 830 |
+
"memory(GiB)": 56.25,
|
| 831 |
+
"step": 375,
|
| 832 |
+
"token_acc": 0.5633509769503426,
|
| 833 |
+
"train_speed(iter/s)": 0.068681
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.2849643794525684,
|
| 837 |
+
"grad_norm": 0.9381728731046745,
|
| 838 |
+
"learning_rate": 9.768895570397585e-05,
|
| 839 |
+
"loss": 1.6603784561157227,
|
| 840 |
+
"memory(GiB)": 56.25,
|
| 841 |
+
"step": 380,
|
| 842 |
+
"token_acc": 0.611448285101822,
|
| 843 |
+
"train_speed(iter/s)": 0.068789
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.2887139107611549,
|
| 847 |
+
"grad_norm": 0.9673492617016157,
|
| 848 |
+
"learning_rate": 9.759482419839141e-05,
|
| 849 |
+
"loss": 1.6619821548461915,
|
| 850 |
+
"memory(GiB)": 56.25,
|
| 851 |
+
"step": 385,
|
| 852 |
+
"token_acc": 0.5536853508807126,
|
| 853 |
+
"train_speed(iter/s)": 0.068895
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.2924634420697413,
|
| 857 |
+
"grad_norm": 0.8485890512065627,
|
| 858 |
+
"learning_rate": 9.749886092388907e-05,
|
| 859 |
+
"loss": 1.646490478515625,
|
| 860 |
+
"memory(GiB)": 56.25,
|
| 861 |
+
"step": 390,
|
| 862 |
+
"token_acc": 0.5676264308956791,
|
| 863 |
+
"train_speed(iter/s)": 0.069
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.2962129733783277,
|
| 867 |
+
"grad_norm": 0.7926314982622505,
|
| 868 |
+
"learning_rate": 9.740106957378099e-05,
|
| 869 |
+
"loss": 1.6406166076660156,
|
| 870 |
+
"memory(GiB)": 56.25,
|
| 871 |
+
"step": 395,
|
| 872 |
+
"token_acc": 0.5915257435470332,
|
| 873 |
+
"train_speed(iter/s)": 0.069106
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.29996250468691416,
|
| 877 |
+
"grad_norm": 0.8052678595507339,
|
| 878 |
+
"learning_rate": 9.730145391173599e-05,
|
| 879 |
+
"loss": 1.6465457916259765,
|
| 880 |
+
"memory(GiB)": 56.25,
|
| 881 |
+
"step": 400,
|
| 882 |
+
"token_acc": 0.5853197197506321,
|
| 883 |
+
"train_speed(iter/s)": 0.069213
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.29996250468691416,
|
| 887 |
+
"eval_loss": 1.6552951335906982,
|
| 888 |
+
"eval_runtime": 42.3718,
|
| 889 |
+
"eval_samples_per_second": 60.984,
|
| 890 |
+
"eval_steps_per_second": 1.274,
|
| 891 |
+
"eval_token_acc": 0.5856434723165544,
|
| 892 |
+
"step": 400
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.30371203599550056,
|
| 896 |
+
"grad_norm": 0.7789536014769777,
|
| 897 |
+
"learning_rate": 9.72000177716347e-05,
|
| 898 |
+
"loss": 1.6445446014404297,
|
| 899 |
+
"memory(GiB)": 56.25,
|
| 900 |
+
"step": 405,
|
| 901 |
+
"token_acc": 0.578255304152203,
|
| 902 |
+
"train_speed(iter/s)": 0.068233
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.30746156730408697,
|
| 906 |
+
"grad_norm": 0.8126568097707741,
|
| 907 |
+
"learning_rate": 9.709676505742193e-05,
|
| 908 |
+
"loss": 1.6398021697998046,
|
| 909 |
+
"memory(GiB)": 56.25,
|
| 910 |
+
"step": 410,
|
| 911 |
+
"token_acc": 0.5794440878666021,
|
| 912 |
+
"train_speed(iter/s)": 0.068346
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.31121109861267343,
|
| 916 |
+
"grad_norm": 0.6420718094071365,
|
| 917 |
+
"learning_rate": 9.699169974295654e-05,
|
| 918 |
+
"loss": 1.6642969131469727,
|
| 919 |
+
"memory(GiB)": 56.25,
|
| 920 |
+
"step": 415,
|
| 921 |
+
"token_acc": 0.5700160032687528,
|
| 922 |
+
"train_speed(iter/s)": 0.068457
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.31496062992125984,
|
| 926 |
+
"grad_norm": 0.6668642301588372,
|
| 927 |
+
"learning_rate": 9.688482587185838e-05,
|
| 928 |
+
"loss": 1.6653812408447266,
|
| 929 |
+
"memory(GiB)": 56.25,
|
| 930 |
+
"step": 420,
|
| 931 |
+
"token_acc": 0.5615830496265943,
|
| 932 |
+
"train_speed(iter/s)": 0.068561
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.31871016122984624,
|
| 936 |
+
"grad_norm": 0.8305477780334533,
|
| 937 |
+
"learning_rate": 9.67761475573528e-05,
|
| 938 |
+
"loss": 1.6734373092651367,
|
| 939 |
+
"memory(GiB)": 56.25,
|
| 940 |
+
"step": 425,
|
| 941 |
+
"token_acc": 0.5612940474081656,
|
| 942 |
+
"train_speed(iter/s)": 0.068663
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.3224596925384327,
|
| 946 |
+
"grad_norm": 0.758546566285689,
|
| 947 |
+
"learning_rate": 9.666566898211218e-05,
|
| 948 |
+
"loss": 1.655715560913086,
|
| 949 |
+
"memory(GiB)": 56.25,
|
| 950 |
+
"step": 430,
|
| 951 |
+
"token_acc": 0.578204214461493,
|
| 952 |
+
"train_speed(iter/s)": 0.068764
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.3262092238470191,
|
| 956 |
+
"grad_norm": 0.6981861186884245,
|
| 957 |
+
"learning_rate": 9.655339439809508e-05,
|
| 958 |
+
"loss": 1.6684629440307617,
|
| 959 |
+
"memory(GiB)": 56.25,
|
| 960 |
+
"step": 435,
|
| 961 |
+
"token_acc": 0.5874935542644986,
|
| 962 |
+
"train_speed(iter/s)": 0.068863
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.3299587551556056,
|
| 966 |
+
"grad_norm": 0.6768506175014746,
|
| 967 |
+
"learning_rate": 9.64393281263826e-05,
|
| 968 |
+
"loss": 1.6376283645629883,
|
| 969 |
+
"memory(GiB)": 56.25,
|
| 970 |
+
"step": 440,
|
| 971 |
+
"token_acc": 0.5950883477422244,
|
| 972 |
+
"train_speed(iter/s)": 0.068955
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.333708286464192,
|
| 976 |
+
"grad_norm": 0.7775597964336293,
|
| 977 |
+
"learning_rate": 9.632347455701197e-05,
|
| 978 |
+
"loss": 1.6142234802246094,
|
| 979 |
+
"memory(GiB)": 56.25,
|
| 980 |
+
"step": 445,
|
| 981 |
+
"token_acc": 0.5972855373930368,
|
| 982 |
+
"train_speed(iter/s)": 0.069047
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3374578177727784,
|
| 986 |
+
"grad_norm": 0.7450978759183977,
|
| 987 |
+
"learning_rate": 9.620583814880763e-05,
|
| 988 |
+
"loss": 1.6266536712646484,
|
| 989 |
+
"memory(GiB)": 56.25,
|
| 990 |
+
"step": 450,
|
| 991 |
+
"token_acc": 0.5649326780390181,
|
| 992 |
+
"train_speed(iter/s)": 0.069136
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.3374578177727784,
|
| 996 |
+
"eval_loss": 1.6390810012817383,
|
| 997 |
+
"eval_runtime": 42.2764,
|
| 998 |
+
"eval_samples_per_second": 61.122,
|
| 999 |
+
"eval_steps_per_second": 1.277,
|
| 1000 |
+
"eval_token_acc": 0.5899320189284537,
|
| 1001 |
+
"step": 450
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.34120734908136485,
|
| 1005 |
+
"grad_norm": 0.6302565087261944,
|
| 1006 |
+
"learning_rate": 9.608642342920975e-05,
|
| 1007 |
+
"loss": 1.6206260681152345,
|
| 1008 |
+
"memory(GiB)": 56.25,
|
| 1009 |
+
"step": 455,
|
| 1010 |
+
"token_acc": 0.5890885706403883,
|
| 1011 |
+
"train_speed(iter/s)": 0.068263
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.34495688038995126,
|
| 1015 |
+
"grad_norm": 0.9796158068421448,
|
| 1016 |
+
"learning_rate": 9.59652349940998e-05,
|
| 1017 |
+
"loss": 1.6623401641845703,
|
| 1018 |
+
"memory(GiB)": 56.25,
|
| 1019 |
+
"step": 460,
|
| 1020 |
+
"token_acc": 0.5619712787660901,
|
| 1021 |
+
"train_speed(iter/s)": 0.068359
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.34870641169853767,
|
| 1025 |
+
"grad_norm": 1.7458874608128234,
|
| 1026 |
+
"learning_rate": 9.584227750762378e-05,
|
| 1027 |
+
"loss": 1.653547477722168,
|
| 1028 |
+
"memory(GiB)": 56.25,
|
| 1029 |
+
"step": 465,
|
| 1030 |
+
"token_acc": 0.6316764851100094,
|
| 1031 |
+
"train_speed(iter/s)": 0.068457
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.35245594300712413,
|
| 1035 |
+
"grad_norm": 0.9304654357956746,
|
| 1036 |
+
"learning_rate": 9.571755570201266e-05,
|
| 1037 |
+
"loss": 1.6338262557983398,
|
| 1038 |
+
"memory(GiB)": 56.25,
|
| 1039 |
+
"step": 470,
|
| 1040 |
+
"token_acc": 0.5678030849070204,
|
| 1041 |
+
"train_speed(iter/s)": 0.068551
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.35620547431571054,
|
| 1045 |
+
"grad_norm": 0.9536211806652244,
|
| 1046 |
+
"learning_rate": 9.559107437740032e-05,
|
| 1047 |
+
"loss": 1.611220932006836,
|
| 1048 |
+
"memory(GiB)": 56.25,
|
| 1049 |
+
"step": 475,
|
| 1050 |
+
"token_acc": 0.5698471904704608,
|
| 1051 |
+
"train_speed(iter/s)": 0.068639
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.35995500562429694,
|
| 1055 |
+
"grad_norm": 0.9633939900855533,
|
| 1056 |
+
"learning_rate": 9.54628384016387e-05,
|
| 1057 |
+
"loss": 1.6381212234497071,
|
| 1058 |
+
"memory(GiB)": 56.25,
|
| 1059 |
+
"step": 480,
|
| 1060 |
+
"token_acc": 0.5882325284615643,
|
| 1061 |
+
"train_speed(iter/s)": 0.068724
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.3637045369328834,
|
| 1065 |
+
"grad_norm": 0.7604082522605755,
|
| 1066 |
+
"learning_rate": 9.533285271011061e-05,
|
| 1067 |
+
"loss": 1.6559059143066406,
|
| 1068 |
+
"memory(GiB)": 56.25,
|
| 1069 |
+
"step": 485,
|
| 1070 |
+
"token_acc": 0.6037551339722277,
|
| 1071 |
+
"train_speed(iter/s)": 0.068816
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 0.3674540682414698,
|
| 1075 |
+
"grad_norm": 0.9268230652480057,
|
| 1076 |
+
"learning_rate": 9.520112230553958e-05,
|
| 1077 |
+
"loss": 1.6622413635253905,
|
| 1078 |
+
"memory(GiB)": 56.25,
|
| 1079 |
+
"step": 490,
|
| 1080 |
+
"token_acc": 0.5744411809964741,
|
| 1081 |
+
"train_speed(iter/s)": 0.068899
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.3712035995500562,
|
| 1085 |
+
"grad_norm": 0.6150310572151141,
|
| 1086 |
+
"learning_rate": 9.506765225779753e-05,
|
| 1087 |
+
"loss": 1.635504913330078,
|
| 1088 |
+
"memory(GiB)": 56.25,
