Upload BIO/pretrain_output/qwen2.5-7b-instruct-bio/bio_all/v0-20250608-015630/checkpoint-1300/trainer_state.json with huggingface_hub
Browse files
BIO/pretrain_output/qwen2.5-7b-instruct-bio/bio_all/v0-20250608-015630/checkpoint-1300/trainer_state.json
ADDED
|
@@ -0,0 +1,2878 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1300,
|
| 3 |
+
"best_metric": 1.61236322,
|
| 4 |
+
"best_model_checkpoint": "/oss/wangyujia/BIO/pretrain_output/qwen2.5-7b-instruct-bio/bio_all/v0-20250608-015630/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": 2.5990765920410994,
|
| 15 |
+
"learning_rate": 1.4925373134328358e-07,
|
| 16 |
+
"loss": 1.9313178062438965,
|
| 17 |
+
"memory(GiB)": 33.98,
|
| 18 |
+
"step": 1,
|
| 19 |
+
"token_acc": 0.4522775684089792,
|
| 20 |
+
"train_speed(iter/s)": 0.035384
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.0037495313085864268,
|
| 24 |
+
"grad_norm": 2.464822528919476,
|
| 25 |
+
"learning_rate": 7.462686567164179e-07,
|
| 26 |
+
"loss": 1.9278267621994019,
|
| 27 |
+
"memory(GiB)": 47.91,
|
| 28 |
+
"step": 5,
|
| 29 |
+
"token_acc": 0.52122773204604,
|
| 30 |
+
"train_speed(iter/s)": 0.062923
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0074990626171728535,
|
| 34 |
+
"grad_norm": 1.7325416016205337,
|
| 35 |
+
"learning_rate": 1.4925373134328358e-06,
|
| 36 |
+
"loss": 1.8955060958862304,
|
| 37 |
+
"memory(GiB)": 47.91,
|
| 38 |
+
"step": 10,
|
| 39 |
+
"token_acc": 0.5187258813526064,
|
| 40 |
+
"train_speed(iter/s)": 0.069591
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.01124859392575928,
|
| 44 |
+
"grad_norm": 1.7497313992660668,
|
| 45 |
+
"learning_rate": 2.238805970149254e-06,
|
| 46 |
+
"loss": 1.9058597564697266,
|
| 47 |
+
"memory(GiB)": 47.91,
|
| 48 |
+
"step": 15,
|
| 49 |
+
"token_acc": 0.5233299897068754,
|
| 50 |
+
"train_speed(iter/s)": 0.072156
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.014998125234345707,
|
| 54 |
+
"grad_norm": 2.2036202707744037,
|
| 55 |
+
"learning_rate": 2.9850746268656716e-06,
|
| 56 |
+
"loss": 1.8909317016601563,
|
| 57 |
+
"memory(GiB)": 47.91,
|
| 58 |
+
"step": 20,
|
| 59 |
+
"token_acc": 0.5260313022881105,
|
| 60 |
+
"train_speed(iter/s)": 0.073089
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.018747656542932135,
|
| 64 |
+
"grad_norm": 0.940413680068897,
|
| 65 |
+
"learning_rate": 3.73134328358209e-06,
|
| 66 |
+
"loss": 1.8793617248535157,
|
| 67 |
+
"memory(GiB)": 47.91,
|
| 68 |
+
"step": 25,
|
| 69 |
+
"token_acc": 0.5194384134524312,
|
| 70 |
+
"train_speed(iter/s)": 0.074012
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.02249718785151856,
|
| 74 |
+
"grad_norm": 0.8391713031013472,
|
| 75 |
+
"learning_rate": 4.477611940298508e-06,
|
| 76 |
+
"loss": 1.8508609771728515,
|
| 77 |
+
"memory(GiB)": 47.91,
|
| 78 |
+
"step": 30,
|
| 79 |
+
"token_acc": 0.5043917156593213,
|
| 80 |
+
"train_speed(iter/s)": 0.074566
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.026246719160104987,
|
| 84 |
+
"grad_norm": 0.8841568952278445,
|
| 85 |
+
"learning_rate": 5.2238805970149255e-06,
|
| 86 |
+
"loss": 1.824697494506836,
|