|
| 1089 |
+
"step": 495,
|
| 1090 |
+
"token_acc": 0.5890812212187838,
|
| 1091 |
+
"train_speed(iter/s)": 0.068984
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.3749531308586427,
|
| 1095 |
+
"grad_norm": 0.6371372361851609,
|
| 1096 |
+
"learning_rate": 9.493244770370946e-05,
|
| 1097 |
+
"loss": 1.6373722076416015,
|
| 1098 |
+
"memory(GiB)": 56.25,
|
| 1099 |
+
"step": 500,
|
| 1100 |
+
"token_acc": 0.6058406678969068,
|
| 1101 |
+
"train_speed(iter/s)": 0.069065
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3749531308586427,
|
| 1105 |
+
"eval_loss": 1.6263278722763062,
|
| 1106 |
+
"eval_runtime": 42.7074,
|
| 1107 |
+
"eval_samples_per_second": 60.505,
|
| 1108 |
+
"eval_steps_per_second": 1.264,
|
| 1109 |
+
"eval_token_acc": 0.5933668829238414,
|
| 1110 |
+
"step": 500
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.3787026621672291,
|
| 1114 |
+
"grad_norm": 0.6273442347450918,
|
| 1115 |
+
"learning_rate": 9.479551384685594e-05,
|
| 1116 |
+
"loss": 1.640282440185547,
|
| 1117 |
+
"memory(GiB)": 56.25,
|
| 1118 |
+
"step": 505,
|
| 1119 |
+
"token_acc": 0.5944903177323424,
|
| 1120 |
+
"train_speed(iter/s)": 0.068275
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.38245219347581555,
|
| 1124 |
+
"grad_norm": 0.550044323406006,
|
| 1125 |
+
"learning_rate": 9.465685595737263e-05,
|
| 1126 |
+
"loss": 1.6416170120239257,
|
| 1127 |
+
"memory(GiB)": 56.25,
|
| 1128 |
+
"step": 510,
|
| 1129 |
+
"token_acc": 0.6041292759368889,
|
| 1130 |
+
"train_speed(iter/s)": 0.068356
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.38620172478440196,
|
| 1134 |
+
"grad_norm": 0.7138927383895873,
|
| 1135 |
+
"learning_rate": 9.451647937174763e-05,
|
| 1136 |
+
"loss": 1.611105728149414,
|
| 1137 |
+
"memory(GiB)": 56.25,
|
| 1138 |
+
"step": 515,
|
| 1139 |
+
"token_acc": 0.5766065465121516,
|
| 1140 |
+
"train_speed(iter/s)": 0.068441
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.38995125609298836,
|
| 1144 |
+
"grad_norm": 0.7243864042311587,
|
| 1145 |
+
"learning_rate": 9.437438949261602e-05,
|
| 1146 |
+
"loss": 1.5815048217773438,
|
| 1147 |
+
"memory(GiB)": 56.25,
|
| 1148 |
+
"step": 520,
|
| 1149 |
+
"token_acc": 0.6105661252227841,
|
| 1150 |
+
"train_speed(iter/s)": 0.068519
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3937007874015748,
|
| 1154 |
+
"grad_norm": 0.8647763365191747,
|
| 1155 |
+
"learning_rate": 9.423059178855193e-05,
|
| 1156 |
+
"loss": 1.6214418411254883,
|
| 1157 |
+
"memory(GiB)": 56.25,
|
| 1158 |
+
"step": 525,
|
| 1159 |
+
"token_acc": 0.6274994705597038,
|
| 1160 |
+
"train_speed(iter/s)": 0.068603
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.39745031871016123,
|
| 1164 |
+
"grad_norm": 0.6745963449668103,
|
| 1165 |
+
"learning_rate": 9.408509179385806e-05,
|
| 1166 |
+
"loss": 1.6437505722045898,
|
| 1167 |
+
"memory(GiB)": 56.25,
|
| 1168 |
+
"step": 530,
|
| 1169 |
+
"token_acc": 0.5829802117384344,
|
| 1170 |
+
"train_speed(iter/s)": 0.068687
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.40119985001874764,
|
| 1174 |
+
"grad_norm": 0.6493931503120355,
|
| 1175 |
+
"learning_rate": 9.393789510835272e-05,
|
| 1176 |
+
"loss": 1.6190866470336913,
|
| 1177 |
+
"memory(GiB)": 56.25,
|
| 1178 |
+
"step": 535,
|
| 1179 |
+
"token_acc": 0.5781899059024808,
|
| 1180 |
+
"train_speed(iter/s)": 0.068771
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.4049493813273341,
|
| 1184 |
+
"grad_norm": 0.7025286692863906,
|
| 1185 |
+
"learning_rate": 9.378900739715428e-05,
|
| 1186 |
+
"loss": 1.6072065353393554,
|
| 1187 |
+
"memory(GiB)": 56.25,
|
| 1188 |
+
"step": 540,
|
| 1189 |
+
"token_acc": 0.6341984099365581,
|
| 1190 |
+
"train_speed(iter/s)": 0.068849
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.4086989126359205,
|
| 1194 |
+
"grad_norm": 2.625951830662365,
|
| 1195 |
+
"learning_rate": 9.363843439046318e-05,
|
| 1196 |
+
"loss": 1.6233955383300782,
|
| 1197 |
+
"memory(GiB)": 56.25,
|
| 1198 |
+
"step": 545,
|
| 1199 |
+
"token_acc": 0.5952244441336876,
|
| 1200 |
+
"train_speed(iter/s)": 0.068928
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.4124484439445069,
|
| 1204 |
+
"grad_norm": 0.613788773144586,
|
| 1205 |
+
"learning_rate": 9.348618188334133e-05,
|
| 1206 |
+
"loss": 1.6083612442016602,
|
| 1207 |
+
"memory(GiB)": 56.25,
|
| 1208 |
+
"step": 550,
|
| 1209 |
+
"token_acc": 0.5988067170283404,
|
| 1210 |
+
"train_speed(iter/s)": 0.068998
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.4124484439445069,
|
| 1214 |
+
"eval_loss": 1.628952980041504,
|
| 1215 |
+
"eval_runtime": 42.187,
|
| 1216 |
+
"eval_samples_per_second": 61.251,
|
| 1217 |
+
"eval_steps_per_second": 1.28,
|
| 1218 |
+
"eval_token_acc": 0.593999306441859,
|
| 1219 |
+
"step": 550
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.4161979752530934,
|
| 1223 |
+
"grad_norm": 0.5996231652311943,
|
| 1224 |
+
"learning_rate": 9.333225573548915e-05,
|
| 1225 |
+
"loss": 1.610010528564453,
|
| 1226 |
+
"memory(GiB)": 56.25,
|
| 1227 |
+
"step": 555,
|
| 1228 |
+
"token_acc": 0.5909896444579076,
|
| 1229 |
+
"train_speed(iter/s)": 0.068283
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.4199475065616798,
|
| 1233 |
+
"grad_norm": 0.7826419473705719,
|
| 1234 |
+
"learning_rate": 9.317666187101996e-05,
|
| 1235 |
+
"loss": 1.6196056365966798,
|
| 1236 |
+
"memory(GiB)": 56.25,
|
| 1237 |
+
"step": 560,
|
| 1238 |
+
"token_acc": 0.5791037169172522,
|
| 1239 |
+
"train_speed(iter/s)": 0.068361
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.4236970378702662,
|
| 1243 |
+
"grad_norm": 0.9756727744796079,
|
| 1244 |
+
"learning_rate": 9.301940627823207e-05,
|
| 1245 |
+
"loss": 1.6257247924804688,
|
| 1246 |
+
"memory(GiB)": 56.25,
|
| 1247 |
+
"step": 565,
|
| 1248 |
+
"token_acc": 0.5991850555252661,
|
| 1249 |
+
"train_speed(iter/s)": 0.06844
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.42744656917885265,
|
| 1253 |
+
"grad_norm": 0.7274497541989631,
|
| 1254 |
+
"learning_rate": 9.286049500937826e-05,
|
| 1255 |
+
"loss": 1.617807388305664,
|
| 1256 |
+
"memory(GiB)": 56.25,
|
| 1257 |
+
"step": 570,
|
| 1258 |
+
"token_acc": 0.5784111888854946,
|
| 1259 |
+
"train_speed(iter/s)": 0.068507
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.43119610048743906,
|
| 1263 |
+
"grad_norm": 0.7849941490304332,
|
| 1264 |
+
"learning_rate": 9.269993418043284e-05,
|
| 1265 |
+
"loss": 1.5984549522399902,
|
| 1266 |
+
"memory(GiB)": 56.25,
|
| 1267 |
+
"step": 575,
|
| 1268 |
+
"token_acc": 0.5971784544107519,
|
| 1269 |
+
"train_speed(iter/s)": 0.068578
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4349456317960255,
|
| 1273 |
+
"grad_norm": 0.8482897502670447,
|
| 1274 |
+
"learning_rate": 9.253772997085633e-05,
|
| 1275 |
+
"loss": 1.6019691467285155,
|
| 1276 |
+
"memory(GiB)": 56.25,
|
| 1277 |
+
"step": 580,
|
| 1278 |
+
"token_acc": 0.606298323458299,
|
| 1279 |
+
"train_speed(iter/s)": 0.068648
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 0.43869516310461193,
|
| 1283 |
+
"grad_norm": 0.5853686509585286,
|
| 1284 |
+
"learning_rate": 9.237388862335756e-05,
|
| 1285 |
+
"loss": 1.6091520309448242,
|
| 1286 |
+
"memory(GiB)": 56.25,
|
| 1287 |
+
"step": 585,
|
| 1288 |
+
"token_acc": 0.6023047714772315,
|
| 1289 |
+
"train_speed(iter/s)": 0.06872
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.44244469441319834,
|
| 1293 |
+
"grad_norm": 0.8637269083352683,
|
| 1294 |
+
"learning_rate": 9.220841644365343e-05,
|
| 1295 |
+
"loss": 1.6088146209716796,
|
| 1296 |
+
"memory(GiB)": 56.25,
|
| 1297 |
+
"step": 590,
|
| 1298 |
+
"token_acc": 0.6193387930086842,
|
| 1299 |
+
"train_speed(iter/s)": 0.068791
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.4461942257217848,
|
| 1303 |
+
"grad_norm": 0.6214093620067173,
|
| 1304 |
+
"learning_rate": 9.204131980022625e-05,
|
| 1305 |
+
"loss": 1.5891772270202638,
|
| 1306 |
+
"memory(GiB)": 56.25,
|
| 1307 |
+
"step": 595,
|
| 1308 |
+
"token_acc": 0.646423532731377,
|
| 1309 |
+
"train_speed(iter/s)": 0.068864
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 0.4499437570303712,
|
| 1313 |
+
"grad_norm": 0.9268874344751206,
|
| 1314 |
+
"learning_rate": 9.18726051240786e-05,
|
| 1315 |
+
"loss": 1.6200408935546875,
|
| 1316 |
+
"memory(GiB)": 56.25,
|
| 1317 |
+
"step": 600,
|
| 1318 |
+
"token_acc": 0.584342001797349,
|
| 1319 |
+
"train_speed(iter/s)": 0.06893
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.4499437570303712,
|
| 1323 |
+
"eval_loss": 1.6117960214614868,
|
| 1324 |
+
"eval_runtime": 42.3686,
|
| 1325 |
+
"eval_samples_per_second": 60.989,
|
| 1326 |
+
"eval_steps_per_second": 1.275,
|
| 1327 |
+
"eval_token_acc": 0.5979014360973838,
|
| 1328 |
+
"step": 600
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.4536932883389576,
|
| 1332 |
+
"grad_norm": 0.7318396102257045,
|
| 1333 |
+
"learning_rate": 9.170227890848588e-05,
|
| 1334 |
+
"loss": 1.6193307876586913,
|
| 1335 |
+
"memory(GiB)": 56.25,
|
| 1336 |
+
"step": 605,
|
| 1337 |
+
"token_acc": 0.604269745447046,
|
| 1338 |
+
"train_speed(iter/s)": 0.068154
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.4574428196475441,
|
| 1342 |
+
"grad_norm": 0.7983303550845163,
|
| 1343 |
+
"learning_rate": 9.153034770874629e-05,
|
| 1344 |
+
"loss": 1.6229751586914063,