| 87 |
+
"memory(GiB)": 47.91,
|
| 88 |
+
"step": 35,
|
| 89 |
+
"token_acc": 0.528556076112075,
|
| 90 |
+
"train_speed(iter/s)": 0.075038
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.029996250468691414,
|
| 94 |
+
"grad_norm": 0.8946312876696884,
|
| 95 |
+
"learning_rate": 5.970149253731343e-06,
|
| 96 |
+
"loss": 1.8307044982910157,
|
| 97 |
+
"memory(GiB)": 47.91,
|
| 98 |
+
"step": 40,
|
| 99 |
+
"token_acc": 0.5308448585032051,
|
| 100 |
+
"train_speed(iter/s)": 0.075325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03374578177727784,
|
| 104 |
+
"grad_norm": 0.7947638691192278,
|
| 105 |
+
"learning_rate": 6.7164179104477625e-06,
|
| 106 |
+
"loss": 1.816474723815918,
|
| 107 |
+
"memory(GiB)": 47.91,
|
| 108 |
+
"step": 45,
|
| 109 |
+
"token_acc": 0.5064540778037032,
|
| 110 |
+
"train_speed(iter/s)": 0.075663
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.03749531308586427,
|
| 114 |
+
"grad_norm": 0.9078284980783232,
|
| 115 |
+
"learning_rate": 7.46268656716418e-06,
|
| 116 |
+
"loss": 1.8235408782958984,
|
| 117 |
+
"memory(GiB)": 47.91,
|
| 118 |
+
"step": 50,
|
| 119 |
+
"token_acc": 0.5335809746292172,
|
| 120 |
+
"train_speed(iter/s)": 0.075798
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.03749531308586427,
|
| 124 |
+
"eval_loss": 1.8199940919876099,
|
| 125 |
+
"eval_runtime": 43.3327,
|
| 126 |
+
"eval_samples_per_second": 59.632,
|
| 127 |
+
"eval_steps_per_second": 1.246,
|
| 128 |
+
"eval_token_acc": 0.5329827582536721,
|
| 129 |
+
"step": 50
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0412448443944507,
|
| 133 |
+
"grad_norm": 0.9483395054820509,
|
| 134 |
+
"learning_rate": 8.208955223880599e-06,
|
| 135 |
+
"loss": 1.820401382446289,
|
| 136 |
+
"memory(GiB)": 55.02,
|
| 137 |
+
"step": 55,
|
| 138 |
+
"token_acc": 0.5211876688973266,
|
| 139 |
+
"train_speed(iter/s)": 0.067429
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.04499437570303712,
|
| 143 |
+
"grad_norm": 1.068284349391847,
|
| 144 |
+
"learning_rate": 8.955223880597016e-06,
|
| 145 |
+
"loss": 1.8140316009521484,
|
| 146 |
+
"memory(GiB)": 55.02,
|
| 147 |
+
"step": 60,
|
| 148 |
+
"token_acc": 0.5256004431606853,
|
| 149 |
+
"train_speed(iter/s)": 0.068168
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.048743907011623545,
|
| 153 |
+
"grad_norm": 1.0374195399684023,
|
| 154 |
+
"learning_rate": 9.701492537313434e-06,
|
| 155 |
+
"loss": 1.7972326278686523,
|
| 156 |
+
"memory(GiB)": 55.02,
|
| 157 |
+
"step": 65,
|
| 158 |
+
"token_acc": 0.5301121869202227,
|
| 159 |
+
"train_speed(iter/s)": 0.068826
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.05249343832020997,
|
| 163 |
+
"grad_norm": 1.1664573975367183,
|
| 164 |
+
"learning_rate": 9.999861447984952e-06,
|
| 165 |
+
"loss": 1.8000951766967774,
|
| 166 |
+
"memory(GiB)": 55.02,
|
| 167 |
+
"step": 70,
|
| 168 |
+
"token_acc": 0.5479384871713907,
|
| 169 |
+
"train_speed(iter/s)": 0.069381
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.0562429696287964,
|
| 173 |
+
"grad_norm": 1.3747388495283737,
|
| 174 |
+
"learning_rate": 9.99901476903367e-06,
|
| 175 |
+
"loss": 1.7784345626831055,
|
| 176 |
+
"memory(GiB)": 55.02,
|
| 177 |
+
"step": 75,
|
| 178 |
+
"token_acc": 0.5400628212450028,
|
| 179 |
+
"train_speed(iter/s)": 0.069877
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.05999250093738283,
|
| 183 |
+
"grad_norm": 1.049733513434474,
|
| 184 |
+
"learning_rate": 9.997398514657146e-06,
|
| 185 |
+
"loss": 1.7863637924194335,
|
| 186 |
+
"memory(GiB)": 55.02,
|
| 187 |
+
"step": 80,
|
| 188 |
+
"token_acc": 0.543317168313022,
|
| 189 |
+
"train_speed(iter/s)": 0.070337
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.06374203224596925,
|
| 193 |
+
"grad_norm": 1.1176985642823551,
|
| 194 |
+
"learning_rate": 9.995012933670341e-06,
|
| 195 |
+
"loss": 1.8001277923583985,
|
| 196 |
+
"memory(GiB)": 55.02,
|
| 197 |
+
"step": 85,
|
| 198 |
+
"token_acc": 0.5290585589203952,
|
| 199 |
+
"train_speed(iter/s)": 0.070763
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.06749156355455568,
|
| 203 |
+
"grad_norm": 1.2869315249626965,
|
| 204 |
+
"learning_rate": 9.991858393322517e-06,
|
| 205 |
+
"loss": 1.8029617309570312,
|
| 206 |
+
"memory(GiB)": 55.02,
|
| 207 |
+
"step": 90,
|
| 208 |
+
"token_acc": 0.5462134374959176,
|
| 209 |
+
"train_speed(iter/s)": 0.071137
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.0712410948631421,
|
| 213 |
+
"grad_norm": 1.2836534355867217,
|
| 214 |
+
"learning_rate": 9.987935379240715e-06,
|
| 215 |
+
"loss": 1.8010761260986328,
|
| 216 |
+
"memory(GiB)": 55.02,
|
| 217 |
+
"step": 95,
|
| 218 |
+
"token_acc": 0.5229859052198638,
|
| 219 |
+
"train_speed(iter/s)": 0.071455
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07499062617172854,
|
| 223 |
+
"grad_norm": 0.9256584210711153,
|
| 224 |
+
"learning_rate": 9.98324449535498e-06,
|
| 225 |
+
"loss": 1.7873167037963866,
|
| 226 |
+
"memory(GiB)": 55.02,
|
| 227 |
+
"step": 100,
|
| 228 |
+
"token_acc": 0.5686663397712219,
|
| 229 |
+
"train_speed(iter/s)": 0.071771
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.07499062617172854,
|
| 233 |
+
"eval_loss": 1.7935261726379395,
|
| 234 |
+
"eval_runtime": 42.5716,
|
| 235 |
+
"eval_samples_per_second": 60.698,
|
| 236 |
+
"eval_steps_per_second": 1.268,
|
| 237 |
+
"eval_token_acc": 0.5386120720851657,
|
| 238 |
+
"step": 100
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.07874015748031496,
|
| 242 |
+
"grad_norm": 0.9939350045074621,
|
| 243 |
+
"learning_rate": 9.977786463805399e-06,
|
| 244 |
+
"loss": 1.7998859405517578,
|
| 245 |
+
"memory(GiB)": 55.02,
|
| 246 |
+
"step": 105,
|
| 247 |
+
"token_acc": 0.5293935626767713,
|
| 248 |
+
"train_speed(iter/s)": 0.06774
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0824896887889014,
|
| 252 |
+
"grad_norm": 1.0230340656701442,
|
| 253 |
+
"learning_rate": 9.97156212483093e-06,
|
| 254 |
+
"loss": 1.7763248443603517,
|
| 255 |
+
"memory(GiB)": 55.02,
|
| 256 |
+
"step": 110,
|
| 257 |
+
"token_acc": 0.5234773241685314,
|
| 258 |
+
"train_speed(iter/s)": 0.068148
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.08623922009748781,
|
| 262 |
+
"grad_norm": 0.7782735221046598,
|