|
| 1345 |
+
"memory(GiB)": 56.25,
|
| 1346 |
+
"step": 610,
|
| 1347 |
+
"token_acc": 0.6151124976497705,
|
| 1348 |
+
"train_speed(iter/s)": 0.06823
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.4611923509561305,
|
| 1352 |
+
"grad_norm": 0.507160284137654,
|
| 1353 |
+
"learning_rate": 9.135681814192869e-05,
|
| 1354 |
+
"loss": 1.602430534362793,
|
| 1355 |
+
"memory(GiB)": 56.25,
|
| 1356 |
+
"step": 615,
|
| 1357 |
+
"token_acc": 0.6092228016033486,
|
| 1358 |
+
"train_speed(iter/s)": 0.068301
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.4649418822647169,
|
| 1362 |
+
"grad_norm": 0.5258826981699198,
|
| 1363 |
+
"learning_rate": 9.118169688661784e-05,
|
| 1364 |
+
"loss": 1.6117786407470702,
|
| 1365 |
+
"memory(GiB)": 56.25,
|
| 1366 |
+
"step": 620,
|
| 1367 |
+
"token_acc": 0.5668433580117113,
|
| 1368 |
+
"train_speed(iter/s)": 0.068374
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.46869141357330335,
|
| 1372 |
+
"grad_norm": 1.390218598435202,
|
| 1373 |
+
"learning_rate": 9.100499068265733e-05,
|
| 1374 |
+
"loss": 1.598494529724121,
|
| 1375 |
+
"memory(GiB)": 56.25,
|
| 1376 |
+
"step": 625,
|
| 1377 |
+
"token_acc": 0.5877093624122839,
|
| 1378 |
+
"train_speed(iter/s)": 0.068447
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.47244094488188976,
|
| 1382 |
+
"grad_norm": 0.8036635924637877,
|
| 1383 |
+
"learning_rate": 9.082670633089028e-05,
|
| 1384 |
+
"loss": 1.6108219146728515,
|
| 1385 |
+
"memory(GiB)": 56.25,
|
| 1386 |
+
"step": 630,
|
| 1387 |
+
"token_acc": 0.5942192935320569,
|
| 1388 |
+
"train_speed(iter/s)": 0.068516
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.47619047619047616,
|
| 1392 |
+
"grad_norm": 0.7380008944811205,
|
| 1393 |
+
"learning_rate": 9.064685069289754e-05,
|
| 1394 |
+
"loss": 1.6038398742675781,
|
| 1395 |
+
"memory(GiB)": 56.25,
|
| 1396 |
+
"step": 635,
|
| 1397 |
+
"token_acc": 0.6017213986528185,
|
| 1398 |
+
"train_speed(iter/s)": 0.068584
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.4799400074990626,
|
| 1402 |
+
"grad_norm": 0.5558563000258591,
|
| 1403 |
+
"learning_rate": 9.046543069073361e-05,
|
| 1404 |
+
"loss": 1.5782081604003906,
|
| 1405 |
+
"memory(GiB)": 56.25,
|
| 1406 |
+
"step": 640,
|
| 1407 |
+
"token_acc": 0.5869027802372941,
|
| 1408 |
+
"train_speed(iter/s)": 0.06865
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.48368953880764903,
|
| 1412 |
+
"grad_norm": 0.6132539290805606,
|
| 1413 |
+
"learning_rate": 9.028245330666025e-05,
|
| 1414 |
+
"loss": 1.597235107421875,
|
| 1415 |
+
"memory(GiB)": 56.25,
|
| 1416 |
+
"step": 645,
|
| 1417 |
+
"token_acc": 0.6101842364722485,
|
| 1418 |
+
"train_speed(iter/s)": 0.068713
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.4874390701162355,
|
| 1422 |
+
"grad_norm": 0.45181597171435856,
|
| 1423 |
+
"learning_rate": 9.009792558287777e-05,
|
| 1424 |
+
"loss": 1.6153556823730468,
|
| 1425 |
+
"memory(GiB)": 56.25,
|
| 1426 |
+
"step": 650,
|
| 1427 |
+
"token_acc": 0.6035294777270052,
|
| 1428 |
+
"train_speed(iter/s)": 0.068777
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.4874390701162355,
|
| 1432 |
+
"eval_loss": 1.5969618558883667,
|
| 1433 |
+
"eval_runtime": 42.5756,
|
| 1434 |
+
"eval_samples_per_second": 60.692,
|
| 1435 |
+
"eval_steps_per_second": 1.268,
|
| 1436 |
+
"eval_token_acc": 0.601238778310023,
|
| 1437 |
+
"step": 650
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.4911886014248219,
|
| 1441 |
+
"grad_norm": 0.41069684753936503,
|
| 1442 |
+
"learning_rate": 8.991185462125393e-05,
|
| 1443 |
+
"loss": 1.5636050224304199,
|
| 1444 |
+
"memory(GiB)": 56.25,
|
| 1445 |
+
"step": 655,
|
| 1446 |
+
"token_acc": 0.5951534341779499,
|
| 1447 |
+
"train_speed(iter/s)": 0.068176
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 0.4949381327334083,
|
| 1451 |
+
"grad_norm": 0.45737215255650476,
|
| 1452 |
+
"learning_rate": 8.972424758305072e-05,
|
| 1453 |
+
"loss": 1.5677555084228516,
|
| 1454 |
+
"memory(GiB)": 56.25,
|
| 1455 |
+
"step": 660,
|
| 1456 |
+
"token_acc": 0.6208551977969955,
|
| 1457 |
+
"train_speed(iter/s)": 0.068244
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.49868766404199477,
|
| 1461 |
+
"grad_norm": 0.7047773041057553,
|
| 1462 |
+
"learning_rate": 8.953511168864865e-05,
|
| 1463 |
+
"loss": 1.6052043914794922,
|
| 1464 |
+
"memory(GiB)": 56.25,
|
| 1465 |
+
"step": 665,
|
| 1466 |
+
"token_acc": 0.634538316284251,
|
| 1467 |
+
"train_speed(iter/s)": 0.068309
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.5024371953505812,
|
| 1471 |
+
"grad_norm": 0.5470681637953386,
|
| 1472 |
+
"learning_rate": 8.934445421726886e-05,
|
| 1473 |
+
"loss": 1.5874359130859375,
|
| 1474 |
+
"memory(GiB)": 56.25,
|
| 1475 |
+
"step": 670,
|
| 1476 |
+
"token_acc": 0.6117839790531931,
|
| 1477 |
+
"train_speed(iter/s)": 0.068375
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.5061867266591676,
|
| 1481 |
+
"grad_norm": 0.7330463344896269,
|
| 1482 |
+
"learning_rate": 8.915228250669314e-05,
|
| 1483 |
+
"loss": 1.5850339889526368,
|
| 1484 |
+
"memory(GiB)": 56.25,
|
| 1485 |
+
"step": 675,
|
| 1486 |
+
"token_acc": 0.6332533951099547,
|
| 1487 |
+
"train_speed(iter/s)": 0.068437
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.509936257967754,
|
| 1491 |
+
"grad_norm": 0.45916529850986554,
|
| 1492 |
+
"learning_rate": 8.895860395298121e-05,
|
| 1493 |
+
"loss": 1.6019588470458985,
|
| 1494 |
+
"memory(GiB)": 56.25,
|
| 1495 |
+
"step": 680,
|
| 1496 |
+
"token_acc": 0.5748860611805885,
|
| 1497 |
+
"train_speed(iter/s)": 0.068501
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.5136857892763405,
|
| 1501 |
+
"grad_norm": 0.5612063670960976,
|
| 1502 |
+
"learning_rate": 8.876342601018639e-05,
|
| 1503 |
+
"loss": 1.584090232849121,
|
| 1504 |
+
"memory(GiB)": 56.25,
|
| 1505 |
+
"step": 685,
|
| 1506 |
+
"token_acc": 0.6142928971273423,
|
| 1507 |
+
"train_speed(iter/s)": 0.068566
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5174353205849269,
|
| 1511 |
+
"grad_norm": 0.617189452034562,
|
| 1512 |
+
"learning_rate": 8.85667561900685e-05,
|
| 1513 |
+
"loss": 1.6062986373901367,
|
| 1514 |
+
"memory(GiB)": 56.25,
|
| 1515 |
+
"step": 690,
|
| 1516 |
+
"token_acc": 0.6123293459414567,
|
| 1517 |
+
"train_speed(iter/s)": 0.068632
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 0.5211848518935133,
|
| 1521 |
+
"grad_norm": 0.5347989405972925,
|
| 1522 |
+
"learning_rate": 8.836860206180486e-05,
|
| 1523 |
+
"loss": 1.566591453552246,
|
| 1524 |
+
"memory(GiB)": 56.25,
|
| 1525 |
+
"step": 695,
|
| 1526 |
+
"token_acc": 0.5955262981124471,
|
| 1527 |
+
"train_speed(iter/s)": 0.068691
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.5249343832020997,
|
| 1531 |
+
"grad_norm": 0.7441035889172485,
|
| 1532 |
+
"learning_rate": 8.816897125169894e-05,
|
| 1533 |
+
"loss": 1.6083082199096679,
|
| 1534 |
+
"memory(GiB)": 56.25,
|
| 1535 |
+
"step": 700,
|
| 1536 |
+
"token_acc": 0.5964849596181591,
|
| 1537 |
+
"train_speed(iter/s)": 0.068751
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.5249343832020997,
|
| 1541 |
+
"eval_loss": 1.5856742858886719,
|
| 1542 |
+
"eval_runtime": 42.8768,
|
| 1543 |
+
"eval_samples_per_second": 60.266,
|
| 1544 |
+
"eval_steps_per_second": 1.259,
|
| 1545 |
+
"eval_token_acc": 0.6039501984568278,
|
| 1546 |
+
"step": 700
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.5286839145106862,
|
| 1550 |
+
"grad_norm": 0.4979964017526753,
|
| 1551 |
+
"learning_rate": 8.796787144288687e-05,
|
| 1552 |
+
"loss": 1.5707754135131835,
|
| 1553 |
+
"memory(GiB)": 56.25,
|
| 1554 |
+
"step": 705,
|
| 1555 |
+
"token_acc": 0.6041424543831413,
|
| 1556 |
+
"train_speed(iter/s)": 0.068164
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5324334458192725,
|
| 1560 |
+
"grad_norm": 0.6632219113708855,
|
| 1561 |
+
"learning_rate": 8.776531037504169e-05,
|
| 1562 |
+
"loss": 1.562371063232422,
|
| 1563 |
+
"memory(GiB)": 56.25,
|
| 1564 |
+
"step": 710,
|
| 1565 |
+
"token_acc": 0.6135038997487191,
|
| 1566 |
+
"train_speed(iter/s)": 0.068224
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 0.536182977127859,
|
| 1570 |
+
"grad_norm": 0.6737198858690441,
|
| 1571 |
+
"learning_rate": 8.75612958440756e-05,
|
| 1572 |
+
"loss": 1.5978498458862305,
|
| 1573 |
+
"memory(GiB)": 56.25,
|
| 1574 |
+
"step": 715,
|
| 1575 |
+
"token_acc": 0.6002256280439149,
|
| 1576 |
+
"train_speed(iter/s)": 0.068286
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.5399325084364455,
|
| 1580 |
+
"grad_norm": 0.6547321060806424,
|
| 1581 |
+
"learning_rate": 8.735583570183974e-05,
|
| 1582 |
+
"loss": 1.5621134757995605,
|
| 1583 |
+
"memory(GiB)": 56.25,
|
| 1584 |
+
"step": 720,
|
| 1585 |
+
"token_acc": 0.565872061965812,
|
| 1586 |
+
"train_speed(iter/s)": 0.068347
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.5436820397450318,
|
| 1590 |
+
"grad_norm": 0.488924503709666,
|
| 1591 |
+
"learning_rate": 8.714893785582215e-05,
|
| 1592 |
+
"loss": 1.5777151107788085,
|
| 1593 |
+
"memory(GiB)": 56.25,
|
| 1594 |
+
"step": 725,
|
| 1595 |
+
"token_acc": 0.6218858668004655,
|
| 1596 |
+
"train_speed(iter/s)": 0.068407
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.5474315710536183,
|
| 1600 |
+
"grad_norm": 0.5270106654729826,