| 263 |
+
"learning_rate": 9.964572436640046e-06,
|
| 264 |
+
"loss": 1.7829460144042968,
|
| 265 |
+
"memory(GiB)": 55.02,
|
| 266 |
+
"step": 115,
|
| 267 |
+
"token_acc": 0.536639456424291,
|
| 268 |
+
"train_speed(iter/s)": 0.068543
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.08998875140607424,
|
| 272 |
+
"grad_norm": 0.7126371800506376,
|
| 273 |
+
"learning_rate": 9.956818475263228e-06,
|
| 274 |
+
"loss": 1.7885452270507813,
|
| 275 |
+
"memory(GiB)": 55.02,
|
| 276 |
+
"step": 120,
|
| 277 |
+
"token_acc": 0.5467124795676686,
|
| 278 |
+
"train_speed(iter/s)": 0.068903
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.09373828271466067,
|
| 282 |
+
"grad_norm": 0.903212458549272,
|
| 283 |
+
"learning_rate": 9.948301434387308e-06,
|
| 284 |
+
"loss": 1.7893110275268556,
|
| 285 |
+
"memory(GiB)": 55.02,
|
| 286 |
+
"step": 125,
|
| 287 |
+
"token_acc": 0.5382898013748363,
|
| 288 |
+
"train_speed(iter/s)": 0.069222
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.09748781402324709,
|
| 292 |
+
"grad_norm": 0.9564978234402557,
|
| 293 |
+
"learning_rate": 9.939022625171723e-06,
|
| 294 |
+
"loss": 1.7851387023925782,
|
| 295 |
+
"memory(GiB)": 55.02,
|
| 296 |
+
"step": 130,
|
| 297 |
+
"token_acc": 0.5361465251716474,
|
| 298 |
+
"train_speed(iter/s)": 0.069535
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.10123734533183353,
|
| 302 |
+
"grad_norm": 0.9221350564014876,
|
| 303 |
+
"learning_rate": 9.928983476046643e-06,
|
| 304 |
+
"loss": 1.7801361083984375,
|
| 305 |
+
"memory(GiB)": 55.02,
|
| 306 |
+
"step": 135,
|
| 307 |
+
"token_acc": 0.5709161091447238,
|
| 308 |
+
"train_speed(iter/s)": 0.069829
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.10498687664041995,
|
| 312 |
+
"grad_norm": 0.6552383137988806,
|
| 313 |
+
"learning_rate": 9.918185532493095e-06,
|
| 314 |
+
"loss": 1.792304229736328,
|
| 315 |
+
"memory(GiB)": 55.02,
|
| 316 |
+
"step": 140,
|
| 317 |
+
"token_acc": 0.5654260190875665,
|
| 318 |
+
"train_speed(iter/s)": 0.070115
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.10873640794900638,
|
| 322 |
+
"grad_norm": 0.6178739968771222,
|
| 323 |
+
"learning_rate": 9.906630456805024e-06,
|
| 324 |
+
"loss": 1.7768011093139648,
|
| 325 |
+
"memory(GiB)": 55.02,
|
| 326 |
+
"step": 145,
|
| 327 |
+
"token_acc": 0.5628114715561824,
|
| 328 |
+
"train_speed(iter/s)": 0.070369
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.1124859392575928,
|
| 332 |
+
"grad_norm": 0.6956981375298377,
|
| 333 |
+
"learning_rate": 9.894320027833405e-06,
|
| 334 |
+
"loss": 1.791607666015625,
|
| 335 |
+
"memory(GiB)": 55.02,
|
| 336 |
+
"step": 150,
|
| 337 |
+
"token_acc": 0.5352548522917879,
|
| 338 |
+
"train_speed(iter/s)": 0.070604
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1124859392575928,
|
| 342 |
+
"eval_loss": 1.7808756828308105,
|
| 343 |
+
"eval_runtime": 42.4149,
|
| 344 |
+
"eval_samples_per_second": 60.922,
|
| 345 |
+
"eval_steps_per_second": 1.273,
|
| 346 |
+
"eval_token_acc": 0.5413158372963355,
|
| 347 |
+
"step": 150
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.11623547056617922,
|
| 351 |
+
"grad_norm": 0.8022685967483424,
|
| 352 |
+
"learning_rate": 9.881256140712389e-06,
|
| 353 |
+
"loss": 1.8093143463134767,
|
| 354 |
+
"memory(GiB)": 55.02,
|
| 355 |
+
"step": 155,
|
| 356 |
+
"token_acc": 0.5188995435520638,
|
| 357 |
+
"train_speed(iter/s)": 0.068014
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.11998500187476566,
|
| 361 |
+
"grad_norm": 1.0203184952354634,
|
| 362 |
+
"learning_rate": 9.86744080656756e-06,
|
| 363 |
+
"loss": 1.7841522216796875,
|
| 364 |
+
"memory(GiB)": 55.02,
|
| 365 |
+
"step": 160,
|
| 366 |
+
"token_acc": 0.5357633114226129,
|
| 367 |
+
"train_speed(iter/s)": 0.06829
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.12373453318335208,
|
| 371 |
+
"grad_norm": 0.9081018676373969,
|
| 372 |
+
"learning_rate": 9.852876152206325e-06,
|
| 373 |
+
"loss": 1.7839433670043945,
|
| 374 |
+
"memory(GiB)": 55.02,
|
| 375 |
+
"step": 165,
|
| 376 |
+
"token_acc": 0.5591644401908787,
|
| 377 |
+
"train_speed(iter/s)": 0.068546
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.1274840644919385,
|
| 381 |
+
"grad_norm": 0.9907488698487368,
|
| 382 |
+
"learning_rate": 9.837564419790506e-06,
|
| 383 |
+
"loss": 1.768780517578125,
|
| 384 |
+
"memory(GiB)": 55.02,
|
| 385 |
+
"step": 170,
|
| 386 |
+
"token_acc": 0.5500457735733904,
|
| 387 |
+
"train_speed(iter/s)": 0.068794
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13123359580052493,
|
| 391 |
+
"grad_norm": 1.405852122526834,
|
| 392 |
+
"learning_rate": 9.821507966491178e-06,
|
| 393 |
+
"loss": 1.7799259185791017,
|
| 394 |
+
"memory(GiB)": 55.02,
|
| 395 |
+
"step": 175,
|
| 396 |
+
"token_acc": 0.5457307133754187,
|
| 397 |
+
"train_speed(iter/s)": 0.069045
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.13498312710911137,
|
| 401 |
+
"grad_norm": 1.2404474949089965,
|
| 402 |
+
"learning_rate": 9.804709264125772e-06,
|
| 403 |
+
"loss": 1.7638275146484375,
|
| 404 |
+
"memory(GiB)": 55.02,
|
| 405 |
+
"step": 180,
|
| 406 |
+
"token_acc": 0.5435971411710707,
|
| 407 |
+
"train_speed(iter/s)": 0.06928
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.13873265841769777,
|
| 411 |
+
"grad_norm": 1.2017488538172039,
|
| 412 |
+
"learning_rate": 9.787170898777571e-06,
|
| 413 |
+
"loss": 1.7629226684570312,
|
| 414 |
+
"memory(GiB)": 55.02,
|
| 415 |
+
"step": 185,
|
| 416 |
+
"token_acc": 0.5377509092423444,
|
| 417 |
+
"train_speed(iter/s)": 0.069496
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.1424821897262842,
|
| 421 |
+
"grad_norm": 1.4587980981731037,
|
| 422 |
+
"learning_rate": 9.768895570397586e-06,
|
| 423 |
+
"loss": 1.782264518737793,
|
| 424 |
+
"memory(GiB)": 55.02,
|
| 425 |
+
"step": 190,
|
| 426 |
+
"token_acc": 0.5378942424306155,
|
| 427 |
+
"train_speed(iter/s)": 0.069699
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.14623172103487064,
|
| 431 |
+
"grad_norm": 0.9636603641214471,
|
| 432 |
+
"learning_rate": 9.749886092388907e-06,
|
| 433 |
+
"loss": 1.7805797576904296,
|
| 434 |
+
"memory(GiB)": 55.02,
|
| 435 |
+
"step": 195,
|
| 436 |
+
"token_acc": 0.5591106553685625,
|