|
| 1601 |
+
"learning_rate": 8.694061026884337e-05,
|
| 1602 |
+
"loss": 1.5883260726928712,
|
| 1603 |
+
"memory(GiB)": 56.25,
|
| 1604 |
+
"step": 730,
|
| 1605 |
+
"token_acc": 0.5985521203466941,
|
| 1606 |
+
"train_speed(iter/s)": 0.068464
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.5511811023622047,
|
| 1610 |
+
"grad_norm": 0.9298667263071514,
|
| 1611 |
+
"learning_rate": 8.673086095874996e-05,
|
| 1612 |
+
"loss": 1.5839271545410156,
|
| 1613 |
+
"memory(GiB)": 56.25,
|
| 1614 |
+
"step": 735,
|
| 1615 |
+
"token_acc": 0.6176172480685197,
|
| 1616 |
+
"train_speed(iter/s)": 0.068519
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 0.5549306336707911,
|
| 1620 |
+
"grad_norm": 0.7413714864196878,
|
| 1621 |
+
"learning_rate": 8.6519697998106e-05,
|
| 1622 |
+
"loss": 1.5891104698181153,
|
| 1623 |
+
"memory(GiB)": 56.25,
|
| 1624 |
+
"step": 740,
|
| 1625 |
+
"token_acc": 0.6105030164792895,
|
| 1626 |
+
"train_speed(iter/s)": 0.068578
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5586801649793776,
|
| 1630 |
+
"grad_norm": 0.6325112297680812,
|
| 1631 |
+
"learning_rate": 8.63071295138823e-05,
|
| 1632 |
+
"loss": 1.562502670288086,
|
| 1633 |
+
"memory(GiB)": 56.25,
|
| 1634 |
+
"step": 745,
|
| 1635 |
+
"token_acc": 0.6175052296992584,
|
| 1636 |
+
"train_speed(iter/s)": 0.068638
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 0.562429696287964,
|
| 1640 |
+
"grad_norm": 0.6304188268964839,
|
| 1641 |
+
"learning_rate": 8.60931636871437e-05,
|
| 1642 |
+
"loss": 1.5628878593444824,
|
| 1643 |
+
"memory(GiB)": 56.25,
|
| 1644 |
+
"step": 750,
|
| 1645 |
+
"token_acc": 0.6012801229766173,
|
| 1646 |
+
"train_speed(iter/s)": 0.068697
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.562429696287964,
|
| 1650 |
+
"eval_loss": 1.5786051750183105,
|
| 1651 |
+
"eval_runtime": 42.4321,
|
| 1652 |
+
"eval_samples_per_second": 60.897,
|
| 1653 |
+
"eval_steps_per_second": 1.273,
|
| 1654 |
+
"eval_token_acc": 0.6057038078900575,
|
| 1655 |
+
"step": 750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.5661792275965505,
|
| 1659 |
+
"grad_norm": 0.5655524603470121,
|
| 1660 |
+
"learning_rate": 8.587780875273417e-05,
|
| 1661 |
+
"loss": 1.5747660636901855,
|
| 1662 |
+
"memory(GiB)": 56.25,
|
| 1663 |
+
"step": 755,
|
| 1664 |
+
"token_acc": 0.6084260843331051,
|
| 1665 |
+
"train_speed(iter/s)": 0.068161
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 0.5699287589051368,
|
| 1669 |
+
"grad_norm": 0.6514440765111553,
|
| 1670 |
+
"learning_rate": 8.566107299895987e-05,
|
| 1671 |
+
"loss": 1.561225700378418,
|
| 1672 |
+
"memory(GiB)": 56.25,
|
| 1673 |
+
"step": 760,
|
| 1674 |
+
"token_acc": 0.5947622375295332,
|
| 1675 |
+
"train_speed(iter/s)": 0.068221
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.5736782902137233,
|
| 1679 |
+
"grad_norm": 0.617533543549024,
|
| 1680 |
+
"learning_rate": 8.544296476727018e-05,
|
| 1681 |
+
"loss": 1.554114532470703,
|
| 1682 |
+
"memory(GiB)": 56.25,
|
| 1683 |
+
"step": 765,
|
| 1684 |
+
"token_acc": 0.6276475389880996,
|
| 1685 |
+
"train_speed(iter/s)": 0.068275
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 0.5774278215223098,
|
| 1689 |
+
"grad_norm": 0.4568963473030676,
|
| 1690 |
+
"learning_rate": 8.52234924519367e-05,
|
| 1691 |
+
"loss": 1.5849886894226075,
|
| 1692 |
+
"memory(GiB)": 56.25,
|
| 1693 |
+
"step": 770,
|
| 1694 |
+
"token_acc": 0.5775132366681662,
|
| 1695 |
+
"train_speed(iter/s)": 0.068332
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 0.5811773528308961,
|
| 1699 |
+
"grad_norm": 0.6476620290579564,
|
| 1700 |
+
"learning_rate": 8.500266449973007e-05,
|
| 1701 |
+
"loss": 1.572746467590332,
|
| 1702 |
+
"memory(GiB)": 56.25,
|
| 1703 |
+
"step": 775,
|
| 1704 |
+
"token_acc": 0.6051636165725559,
|
| 1705 |
+
"train_speed(iter/s)": 0.068387
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.5849268841394826,
|
| 1709 |
+
"grad_norm": 0.6023916375767202,
|
| 1710 |
+
"learning_rate": 8.478048940959503e-05,
|
| 1711 |
+
"loss": 1.5585566520690919,
|
| 1712 |
+
"memory(GiB)": 56.25,
|
| 1713 |
+
"step": 780,
|
| 1714 |
+
"token_acc": 0.6194751720316215,
|
| 1715 |
+
"train_speed(iter/s)": 0.068443
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 0.588676415448069,
|
| 1719 |
+
"grad_norm": 0.6972915914312097,
|
| 1720 |
+
"learning_rate": 8.45569757323232e-05,
|
| 1721 |
+
"loss": 1.5557513236999512,
|
| 1722 |
+
"memory(GiB)": 56.25,
|
| 1723 |
+
"step": 785,
|
| 1724 |
+
"token_acc": 0.6362351325700208,
|
| 1725 |
+
"train_speed(iter/s)": 0.068497
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.5924259467566554,
|
| 1729 |
+
"grad_norm": 0.571582909097701,
|
| 1730 |
+
"learning_rate": 8.433213207022404e-05,
|
| 1731 |
+
"loss": 1.5855814933776855,
|
| 1732 |
+
"memory(GiB)": 56.25,
|
| 1733 |
+
"step": 790,
|
| 1734 |
+
"token_acc": 0.598226788115737,
|
| 1735 |
+
"train_speed(iter/s)": 0.06855
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 0.5961754780652418,
|
| 1739 |
+
"grad_norm": 0.7295276130202991,
|
| 1740 |
+
"learning_rate": 8.410596707679376e-05,
|
| 1741 |
+
"loss": 1.571615505218506,
|
| 1742 |
+
"memory(GiB)": 56.25,
|
| 1743 |
+
"step": 795,
|
| 1744 |
+
"token_acc": 0.6098404656136297,
|
| 1745 |
+
"train_speed(iter/s)": 0.068603
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5999250093738283,
|
| 1749 |
+
"grad_norm": 0.6027046756459332,
|
| 1750 |
+
"learning_rate": 8.387848945638235e-05,
|
| 1751 |
+
"loss": 1.593567180633545,
|
| 1752 |
+
"memory(GiB)": 56.25,
|
| 1753 |
+
"step": 800,
|
| 1754 |
+
"token_acc": 0.59409267045923,
|
| 1755 |
+
"train_speed(iter/s)": 0.068658
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 0.5999250093738283,
|
| 1759 |
+
"eval_loss": 1.5670666694641113,
|
| 1760 |
+
"eval_runtime": 42.3194,
|
| 1761 |
+
"eval_samples_per_second": 61.059,
|
| 1762 |
+
"eval_steps_per_second": 1.276,
|
| 1763 |
+
"eval_token_acc": 0.6087456486147556,
|
| 1764 |
+
"step": 800
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.6036745406824147,
|
| 1768 |
+
"grad_norm": 0.6498038905423785,
|
| 1769 |
+
"learning_rate": 8.364970796385842e-05,
|
| 1770 |
+
"loss": 1.5564023971557617,
|
| 1771 |
+
"memory(GiB)": 69.47,
|
| 1772 |
+
"step": 805,
|
| 1773 |
+
"token_acc": 0.6138023426326557,
|
| 1774 |
+
"train_speed(iter/s)": 0.068168
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.6074240719910011,
|
| 1778 |
+
"grad_norm": 0.6135051165962923,
|
| 1779 |
+
"learning_rate": 8.341963140427242e-05,
|
| 1780 |
+
"loss": 1.5415067672729492,
|
| 1781 |
+
"memory(GiB)": 69.47,
|
| 1782 |
+
"step": 810,
|
| 1783 |
+
"token_acc": 0.626180975496044,
|
| 1784 |
+
"train_speed(iter/s)": 0.068221
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 0.6111736032995876,
|
| 1788 |
+
"grad_norm": 0.6815651297208333,
|
| 1789 |
+
"learning_rate": 8.318826863251767e-05,
|
| 1790 |
+
"loss": 1.5766416549682618,
|
| 1791 |
+
"memory(GiB)": 69.47,
|
| 1792 |
+
"step": 815,
|
| 1793 |
+
"token_acc": 0.6249666577754067,
|
| 1794 |
+
"train_speed(iter/s)": 0.068273
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6149231346081739,
|
| 1798 |
+
"grad_norm": 0.572702815102001,
|
| 1799 |
+
"learning_rate": 8.295562855298954e-05,
|
| 1800 |
+
"loss": 1.553378200531006,
|
| 1801 |
+
"memory(GiB)": 69.47,
|
| 1802 |
+
"step": 820,
|
| 1803 |
+
"token_acc": 0.6252176941832114,
|
| 1804 |
+
"train_speed(iter/s)": 0.068327
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 0.6186726659167604,
|
| 1808 |
+
"grad_norm": 0.6456178217228715,
|
| 1809 |
+
"learning_rate": 8.272172011924286e-05,
|
| 1810 |
+
"loss": 1.5423843383789062,
|
| 1811 |
+
"memory(GiB)": 69.47,
|
| 1812 |
+
"step": 825,
|
| 1813 |
+
"token_acc": 0.630381553448908,
|
| 1814 |
+
"train_speed(iter/s)": 0.068375
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.6224221972253469,
|
| 1818 |
+
"grad_norm": 0.7409789314612456,
|
| 1819 |
+
"learning_rate": 8.248655233364724e-05,
|
| 1820 |
+
"loss": 1.5722437858581544,
|
| 1821 |
+
"memory(GiB)": 69.47,
|
| 1822 |
+
"step": 830,
|
| 1823 |
+
"token_acc": 0.6120017618804579,
|
| 1824 |
+
"train_speed(iter/s)": 0.068426
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.6261717285339332,
|
| 1828 |
+
"grad_norm": 0.5703111712490284,
|
| 1829 |
+
"learning_rate": 8.22501342470406e-05,
|
| 1830 |
+
"loss": 1.5867332458496093,
|
| 1831 |
+
"memory(GiB)": 69.47,
|
| 1832 |
+
"step": 835,
|
| 1833 |
+
"token_acc": 0.6055975104556897,
|
| 1834 |
+
"train_speed(iter/s)": 0.068477
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 0.6299212598425197,
|
| 1838 |
+
"grad_norm": 0.6106372850542118,
|
| 1839 |
+
"learning_rate": 8.201247495838087e-05,
|
| 1840 |
+
"loss": 1.5748239517211915,
|
| 1841 |
+
"memory(GiB)": 69.47,
|
| 1842 |
+
"step": 840,
|
| 1843 |
+
"token_acc": 0.6177306230012216,
|
| 1844 |
+
"train_speed(iter/s)": 0.068529
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.6336707911511061,
|
| 1848 |
+
"grad_norm": 0.9343402787112188,
|
| 1849 |
+
"learning_rate": 8.177358361439576e-05,
|
| 1850 |
+
"loss": 1.5393968582153321,
|
| 1851 |
+
"memory(GiB)": 69.47,
|
| 1852 |
+
"step": 845,
|
| 1853 |
+
"token_acc": 0.5842235650338834,
|
| 1854 |
+
"train_speed(iter/s)": 0.068581