| 437 |
+
"train_speed(iter/s)": 0.069901
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.14998125234345708,
|
| 441 |
+
"grad_norm": 0.8643693943688852,
|
| 442 |
+
"learning_rate": 9.7301453911736e-06,
|
| 443 |
+
"loss": 1.7657100677490234,
|
| 444 |
+
"memory(GiB)": 55.02,
|
| 445 |
+
"step": 200,
|
| 446 |
+
"token_acc": 0.572268859208869,
|
| 447 |
+
"train_speed(iter/s)": 0.070095
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.14998125234345708,
|
| 451 |
+
"eval_loss": 1.7679647207260132,
|
| 452 |
+
"eval_runtime": 42.6698,
|
| 453 |
+
"eval_samples_per_second": 60.558,
|
| 454 |
+
"eval_steps_per_second": 1.266,
|
| 455 |
+
"eval_token_acc": 0.5451552593969315,
|
| 456 |
+
"step": 200
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.15373078365204348,
|
| 460 |
+
"grad_norm": 0.8657531020540058,
|
| 461 |
+
"learning_rate": 9.709676505742194e-06,
|
| 462 |
+
"loss": 1.7681110382080079,
|
| 463 |
+
"memory(GiB)": 55.02,
|
| 464 |
+
"step": 205,
|
| 465 |
+
"token_acc": 0.539625558780686,
|
| 466 |
+
"train_speed(iter/s)": 0.068103
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.15748031496062992,
|
| 470 |
+
"grad_norm": 0.7555909763046894,
|
| 471 |
+
"learning_rate": 9.688482587185839e-06,
|
| 472 |
+
"loss": 1.7621929168701171,
|
| 473 |
+
"memory(GiB)": 55.02,
|
| 474 |
+
"step": 210,
|
| 475 |
+
"token_acc": 0.5633362247982546,
|
| 476 |
+
"train_speed(iter/s)": 0.068316
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.16122984626921635,
|
| 480 |
+
"grad_norm": 0.8576972485284384,
|
| 481 |
+
"learning_rate": 9.666566898211219e-06,
|
| 482 |
+
"loss": 1.7515193939208984,
|
| 483 |
+
"memory(GiB)": 55.02,
|
| 484 |
+
"step": 215,
|
| 485 |
+
"token_acc": 0.5416328755467514,
|
| 486 |
+
"train_speed(iter/s)": 0.068518
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.1649793775778028,
|
| 490 |
+
"grad_norm": 0.8321394026566008,
|
| 491 |
+
"learning_rate": 9.64393281263826e-06,
|
| 492 |
+
"loss": 1.760385513305664,
|
| 493 |
+
"memory(GiB)": 55.02,
|
| 494 |
+
"step": 220,
|
| 495 |
+
"token_acc": 0.5313785935489345,
|
| 496 |
+
"train_speed(iter/s)": 0.068709
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.1687289088863892,
|
| 500 |
+
"grad_norm": 1.1758097242819285,
|
| 501 |
+
"learning_rate": 9.620583814880763e-06,
|
| 502 |
+
"loss": 1.7725101470947267,
|
| 503 |
+
"memory(GiB)": 55.02,
|
| 504 |
+
"step": 225,
|
| 505 |
+
"token_acc": 0.5523288911890283,
|
| 506 |
+
"train_speed(iter/s)": 0.068896
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.17247844019497563,
|
| 510 |
+
"grad_norm": 1.1624264053744016,
|
| 511 |
+
"learning_rate": 9.59652349940998e-06,
|
| 512 |
+
"loss": 1.7543354034423828,
|
| 513 |
+
"memory(GiB)": 55.02,
|
| 514 |
+
"step": 230,
|
| 515 |
+
"token_acc": 0.5785625028925812,
|
| 516 |
+
"train_speed(iter/s)": 0.069071
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.17622797150356206,
|
| 520 |
+
"grad_norm": 0.7310036085359608,
|
| 521 |
+
"learning_rate": 9.571755570201266e-06,
|
| 522 |
+
"loss": 1.7507953643798828,
|
| 523 |
+
"memory(GiB)": 55.02,
|
| 524 |
+