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 0.6374203224596925,
|
| 1858 |
+
"grad_norm": 0.5019319495784018,
|
| 1859 |
+
"learning_rate": 8.153346940923076e-05,
|
| 1860 |
+
"loss": 1.5542526245117188,
|
| 1861 |
+
"memory(GiB)": 69.47,
|
| 1862 |
+
"step": 850,
|
| 1863 |
+
"token_acc": 0.6162684137817839,
|
| 1864 |
+
"train_speed(iter/s)": 0.068632
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6374203224596925,
|
| 1868 |
+
"eval_loss": 1.5579167604446411,
|
| 1869 |
+
"eval_runtime": 42.5324,
|
| 1870 |
+
"eval_samples_per_second": 60.754,
|
| 1871 |
+
"eval_steps_per_second": 1.27,
|
| 1872 |
+
"eval_token_acc": 0.6111814910786968,
|
| 1873 |
+
"step": 850
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.641169853768279,
|
| 1877 |
+
"grad_norm": 0.7277039515762256,
|
| 1878 |
+
"learning_rate": 8.12921415840953e-05,
|
| 1879 |
+
"loss": 1.5459892272949218,
|
| 1880 |
+
"memory(GiB)": 69.47,
|
| 1881 |
+
"step": 855,
|
| 1882 |
+
"token_acc": 0.6116581772928544,
|
| 1883 |
+
"train_speed(iter/s)": 0.068165
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 0.6449193850768654,
|
| 1887 |
+
"grad_norm": 0.8341002109732766,
|
| 1888 |
+
"learning_rate": 8.10496094269071e-05,
|
| 1889 |
+
"loss": 1.6068138122558593,
|
| 1890 |
+
"memory(GiB)": 69.47,
|
| 1891 |
+
"step": 860,
|
| 1892 |
+
"token_acc": 0.592729276834424,
|
| 1893 |
+
"train_speed(iter/s)": 0.06822
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.6486689163854518,
|
| 1897 |
+
"grad_norm": 0.6786359044090196,
|
| 1898 |
+
"learning_rate": 8.080588227193457e-05,
|
| 1899 |
+
"loss": 1.5741477012634277,
|
| 1900 |
+
"memory(GiB)": 69.47,
|
| 1901 |
+
"step": 865,
|
| 1902 |
+
"token_acc": 0.6126883277036084,
|
| 1903 |
+
"train_speed(iter/s)": 0.06827
|
| 1904 |
+
},
|
| 1905 |
+
{
|
| 1906 |
+
"epoch": 0.6524184476940382,
|
| 1907 |
+
"grad_norm": 0.5280375262470912,
|
| 1908 |
+
"learning_rate": 8.056096949943776e-05,
|
| 1909 |
+
"loss": 1.5488641738891602,
|
| 1910 |
+
"memory(GiB)": 69.47,
|
| 1911 |
+
"step": 870,
|
| 1912 |
+
"token_acc": 0.6044325273789405,
|
| 1913 |
+
"train_speed(iter/s)": 0.068322
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6561679790026247,
|
| 1917 |
+
"grad_norm": 0.8292559986953764,
|
| 1918 |
+
"learning_rate": 8.031488053530727e-05,
|
| 1919 |
+
"loss": 1.558860683441162,
|
| 1920 |
+
"memory(GiB)": 69.47,
|
| 1921 |
+
"step": 875,
|
| 1922 |
+
"token_acc": 0.6127706521458618,
|
| 1923 |
+
"train_speed(iter/s)": 0.068369
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 0.6599175103112112,
|
| 1927 |
+
"grad_norm": 0.5170318353910285,
|
| 1928 |
+
"learning_rate": 8.006762485070138e-05,
|
| 1929 |
+
"loss": 1.5863139152526855,
|
| 1930 |
+
"memory(GiB)": 69.47,
|
| 1931 |
+
"step": 880,
|
| 1932 |
+
"token_acc": 0.6334258737694805,
|
| 1933 |
+
"train_speed(iter/s)": 0.068416
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.6636670416197975,
|
| 1937 |
+
"grad_norm": 0.5479749245017502,
|
| 1938 |
+
"learning_rate": 7.981921196168169e-05,
|
| 1939 |
+
"loss": 1.5688499450683593,
|
| 1940 |
+
"memory(GiB)": 69.47,
|
| 1941 |
+
"step": 885,
|
| 1942 |
+
"token_acc": 0.5992185184929949,
|
| 1943 |
+
"train_speed(iter/s)": 0.068465
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.667416572928384,
|
| 1947 |
+
"grad_norm": 0.5133907505191853,
|
| 1948 |
+
"learning_rate": 7.956965142884677e-05,
|
| 1949 |
+
"loss": 1.5496557235717774,
|
| 1950 |
+
"memory(GiB)": 69.47,
|
| 1951 |
+
"step": 890,
|
| 1952 |
+
"token_acc": 0.5960713161894178,
|
| 1953 |
+
"train_speed(iter/s)": 0.06851
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 0.6711661042369704,
|
| 1957 |
+
"grad_norm": 0.7025461017808501,
|
| 1958 |
+
"learning_rate": 7.931895285696429e-05,
|
| 1959 |
+
"loss": 1.532984161376953,
|
| 1960 |
+
"memory(GiB)": 69.47,
|
| 1961 |
+
"step": 895,
|
| 1962 |
+
"token_acc": 0.613921736807323,
|
| 1963 |
+
"train_speed(iter/s)": 0.068558
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.6749156355455568,
|
| 1967 |
+
"grad_norm": 0.5357872891985284,
|
| 1968 |
+
"learning_rate": 7.906712589460123e-05,
|
| 1969 |
+
"loss": 1.568619728088379,
|
| 1970 |
+
"memory(GiB)": 69.47,
|
| 1971 |
+
"step": 900,
|
| 1972 |
+
"token_acc": 0.6105768574043465,
|
| 1973 |
+
"train_speed(iter/s)": 0.068606
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 0.6749156355455568,
|
| 1977 |
+
"eval_loss": 1.5493766069412231,
|
| 1978 |
+
"eval_runtime": 42.2631,
|
| 1979 |
+
"eval_samples_per_second": 61.141,
|
| 1980 |
+
"eval_steps_per_second": 1.278,
|
| 1981 |
+
"eval_token_acc": 0.6133544441777451,
|
| 1982 |
+
"step": 900
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.6786651668541432,
|
| 1986 |
+
"grad_norm": 0.6669974027083309,
|
| 1987 |
+
"learning_rate": 7.881418023375274e-05,
|
| 1988 |
+
"loss": 1.5243403434753418,
|
| 1989 |
+
"memory(GiB)": 69.47,
|
| 1990 |
+
"step": 905,
|
| 1991 |
+
"token_acc": 0.6110998751901755,
|
| 1992 |
+
"train_speed(iter/s)": 0.068168
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.6824146981627297,
|
| 1996 |
+
"grad_norm": 0.5782224595436849,
|
| 1997 |
+
"learning_rate": 7.85601256094689e-05,
|
| 1998 |
+
"loss": 1.5333332061767577,
|
| 1999 |
+
"memory(GiB)": 69.47,
|
| 2000 |
+
"step": 910,
|
| 2001 |
+
"token_acc": 0.6019987331972694,
|
| 2002 |
+
"train_speed(iter/s)": 0.068215
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 0.6861642294713161,
|
| 2006 |
+
"grad_norm": 0.501011565104305,
|
| 2007 |
+
"learning_rate": 7.830497179948024e-05,
|
| 2008 |
+
"loss": 1.53333740234375,
|
| 2009 |
+
"memory(GiB)": 69.47,
|
| 2010 |
+
"step": 915,
|
| 2011 |
+
"token_acc": 0.6425022200831091,
|
| 2012 |
+
"train_speed(iter/s)": 0.068262
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.6899137607799025,
|
| 2016 |
+
"grad_norm": 0.6099089681168807,
|
| 2017 |
+
"learning_rate": 7.804872862382131e-05,
|
| 2018 |
+
"loss": 1.5432639122009277,
|
| 2019 |
+
"memory(GiB)": 69.47,
|
| 2020 |
+
"step": 920,
|
| 2021 |
+
"token_acc": 0.6193367306314197,
|
| 2022 |
+
"train_speed(iter/s)": 0.068312
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 0.693663292088489,
|
| 2026 |
+
"grad_norm": 0.5502623584384638,
|
| 2027 |
+
"learning_rate": 7.779140594445276e-05,
|
| 2028 |
+
"loss": 1.5435014724731446,
|
| 2029 |
+
"memory(GiB)": 69.47,
|
| 2030 |
+
"step": 925,
|
| 2031 |
+
"token_acc": 0.5937242005144545,
|
| 2032 |
+
"train_speed(iter/s)": 0.068359
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6974128233970753,
|
| 2036 |
+
"grad_norm": 0.9566069013819732,
|
| 2037 |
+
"learning_rate": 7.753301366488187e-05,
|
| 2038 |
+
"loss": 1.557739543914795,
|
| 2039 |
+
"memory(GiB)": 69.47,
|
| 2040 |
+
"step": 930,
|
| 2041 |
+
"token_acc": 0.5973987625581124,
|
| 2042 |
+
"train_speed(iter/s)": 0.068405
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 0.7011623547056618,
|
| 2046 |
+
"grad_norm": 0.5360843012267709,
|
| 2047 |
+
"learning_rate": 7.727356172978126e-05,
|
| 2048 |
+
"loss": 1.5812086105346679,
|
| 2049 |
+
"memory(GiB)": 69.47,
|
| 2050 |
+
"step": 935,
|
| 2051 |
+
"token_acc": 0.6558960928652322,
|
| 2052 |
+
"train_speed(iter/s)": 0.068448
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.7049118860142483,
|
| 2056 |
+
"grad_norm": 0.5966513399888408,
|
| 2057 |
+
"learning_rate": 7.701306012460628e-05,
|
| 2058 |
+
"loss": 1.5495784759521485,
|
| 2059 |
+
"memory(GiB)": 69.47,
|
| 2060 |
+
"step": 940,
|
| 2061 |
+
"token_acc": 0.6080101474135268,
|
| 2062 |
+
"train_speed(iter/s)": 0.068494
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.7086614173228346,
|
| 2066 |
+
"grad_norm": 0.6021147871835324,
|
| 2067 |
+
"learning_rate": 7.675151887521055e-05,
|
| 2068 |
+
"loss": 1.5489521026611328,
|
| 2069 |
+
"memory(GiB)": 69.47,
|
| 2070 |
+
"step": 945,
|
| 2071 |
+
"token_acc": 0.6134959283054384,
|
| 2072 |
+
"train_speed(iter/s)": 0.068536
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 0.7124109486314211,
|
| 2076 |
+
"grad_norm": 0.518627623432029,
|
| 2077 |
+
"learning_rate": 7.648894804746031e-05,
|
| 2078 |
+
"loss": 1.532328987121582,
|
| 2079 |
+
"memory(GiB)": 69.47,
|
| 2080 |
+
"step": 950,
|
| 2081 |
+
"token_acc": 0.6318991712222541,
|
| 2082 |
+
"train_speed(iter/s)": 0.068579
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.7124109486314211,
|
| 2086 |
+
"eval_loss": 1.5434088706970215,
|
| 2087 |
+
"eval_runtime": 42.2906,
|
| 2088 |
+
"eval_samples_per_second": 61.101,
|
| 2089 |
+
"eval_steps_per_second": 1.277,
|
| 2090 |
+
"eval_token_acc": 0.6149505506887293,
|
| 2091 |
+
"step": 950
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 0.7161604799400075,
|
| 2095 |
+
"grad_norm": 0.8914870632575633,
|
| 2096 |
+
"learning_rate": 7.62253577468468e-05,
|
| 2097 |
+
"loss": 1.5970706939697266,
|
| 2098 |
+
"memory(GiB)": 69.47,
|
| 2099 |
+
"step": 955,
|
| 2100 |
+
"token_acc": 0.6109180049315731,
|
| 2101 |
+
"train_speed(iter/s)": 0.068152
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.7199100112485939,
|
| 2105 |
+
"grad_norm": 0.690063303670817,
|
| 2106 |
+
"learning_rate": 7.596075811809753e-05,
|
| 2107 |
+
"loss": 1.549104881286621,
|
| 2108 |
+
"memory(GiB)": 69.47,
|
| 2109 |
+
"step": 960,
|
| 2110 |
+
"token_acc": 0.6511777992900936,
|
| 2111 |
+
"train_speed(iter/s)": 0.068198
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.7236595425571803,
|
| 2115 |
+
"grad_norm": 0.6659235860977825,
|
| 2116 |
+
"learning_rate": 7.569515934478564e-05,
|
| 2117 |
+
"loss": 1.5497787475585938,
|
| 2118 |
+
"memory(GiB)": 69.47,
|
| 2119 |
+
"step": 965,
|
| 2120 |
+
"token_acc": 0.6138149161542004,
|
| 2121 |
+
"train_speed(iter/s)": 0.068244
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 0.7274090738657668,
|
| 2125 |
+
"grad_norm": 0.6033721066219084,
|
| 2126 |
+
"learning_rate": 7.542857164893816e-05,
|
| 2127 |
+
"loss": 1.527195644378662,
|
| 2128 |
+
"memory(GiB)": 69.47,
|
| 2129 |
+
"step": 970,
|
| 2130 |
+
"token_acc": 0.6292321961721516,
|
| 2131 |
+
"train_speed(iter/s)": 0.06829
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.7311586051743532,
|
| 2135 |
+
"grad_norm": 0.6454995098716503,
|
| 2136 |
+
"learning_rate": 7.51610052906424e-05,
|
| 2137 |
+
"loss": 1.5415388107299806,
|
| 2138 |
+
"memory(GiB)": 69.47,
|
| 2139 |
+
"step": 975,
|
| 2140 |
+
"token_acc": 0.5943276411877579,
|
| 2141 |
+
"train_speed(iter/s)": 0.068334
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 0.7349081364829396,
|
| 2145 |
+
"grad_norm": 0.5803433939439756,
|
| 2146 |
+
"learning_rate": 7.489247056765135e-05,
|
| 2147 |
+
"loss": 1.5216063499450683,
|
| 2148 |
+
"memory(GiB)": 69.47,
|
| 2149 |
+
"step": 980,
|
| 2150 |
+
"token_acc": 0.6121325935038082,
|
| 2151 |
+
"train_speed(iter/s)": 0.068376
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7386576677915261,
|
| 2155 |
+
"grad_norm": 0.5566197748774061,
|
| 2156 |
+
"learning_rate": 7.462297781498701e-05,
|
| 2157 |
+
"loss": 1.532740306854248,
|
| 2158 |
+
"memory(GiB)": 69.47,
|
| 2159 |
+
"step": 985,
|
| 2160 |
+
"token_acc": 0.6146595230537002,
|
| 2161 |
+
"train_speed(iter/s)": 0.068419
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 0.7424071991001124,
|
| 2165 |
+
"grad_norm": 0.6227540700751105,
|
| 2166 |
+
"learning_rate": 7.435253740454293e-05,
|
| 2167 |
+
"loss": 1.5283782005310058,
|
| 2168 |
+
"memory(GiB)": 69.47,
|
| 2169 |
+
"step": 990,
|
| 2170 |
+
"token_acc": 0.6046981841976117,
|
| 2171 |
+
"train_speed(iter/s)": 0.068463
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.7461567304086989,
|
| 2175 |
+
"grad_norm": 0.4833210780346364,
|
| 2176 |
+
"learning_rate": 7.408115974468483e-05,
|
| 2177 |
+
"loss": 1.5609919548034668,
|
| 2178 |
+
"memory(GiB)": 69.47,
|
| 2179 |
+
"step": 995,
|
| 2180 |
+
"token_acc": 0.6033836124597676,
|
| 2181 |
+
"train_speed(iter/s)": 0.068502
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 0.7499062617172854,
|
| 2185 |
+
"grad_norm": 2.343474195794338,
|
| 2186 |
+
"learning_rate": 7.380885527985015e-05,
|
| 2187 |
+
"loss": 1.559861660003662,
|
| 2188 |
+
"memory(GiB)": 69.47,
|
| 2189 |
+
"step": 1000,
|
| 2190 |
+
"token_acc": 0.6262118076357497,
|
| 2191 |
+
"train_speed(iter/s)": 0.068543
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 0.7499062617172854,
|
| 2195 |
+
"eval_loss": 1.5367435216903687,
|
| 2196 |
+
"eval_runtime": 43.3247,
|
| 2197 |
+
"eval_samples_per_second": 59.643,
|
| 2198 |
+
"eval_steps_per_second": 1.246,
|
| 2199 |
+
"eval_token_acc": 0.6163190015111681,
|
| 2200 |
+
"step": 1000
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7536557930258717,
|
| 2204 |
+
"grad_norm": 0.8734104988651366,
|
| 2205 |
+
"learning_rate": 7.353563449014596e-05,
|
| 2206 |
+
"loss": 1.5373735427856445,
|
| 2207 |
+
"memory(GiB)": 69.47,
|
| 2208 |
+
"step": 1005,
|
| 2209 |
+
"token_acc": 0.6212234254333892,
|
| 2210 |
+
"train_speed(iter/s)": 0.068149
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 0.7574053243344582,
|
| 2214 |
+
"grad_norm": 0.8087734921502168,
|
| 2215 |
+
"learning_rate": 7.32615078909457e-05,
|
| 2216 |
+
"loss": 1.5338141441345214,
|
| 2217 |
+
"memory(GiB)": 69.47,
|
| 2218 |
+
"step": 1010,
|
| 2219 |
+
"token_acc": 0.6215583599168054,
|
| 2220 |
+
"train_speed(iter/s)": 0.068193
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 0.7611548556430446,
|
| 2224 |
+
"grad_norm": 0.5858522680768657,
|
| 2225 |
+
"learning_rate": 7.298648603248448e-05,
|
| 2226 |
+
"loss": 1.523855972290039,
|
| 2227 |
+
"memory(GiB)": 69.47,
|
| 2228 |
+
"step": 1015,
|
| 2229 |
+
"token_acc": 0.5851850820847223,
|
| 2230 |
+
"train_speed(iter/s)": 0.068234
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.7649043869516311,
|
| 2234 |
+
"grad_norm": 0.43609230134938953,
|
| 2235 |
+
"learning_rate": 7.271057949945297e-05,
|
| 2236 |
+
"loss": 1.5160707473754882,
|
| 2237 |
+
"memory(GiB)": 69.47,
|
| 2238 |
+
"step": 1020,
|
| 2239 |
+
"token_acc": 0.5845537590381331,
|
| 2240 |
+
"train_speed(iter/s)": 0.068275
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 0.7686539182602175,
|
| 2244 |
+
"grad_norm": 0.660274468996072,
|
| 2245 |
+
"learning_rate": 7.243379891059006e-05,
|
| 2246 |
+
"loss": 1.5410688400268555,
|
| 2247 |
+
"memory(GiB)": 69.47,
|
| 2248 |
+
"step": 1025,
|
| 2249 |
+
"token_acc": 0.5800014434180139,
|
| 2250 |
+
"train_speed(iter/s)": 0.068315
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.7724034495688039,
|
| 2254 |
+
"grad_norm": 0.6306033236366111,
|
| 2255 |
+
"learning_rate": 7.215615491827419e-05,
|
| 2256 |
+
"loss": 1.5749454498291016,
|
| 2257 |
+
"memory(GiB)": 69.47,
|
| 2258 |
+
"step": 1030,
|
| 2259 |
+
"token_acc": 0.5871338912133891,
|
| 2260 |
+
"train_speed(iter/s)": 0.068358
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 0.7761529808773904,
|
| 2264 |
+
"grad_norm": 0.6445997254156263,
|
| 2265 |
+
"learning_rate": 7.187765820811342e-05,
|
| 2266 |
+
"loss": 1.5300399780273437,
|
| 2267 |
+
"memory(GiB)": 69.47,
|
| 2268 |
+
"step": 1035,
|
| 2269 |
+
"token_acc": 0.6455525506612826,
|
| 2270 |
+
"train_speed(iter/s)": 0.0684
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7799025121859767,
|
| 2274 |
+
"grad_norm": 0.5007502340595024,
|
| 2275 |
+
"learning_rate": 7.159831949853409e-05,
|
| 2276 |
+
"loss": 1.5059438705444337,
|
| 2277 |
+
"memory(GiB)": 69.47,
|
| 2278 |
+
"step": 1040,
|
| 2279 |
+
"token_acc": 0.5968595696953947,
|
| 2280 |
+
"train_speed(iter/s)": 0.068441
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 0.7836520434945632,
|
| 2284 |
+
"grad_norm": 0.7039775402140321,
|
| 2285 |
+
"learning_rate": 7.131814954036831e-05,
|
| 2286 |
+
"loss": 1.5242332458496093,
|
| 2287 |
+
"memory(GiB)": 69.47,
|
| 2288 |
+
"step": 1045,
|
| 2289 |
+
"token_acc": 0.6320911041425978,
|
| 2290 |
+
"train_speed(iter/s)": 0.068482
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 0.7874015748031497,
|
| 2294 |
+
"grad_norm": 0.5105682948416441,
|
| 2295 |
+
"learning_rate": 7.103715911644028e-05,
|
| 2296 |
+
"loss": 1.5528465270996095,
|
| 2297 |
+
"memory(GiB)": 69.47,
|
| 2298 |
+
"step": 1050,
|
| 2299 |
+
"token_acc": 0.5929656160458453,
|
| 2300 |
+
"train_speed(iter/s)": 0.068521
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.7874015748031497,
|
| 2304 |
+
"eval_loss": 1.5263327360153198,
|
| 2305 |
+
"eval_runtime": 42.5082,
|
| 2306 |
+
"eval_samples_per_second": 60.788,
|
| 2307 |
+
"eval_steps_per_second": 1.27,
|
| 2308 |
+
"eval_token_acc": 0.6192035490378847,
|
| 2309 |
+
"step": 1050
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 0.791151106111736,
|
| 2313 |
+
"grad_norm": 0.4313496062042051,
|
| 2314 |
+
"learning_rate": 7.075535904115123e-05,
|
| 2315 |
+
"loss": 1.5179916381835938,
|
| 2316 |
+
"memory(GiB)": 69.47,
|
| 2317 |
+
"step": 1055,
|
| 2318 |
+
"token_acc": 0.6148530892217319,
|
| 2319 |
+
"train_speed(iter/s)": 0.06811
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7949006374203225,
|
| 2323 |
+
"grad_norm": 0.5479905912215369,
|
| 2324 |
+
"learning_rate": 7.047276016006318e-05,
|
| 2325 |
+
"loss": 1.5539355278015137,
|
| 2326 |
+
"memory(GiB)": 69.47,
|
| 2327 |
+
"step": 1060,
|
| 2328 |
+
"token_acc": 0.6127638959018519,
|
| 2329 |
+
"train_speed(iter/s)": 0.068151
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 0.7986501687289089,
|
| 2333 |
+
"grad_norm": 0.49760121615852204,
|
| 2334 |
+
"learning_rate": 7.018937334948162e-05,
|
| 2335 |
+
"loss": 1.5182838439941406,
|
| 2336 |
+
"memory(GiB)": 69.47,
|
| 2337 |
+
"step": 1065,
|
| 2338 |
+
"token_acc": 0.6090527574084666,
|
| 2339 |
+
"train_speed(iter/s)": 0.068191
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 0.8023997000374953,
|
| 2343 |
+
"grad_norm": 0.5415301294034648,
|
| 2344 |
+
"learning_rate": 6.990520951603682e-05,
|
| 2345 |
+
"loss": 1.5360933303833009,
|
| 2346 |
+
"memory(GiB)": 69.47,
|
| 2347 |
+
"step": 1070,
|
| 2348 |
+
"token_acc": 0.6196442525462564,
|
| 2349 |
+
"train_speed(iter/s)": 0.068231
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 0.8061492313460817,
|
| 2353 |
+
"grad_norm": 0.4208428056833552,
|
| 2354 |
+
"learning_rate": 6.962027959626415e-05,
|
| 2355 |
+
"loss": 1.5064073562622071,
|
| 2356 |
+
"memory(GiB)": 69.47,
|
| 2357 |
+
"step": 1075,
|
| 2358 |
+
"token_acc": 0.6160373269767316,
|
| 2359 |
+
"train_speed(iter/s)": 0.068273
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 0.8098987626546682,
|
| 2363 |
+
"grad_norm": 0.6586931742476955,
|
| 2364 |
+
"learning_rate": 6.933459455618312e-05,
|
| 2365 |
+
"loss": 1.4846086502075195,
|