"step": 235,
|
| 525 |
+
"token_acc": 0.5411558669001751,
|
| 526 |
+
"train_speed(iter/s)": 0.069246
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.17997750281214847,
|
| 530 |
+
"grad_norm": 0.8077849501412758,
|
| 531 |
+
"learning_rate": 9.54628384016387e-06,
|
| 532 |
+
"loss": 1.7498720169067383,
|
| 533 |
+
"memory(GiB)": 55.02,
|
| 534 |
+
"step": 240,
|
| 535 |
+
"token_acc": 0.5107944339998463,
|
| 536 |
+
"train_speed(iter/s)": 0.069418
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.1837270341207349,
|
| 540 |
+
"grad_norm": 0.9225839356980445,
|
| 541 |
+
"learning_rate": 9.520112230553959e-06,
|
| 542 |
+
"loss": 1.7527772903442382,
|
| 543 |
+
"memory(GiB)": 55.02,
|
| 544 |
+
"step": 245,
|
| 545 |
+
"token_acc": 0.5650339768500239,
|
| 546 |
+
"train_speed(iter/s)": 0.069581
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.18747656542932134,
|
| 550 |
+
"grad_norm": 0.9228761761744316,
|
| 551 |
+
"learning_rate": 9.493244770370947e-06,
|
| 552 |
+
"loss": 1.739366340637207,
|
| 553 |
+
"memory(GiB)": 55.02,
|
| 554 |
+
"step": 250,
|
| 555 |
+
"token_acc": 0.5212063259548612,
|
| 556 |
+
"train_speed(iter/s)": 0.069727
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.18747656542932134,
|
| 560 |
+
"eval_loss": 1.7509726285934448,
|
| 561 |
+
"eval_runtime": 42.5093,
|
| 562 |
+
"eval_samples_per_second": 60.787,
|
| 563 |
+
"eval_steps_per_second": 1.27,
|
| 564 |
+
"eval_token_acc": 0.5502310060336623,
|
| 565 |
+
"step": 250
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.19122609673790777,
|
| 569 |
+
"grad_norm": 0.753239532675668,
|
| 570 |
+
"learning_rate": 9.465685595737263e-06,
|
| 571 |
+
"loss": 1.743215560913086,
|
| 572 |
+
"memory(GiB)": 56.25,
|
| 573 |
+
"step": 255,
|
| 574 |
+
"token_acc": 0.5399775807972443,
|
| 575 |
+
"train_speed(iter/s)": 0.068194
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.19497562804649418,
|
| 579 |
+
"grad_norm": 0.8488311405689892,
|
| 580 |
+
"learning_rate": 9.437438949261602e-06,
|
| 581 |
+
"loss": 1.7420181274414062,
|
| 582 |
+
"memory(GiB)": 56.25,
|
| 583 |
+
"step": 260,
|
| 584 |
+
"token_acc": 0.5663597161966146,
|
| 585 |
+
"train_speed(iter/s)": 0.068357
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.19872515935508062,
|
| 589 |
+
"grad_norm": 0.6800443905154068,
|
| 590 |
+
"learning_rate": 9.408509179385806e-06,
|
| 591 |
+
"loss": 1.7466453552246093,
|
| 592 |
+
"memory(GiB)": 56.25,
|
| 593 |
+
"step": 265,
|
| 594 |
+
"token_acc": 0.5432435786077677,
|
| 595 |
+
"train_speed(iter/s)": 0.068528
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.20247469066366705,
|
| 599 |
+
"grad_norm": 0.87203145697415,
|
| 600 |
+
"learning_rate": 9.378900739715429e-06,
|
| 601 |
+
"loss": 1.7354595184326171,
|
| 602 |
+
"memory(GiB)": 56.25,
|
| 603 |
+
"step": 270,
|
| 604 |
+
"token_acc": 0.5630195594992452,
|
| 605 |
+
"train_speed(iter/s)": 0.068693
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.20622422197225346,
|
| 609 |
+
"grad_norm": 0.8095685158538298,
|
| 610 |
+
"learning_rate": 9.348618188334135e-06,
|
| 611 |
+
"loss": 1.740234375,