| 2366 |
+
"memory(GiB)": 69.47,
|
| 2367 |
+
"step": 1080,
|
| 2368 |
+
"token_acc": 0.6252297918356312,
|
| 2369 |
+
"train_speed(iter/s)": 0.068316
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.8136482939632546,
|
| 2373 |
+
"grad_norm": 0.45819309154158716,
|
| 2374 |
+
"learning_rate": 6.904816539087535e-05,
|
| 2375 |
+
"loss": 1.5363639831542968,
|
| 2376 |
+
"memory(GiB)": 69.47,
|
| 2377 |
+
"step": 1085,
|
| 2378 |
+
"token_acc": 0.5925773337453222,
|
| 2379 |
+
"train_speed(iter/s)": 0.068356
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 0.817397825271841,
|
| 2383 |
+
"grad_norm": 0.4768245890924099,
|
| 2384 |
+
"learning_rate": 6.876100312406141e-05,
|
| 2385 |
+
"loss": 1.4858908653259277,
|
| 2386 |
+
"memory(GiB)": 69.47,
|
| 2387 |
+
"step": 1090,
|
| 2388 |
+
"token_acc": 0.6263631660631144,
|
| 2389 |
+
"train_speed(iter/s)": 0.068396
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8211473565804275,
|
| 2393 |
+
"grad_norm": 0.4287096498191579,
|
| 2394 |
+
"learning_rate": 6.847311880767653e-05,
|
| 2395 |
+
"loss": 1.530019187927246,
|
| 2396 |
+
"memory(GiB)": 69.47,
|
| 2397 |
+
"step": 1095,
|
| 2398 |
+
"token_acc": 0.6366432004192207,
|
| 2399 |
+
"train_speed(iter/s)": 0.068437
|
| 2400 |
+
},
|
| 2401 |
+
{
|
| 2402 |
+
"epoch": 0.8248968878890138,
|
| 2403 |
+
"grad_norm": 0.7751814748031907,
|
| 2404 |
+
"learning_rate": 6.818452352144527e-05,
|
| 2405 |
+
"loss": 1.5163839340209961,
|
| 2406 |
+
"memory(GiB)": 69.47,
|
| 2407 |
+
"step": 1100,
|
| 2408 |
+
"token_acc": 0.6419833221838799,
|
| 2409 |
+
"train_speed(iter/s)": 0.068476
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 0.8248968878890138,
|
| 2413 |
+
"eval_loss": 1.520379662513733,
|
| 2414 |
+
"eval_runtime": 42.5564,
|
| 2415 |
+
"eval_samples_per_second": 60.719,
|
| 2416 |
+
"eval_steps_per_second": 1.269,
|
| 2417 |
+
"eval_token_acc": 0.6207940978558735,
|
| 2418 |
+
"step": 1100
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.8286464191976003,
|
| 2422 |
+
"grad_norm": 0.8005900073361203,
|
| 2423 |
+
"learning_rate": 6.789522837245509e-05,
|
| 2424 |
+
"loss": 1.564078426361084,
|
| 2425 |
+
"memory(GiB)": 69.47,
|
| 2426 |
+
"step": 1105,
|
| 2427 |
+
"token_acc": 0.6168006265048804,
|
| 2428 |
+
"train_speed(iter/s)": 0.068129
|
| 2429 |
+
},
|
| 2430 |
+
{
|
| 2431 |
+
"epoch": 0.8323959505061868,
|
| 2432 |
+
"grad_norm": 0.4437111510930812,
|
| 2433 |
+
"learning_rate": 6.760524449472889e-05,
|
| 2434 |
+
"loss": 1.5362157821655273,
|
| 2435 |
+
"memory(GiB)": 69.47,
|
| 2436 |
+
"step": 1110,
|
| 2437 |
+
"token_acc": 0.5717586525595575,
|
| 2438 |
+
"train_speed(iter/s)": 0.068169
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8361454818147731,
|
| 2442 |
+
"grad_norm": 0.5530244589417227,
|
| 2443 |
+
"learning_rate": 6.731458304879648e-05,
|
| 2444 |
+
"loss": 1.5107562065124511,
|
| 2445 |
+
"memory(GiB)": 69.47,
|
| 2446 |
+
"step": 1115,
|
| 2447 |
+
"token_acc": 0.5941068640097766,
|
| 2448 |
+
"train_speed(iter/s)": 0.068208
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 0.8398950131233596,
|
| 2452 |
+
"grad_norm": 0.49930017760849993,
|
| 2453 |
+
"learning_rate": 6.702325522126502e-05,
|
| 2454 |
+
"loss": 1.5095773696899415,
|
| 2455 |
+
"memory(GiB)": 69.47,
|
| 2456 |
+
"step": 1120,
|
| 2457 |
+
"token_acc": 0.6088115657218822,
|
| 2458 |
+
"train_speed(iter/s)": 0.068248
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 0.843644544431946,
|
| 2462 |
+
"grad_norm": 0.4469462680373623,
|
| 2463 |
+
"learning_rate": 6.673127222438856e-05,
|
| 2464 |
+
"loss": 1.4989742279052733,
|
| 2465 |
+
"memory(GiB)": 69.47,
|
| 2466 |
+
"step": 1125,
|
| 2467 |
+
"token_acc": 0.6028838351560761,
|
| 2468 |
+
"train_speed(iter/s)": 0.068286
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 0.8473940757405324,
|
| 2472 |
+
"grad_norm": 0.4460794266475831,
|
| 2473 |
+
"learning_rate": 6.643864529563644e-05,
|
| 2474 |
+
"loss": 1.496596336364746,
|
| 2475 |
+
"memory(GiB)": 69.47,
|
| 2476 |
+
"step": 1130,
|
| 2477 |
+
"token_acc": 0.5999792157406124,
|
| 2478 |
+
"train_speed(iter/s)": 0.068322
|
| 2479 |
+
},
|
| 2480 |
+
{
|
| 2481 |
+
"epoch": 0.8511436070491188,
|
| 2482 |
+
"grad_norm": 0.9287526197832127,
|
| 2483 |
+
"learning_rate": 6.614538569726086e-05,
|
| 2484 |
+
"loss": 1.5509444236755372,
|
| 2485 |
+
"memory(GiB)": 69.47,
|
| 2486 |
+
"step": 1135,
|
| 2487 |
+
"token_acc": 0.584876554545922,
|
| 2488 |
+
"train_speed(iter/s)": 0.068358
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.8548931383577053,
|
| 2492 |
+
"grad_norm": 0.5551548580097715,
|
| 2493 |
+
"learning_rate": 6.585150471586334e-05,
|
| 2494 |
+
"loss": 1.560178279876709,
|
| 2495 |
+
"memory(GiB)": 69.47,
|
| 2496 |
+
"step": 1140,
|
| 2497 |
+
"token_acc": 0.616736937464666,
|
| 2498 |
+
"train_speed(iter/s)": 0.068396
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 0.8586426696662918,
|
| 2502 |
+
"grad_norm": 0.6024443303940071,
|
| 2503 |
+
"learning_rate": 6.555701366196049e-05,
|
| 2504 |
+
"loss": 1.525533103942871,
|
| 2505 |
+
"memory(GiB)": 69.47,
|
| 2506 |
+
"step": 1145,
|
| 2507 |
+
"token_acc": 0.6429341444825147,
|
| 2508 |
+
"train_speed(iter/s)": 0.068431
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8623922009748781,
|
| 2512 |
+
"grad_norm": 0.5584796840321534,
|
| 2513 |
+
"learning_rate": 6.526192386954853e-05,
|
| 2514 |
+
"loss": 1.51143798828125,
|
| 2515 |
+
"memory(GiB)": 69.47,
|
| 2516 |
+
"step": 1150,
|
| 2517 |
+
"token_acc": 0.6453966470616104,
|
| 2518 |
+
"train_speed(iter/s)": 0.068467
|
| 2519 |
+
},
|
| 2520 |
+
{
|
| 2521 |
+
"epoch": 0.8623922009748781,
|
| 2522 |
+
"eval_loss": 1.513854742050171,
|
| 2523 |
+
"eval_runtime": 42.6717,
|
| 2524 |
+
"eval_samples_per_second": 60.555,
|
| 2525 |
+
"eval_steps_per_second": 1.265,
|
| 2526 |
+
"eval_token_acc": 0.6222880686503016,
|
| 2527 |
+
"step": 1150
|
| 2528 |
+
},
|
| 2529 |
+
{
|
| 2530 |
+
"epoch": 0.8661417322834646,
|
| 2531 |
+
"grad_norm": 0.5912782454437087,
|
| 2532 |
+
"learning_rate": 6.49662466956672e-05,
|
| 2533 |
+
"loss": 1.519371223449707,
|
| 2534 |
+
"memory(GiB)": 69.47,
|
| 2535 |
+
"step": 1155,
|
| 2536 |
+
"token_acc": 0.6188296676949063,
|
| 2537 |
+
"train_speed(iter/s)": 0.06813
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.869891263592051,
|
| 2541 |
+
"grad_norm": 0.6323258494130859,
|
| 2542 |
+
"learning_rate": 6.466999351996266e-05,
|
| 2543 |
+
"loss": 1.5339200973510743,
|
| 2544 |
+
"memory(GiB)": 69.47,
|
| 2545 |
+
"step": 1160,
|
| 2546 |
+
"token_acc": 0.6172287027696723,
|
| 2547 |
+
"train_speed(iter/s)": 0.068169
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 0.8736407949006374,
|
| 2551 |
+
"grad_norm": 0.8076432386359942,
|
| 2552 |
+
"learning_rate": 6.437317574424942e-05,
|
| 2553 |
+
"loss": 1.5256396293640138,
|
| 2554 |
+
"memory(GiB)": 69.47,
|
| 2555 |
+
"step": 1165,
|
| 2556 |
+
"token_acc": 0.6100851645177295,
|
| 2557 |
+
"train_speed(iter/s)": 0.068209
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8773903262092239,
|
| 2561 |
+
"grad_norm": 0.7353169163252865,
|
| 2562 |
+
"learning_rate": 6.407580479207166e-05,
|
| 2563 |
+
"loss": 1.4921053886413573,
|
| 2564 |
+
"memory(GiB)": 69.47,
|
| 2565 |
+
"step": 1170,
|
| 2566 |
+
"token_acc": 0.6401115189158667,
|
| 2567 |
+
"train_speed(iter/s)": 0.068248
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 0.8811398575178103,
|
| 2571 |
+
"grad_norm": 0.8585271026745834,
|
| 2572 |
+
"learning_rate": 6.377789210826348e-05,
|
| 2573 |
+
"loss": 1.506899070739746,
|
| 2574 |
+
"memory(GiB)": 69.47,
|
| 2575 |
+
"step": 1175,
|
| 2576 |
+
"token_acc": 0.6469974922313689,
|
| 2577 |
+
"train_speed(iter/s)": 0.068285
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 0.8848893888263967,
|
| 2581 |
+
"grad_norm": 0.7332150477460917,
|
| 2582 |
+
"learning_rate": 6.347944915850845e-05,
|
| 2583 |
+
"loss": 1.5066943168640137,
|
| 2584 |
+
"memory(GiB)": 69.47,
|
| 2585 |
+
"step": 1180,
|
| 2586 |
+
"token_acc": 0.6266228771329003,
|
| 2587 |
+
"train_speed(iter/s)": 0.06832
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 0.8886389201349831,
|
| 2591 |
+
"grad_norm": 0.5470041289498336,
|
| 2592 |
+
"learning_rate": 6.318048742889837e-05,
|
| 2593 |
+
"loss": 1.513447093963623,
|
| 2594 |
+
"memory(GiB)": 69.47,
|
| 2595 |
+
"step": 1185,
|
| 2596 |
+
"token_acc": 0.6192370090975678,
|
| 2597 |
+
"train_speed(iter/s)": 0.068356
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 0.8923884514435696,
|
| 2601 |
+
"grad_norm": 0.5304051415658053,
|
| 2602 |
+
"learning_rate": 6.288101842549116e-05,
|
| 2603 |
+
"loss": 1.5381876945495605,
|
| 2604 |
+
"memory(GiB)": 69.47,
|
| 2605 |
+
"step": 1190,
|
| 2606 |
+
"token_acc": 0.6580005979723302,
|
| 2607 |
+
"train_speed(iter/s)": 0.068393
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.896137982752156,
|
| 2611 |
+
"grad_norm": 0.5560911716174746,
|
| 2612 |
+
"learning_rate": 6.258105367386805e-05,
|
| 2613 |
+
"loss": 1.5324213981628418,
|
| 2614 |
+
"memory(GiB)": 69.47,
|
| 2615 |
+
"step": 1195,
|
| 2616 |
+
"token_acc": 0.6150402163855079,
|
| 2617 |
+
"train_speed(iter/s)": 0.068427
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 0.8998875140607424,
|
| 2621 |
+
"grad_norm": 0.5509160992216331,
|
| 2622 |
+
"learning_rate": 6.228060471868998e-05,
|
| 2623 |
+
"loss": 1.4963160514831544,
|
| 2624 |
+
"memory(GiB)": 69.47,
|
| 2625 |
+
"step": 1200,
|
| 2626 |
+
"token_acc": 0.6280981700261424,
|
| 2627 |
+
"train_speed(iter/s)": 0.068461
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8998875140607424,
|
| 2631 |
+
"eval_loss": 1.504333257675171,
|
| 2632 |
+
"eval_runtime": 42.9534,
|
| 2633 |
+
"eval_samples_per_second": 60.158,
|
| 2634 |
+
"eval_steps_per_second": 1.257,
|
| 2635 |
+
"eval_token_acc": 0.6246635105262182,
|
| 2636 |
+
"step": 1200
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 0.9036370453693289,
|
| 2640 |
+
"grad_norm": 0.5106735516463429,
|
| 2641 |
+
"learning_rate": 6.197968312325331e-05,
|
| 2642 |
+
"loss": 1.5228327751159667,
|
| 2643 |
+
"memory(GiB)": 69.47,
|
| 2644 |
+
"step": 1205,
|
| 2645 |
+
"token_acc": 0.6340643951903793,
|
| 2646 |
+
"train_speed(iter/s)": 0.068131
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 0.9073865766779152,
|
| 2650 |
+
"grad_norm": 0.5082226063892653,
|
| 2651 |
+
"learning_rate": 6.16783004690448e-05,
|
| 2652 |
+
"loss": 1.486492919921875,
|
| 2653 |
+
"memory(GiB)": 69.47,
|
| 2654 |
+
"step": 1210,
|
| 2655 |
+
"token_acc": 0.6153356827521278,
|
| 2656 |
+
"train_speed(iter/s)": 0.068168
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.9111361079865017,
|
| 2660 |
+
"grad_norm": 0.519704267787884,
|
| 2661 |
+
"learning_rate": 6.13764683552958e-05,
|
| 2662 |
+
"loss": 1.4750579833984374,
|
| 2663 |
+
"memory(GiB)": 69.47,
|
| 2664 |
+
"step": 1215,
|
| 2665 |
+
"token_acc": 0.6144492828225212,
|
| 2666 |
+
"train_speed(iter/s)": 0.068202
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 0.9148856392950881,
|
| 2670 |
+
"grad_norm": 0.45915590635251785,
|
| 2671 |
+
"learning_rate": 6.107419839853597e-05,
|
| 2672 |
+
"loss": 1.5226827621459962,
|
| 2673 |
+
"memory(GiB)": 69.47,
|
| 2674 |
+
"step": 1220,
|
| 2675 |
+
"token_acc": 0.6172223102192043,
|
| 2676 |
+
"train_speed(iter/s)": 0.068239
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9186351706036745,
|
| 2680 |
+
"grad_norm": 0.6022516869442112,
|
| 2681 |
+
"learning_rate": 6.0771502232145986e-05,
|
| 2682 |
+
"loss": 1.477038860321045,
|
| 2683 |
+
"memory(GiB)": 69.47,
|
| 2684 |
+
"step": 1225,
|
| 2685 |
+
"token_acc": 0.6374207480827265,
|
| 2686 |
+
"train_speed(iter/s)": 0.068276
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 0.922384701912261,
|
| 2690 |
+
"grad_norm": 0.5112193945587918,
|
| 2691 |
+
"learning_rate": 6.046839150591006e-05,
|
| 2692 |
+
"loss": 1.5150461196899414,
|
| 2693 |
+
"memory(GiB)": 69.47,
|
| 2694 |
+
"step": 1230,
|
| 2695 |
+
"token_acc": 0.5920762137335217,
|
| 2696 |
+
"train_speed(iter/s)": 0.068312
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 0.9261342332208474,
|
| 2700 |
+
"grad_norm": 0.6035285878286432,
|
| 2701 |
+
"learning_rate": 6.016487788556737e-05,
|
| 2702 |
+
"loss": 1.481464672088623,
|
| 2703 |
+
"memory(GiB)": 69.47,
|
| 2704 |
+
"step": 1235,
|
| 2705 |
+
"token_acc": 0.6499792111512794,
|
| 2706 |
+
"train_speed(iter/s)": 0.068347
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 0.9298837645294338,
|
| 2710 |
+
"grad_norm": 0.7514308474748557,
|
| 2711 |
+
"learning_rate": 5.9860973052363266e-05,
|
| 2712 |
+
"loss": 1.4752119064331055,
|
| 2713 |
+
"memory(GiB)": 69.47,
|
| 2714 |
+
"step": 1240,
|
| 2715 |
+
"token_acc": 0.6230399209428725,
|
| 2716 |
+
"train_speed(iter/s)": 0.068382
|
| 2717 |
+
},
|
| 2718 |
+
{
|
| 2719 |
+
"epoch": 0.9336332958380202,
|
| 2720 |
+
"grad_norm": 0.6413337005680765,
|
| 2721 |
+
"learning_rate": 5.955668870259946e-05,
|
| 2722 |
+
"loss": 1.4961332321166991,
|
| 2723 |
+
"memory(GiB)": 69.47,
|
| 2724 |
+
"step": 1245,
|
| 2725 |
+
"token_acc": 0.6427623211170498,
|
| 2726 |
+
"train_speed(iter/s)": 0.068417
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.9373828271466067,
|
| 2730 |
+
"grad_norm": 0.6910314897526542,
|
| 2731 |
+
"learning_rate": 5.925203654718415e-05,
|
| 2732 |
+
"loss": 1.4927474975585937,
|
| 2733 |
+
"memory(GiB)": 69.47,
|
| 2734 |
+
"step": 1250,
|
| 2735 |
+
"token_acc": 0.6272580753475253,
|
| 2736 |
+
"train_speed(iter/s)": 0.068451
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 0.9373828271466067,
|
| 2740 |
+
"eval_loss": 1.497246503829956,
|
| 2741 |
+
"eval_runtime": 42.3687,
|
| 2742 |
+
"eval_samples_per_second": 60.988,
|
| 2743 |
+
"eval_steps_per_second": 1.275,
|
| 2744 |
+
"eval_token_acc": 0.6262709421474569,
|
| 2745 |
+
"step": 1250
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 0.941132358455193,
|
| 2749 |
+
"grad_norm": 0.517334708850761,
|
| 2750 |
+
"learning_rate": 5.894702831118109e-05,
|
| 2751 |
+
"loss": 1.4793219566345215,
|
| 2752 |
+
"memory(GiB)": 69.47,
|
| 2753 |
+
"step": 1255,
|
| 2754 |
+
"token_acc": 0.6311689610763095,
|
| 2755 |
+
"train_speed(iter/s)": 0.068142
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 0.9448818897637795,
|
| 2759 |
+
"grad_norm": 0.4491902826842717,
|
| 2760 |
+
"learning_rate": 5.864167573335841e-05,
|
| 2761 |
+
"loss": 1.4944025993347168,
|
| 2762 |
+
"memory(GiB)": 69.47,
|
| 2763 |
+
"step": 1260,
|
| 2764 |
+
"token_acc": 0.5857831760276445,
|
| 2765 |
+
"train_speed(iter/s)": 0.068177
|
| 2766 |
+
},
|
| 2767 |
+
{
|
| 2768 |
+
"epoch": 0.948631421072366,
|
| 2769 |
+
"grad_norm": 0.7690790493678104,
|
| 2770 |
+
"learning_rate": 5.8335990565736895e-05,
|
| 2771 |
+
"loss": 1.4978620529174804,
|
| 2772 |
+
"memory(GiB)": 69.47,
|
| 2773 |
+
"step": 1265,
|
| 2774 |
+
"token_acc": 0.6538104909925821,
|
| 2775 |
+
"train_speed(iter/s)": 0.068214
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.9523809523809523,
|
| 2779 |
+
"grad_norm": 0.436716664601782,
|
| 2780 |
+
"learning_rate": 5.802998457313754e-05,
|
| 2781 |
+
"loss": 1.4805458068847657,
|
| 2782 |
+
"memory(GiB)": 69.47,
|
| 2783 |
+
"step": 1270,
|
| 2784 |
+
"token_acc": 0.6780600612337958,
|
| 2785 |
+
"train_speed(iter/s)": 0.068247
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 0.9561304836895388,
|
| 2789 |
+
"grad_norm": 0.5589916229205718,
|
| 2790 |
+
"learning_rate": 5.772366953272888e-05,
|
| 2791 |
+
"loss": 1.5009796142578125,
|
| 2792 |
+
"memory(GiB)": 69.47,
|
| 2793 |
+
"step": 1275,
|
| 2794 |
+
"token_acc": 0.636214271353997,
|
| 2795 |
+
"train_speed(iter/s)": 0.068284
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9598800149981253,
|
| 2799 |
+
"grad_norm": 0.5853408815182066,
|
| 2800 |
+
"learning_rate": 5.741705723357371e-05,
|
| 2801 |
+
"loss": 1.5083969116210938,
|
| 2802 |
+
"memory(GiB)": 69.47,
|
| 2803 |
+
"step": 1280,
|
| 2804 |
+
"token_acc": 0.621927876885384,
|
| 2805 |
+
"train_speed(iter/s)": 0.068318
|
| 2806 |
+
},
|
| 2807 |
+
{
|
| 2808 |
+
"epoch": 0.9636295463067117,
|
| 2809 |
+
"grad_norm": 0.5097897774492353,
|
| 2810 |
+
"learning_rate": 5.7110159476175294e-05,
|
| 2811 |
+
"loss": 1.4834692001342773,
|
| 2812 |
+
"memory(GiB)": 69.47,
|
| 2813 |
+
"step": 1285,
|
| 2814 |
+
"token_acc": 0.6049098142523051,
|
| 2815 |
+
"train_speed(iter/s)": 0.068352
|
| 2816 |
+
},
|
| 2817 |
+
{
|
| 2818 |
+
"epoch": 0.9673790776152981,
|
| 2819 |
+
"grad_norm": 0.509588397268704,
|
| 2820 |
+
"learning_rate": 5.680298807202331e-05,
|
| 2821 |
+
"loss": 1.4848081588745117,
|
| 2822 |
+
"memory(GiB)": 69.47,
|
| 2823 |
+
"step": 1290,
|
| 2824 |
+
"token_acc": 0.6242430593501351,
|
| 2825 |
+
"train_speed(iter/s)": 0.068384
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 0.9711286089238845,
|
| 2829 |
+
"grad_norm": 0.6159708622001356,
|
| 2830 |
+
"learning_rate": 5.649555484313911e-05,
|
| 2831 |
+
"loss": 1.4873367309570313,
|
| 2832 |
+
"memory(GiB)": 69.47,
|
| 2833 |
+
"step": 1295,
|
| 2834 |
+
"token_acc": 0.5676951224583143,
|
| 2835 |
+
"train_speed(iter/s)": 0.068418
|
| 2836 |
+
},
|
| 2837 |
+
{
|
| 2838 |
+
"epoch": 0.974878140232471,
|
| 2839 |
+
"grad_norm": 0.6861668041526542,
|
| 2840 |
+
"learning_rate": 5.6187871621620924e-05,
|
| 2841 |
+
"loss": 1.4572088241577148,
|
| 2842 |
+
"memory(GiB)": 69.47,
|
| 2843 |
+
"step": 1300,
|
| 2844 |
+
"token_acc": 0.6208218302950101,
|
| 2845 |
+
"train_speed(iter/s)": 0.068449
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.974878140232471,
|
| 2849 |
+
"eval_loss": 1.4900087118148804,
|
| 2850 |
+
"eval_runtime": 42.4695,
|
| 2851 |
+
"eval_samples_per_second": 60.844,
|
| 2852 |
+
"eval_steps_per_second": 1.272,
|
| 2853 |
+
"eval_token_acc": 0.6282091089329848,
|
| 2854 |
+
"step": 1300
|
| 2855 |
+
}
|
| 2856 |
+
],
|
| 2857 |
+
"logging_steps": 5,
|
| 2858 |
+
"max_steps": 2666,
|
| 2859 |
+
"num_input_tokens_seen": 0,
|
| 2860 |
+
"num_train_epochs": 2,
|
| 2861 |
+
"save_steps": 50,
|
| 2862 |
+
"stateful_callbacks": {
|
| 2863 |
+
"TrainerControl": {
|
| 2864 |
+
"args": {
|
| 2865 |
+
"should_epoch_stop": false,
|
| 2866 |
+
"should_evaluate": false,
|
| 2867 |
+
"should_log": false,
|
| 2868 |
+
"should_save": true,
|
| 2869 |
+
"should_training_stop": false
|
| 2870 |
+
},
|
| 2871 |
+
"attributes": {}
|
| 2872 |
+
}
|
| 2873 |
+
},
|
| 2874 |
+
"total_flos": 1988579831316480.0,
|
| 2875 |
+
"train_batch_size": 2,
|
| 2876 |
+
"trial_name": null,
|
| 2877 |
+
"trial_params": null
|
| 2878 |
+
}
|