|
| 612 |
+
"memory(GiB)": 56.25,
|
| 613 |
+
"step": 275,
|
| 614 |
+
"token_acc": 0.5514069894056335,
|
| 615 |
+
"train_speed(iter/s)": 0.068842
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.2099737532808399,
|
| 619 |
+
"grad_norm": 1.00206570868519,
|
| 620 |
+
"learning_rate": 9.317666187101996e-06,
|
| 621 |
+
"loss": 1.7255264282226563,
|
| 622 |
+
"memory(GiB)": 56.25,
|
| 623 |
+
"step": 280,
|
| 624 |
+
"token_acc": 0.5434367055763205,
|
| 625 |
+
"train_speed(iter/s)": 0.068989
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.21372328458942633,
|
| 629 |
+
"grad_norm": 0.7742674981350653,
|
| 630 |
+
"learning_rate": 9.286049500937826e-06,
|
| 631 |
+
"loss": 1.7388019561767578,
|
| 632 |
+
"memory(GiB)": 56.25,
|
| 633 |
+
"step": 285,
|
| 634 |
+
"token_acc": 0.5465616294659914,
|
| 635 |
+
"train_speed(iter/s)": 0.069143
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 0.21747281589801276,
|
| 639 |
+
"grad_norm": 0.7030518360692997,
|
| 640 |
+
"learning_rate": 9.253772997085635e-06,
|
| 641 |
+
"loss": 1.7331905364990234,
|
| 642 |
+
"memory(GiB)": 56.25,
|
| 643 |
+
"step": 290,
|
| 644 |
+
"token_acc": 0.5593017630617072,
|
| 645 |
+
"train_speed(iter/s)": 0.069286
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.22122234720659917,
|
| 649 |
+
"grad_norm": 0.8016063917603425,
|
| 650 |
+
"learning_rate": 9.220841644365343e-06,
|
| 651 |
+
"loss": 1.7626869201660156,
|
| 652 |
+
"memory(GiB)": 56.25,
|
| 653 |
+
"step": 295,
|
| 654 |
+
"token_acc": 0.5474898329303144,
|
| 655 |
+
"train_speed(iter/s)": 0.069424
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.2249718785151856,
|
| 659 |
+
"grad_norm": 0.7507835036035847,
|
| 660 |
+
"learning_rate": 9.18726051240786e-06,
|
| 661 |
+
"loss": 1.7293975830078125,
|
| 662 |
+
"memory(GiB)": 56.25,
|
| 663 |
+
"step": 300,
|
| 664 |
+
"token_acc": 0.5329011645735721,
|
| 665 |
+
"train_speed(iter/s)": 0.069556
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.2249718785151856,
|
| 669 |
+
"eval_loss": 1.7327964305877686,
|
| 670 |
+
"eval_runtime": 42.3753,
|
| 671 |
+
"eval_samples_per_second": 60.979,
|
| 672 |
+
"eval_steps_per_second": 1.274,
|
| 673 |
+
"eval_token_acc": 0.5556736652702177,
|
| 674 |
+
"step": 300
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.22872140982377204,
|
| 678 |
+
"grad_norm": 0.6548926760419295,
|
| 679 |
+
"learning_rate": 9.15303477087463e-06,
|
| 680 |
+
"loss": 1.7221017837524415,
|
| 681 |
+
"memory(GiB)": 56.25,
|
| 682 |
+
"step": 305,
|
| 683 |
+
"token_acc": 0.5494975649750837,
|
| 684 |
+
"train_speed(iter/s)": 0.068232
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.23247094113235844,
|
| 688 |
+
"grad_norm": 0.6751231265805964,
|
| 689 |
+
"learning_rate": 9.118169688661785e-06,
|
| 690 |
+
"loss": 1.7264883041381835,
|
| 691 |
+
"memory(GiB)": 56.25,
|
| 692 |
+
"step": 310,
|
| 693 |
+
"token_acc": 0.5448713201882233,
|
| 694 |
+
"train_speed(iter/s)": 0.068365
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.23622047244094488,
|
| 698 |
+
"grad_norm": 0.605793575125745,
|
| 699 |
+
"learning_rate": 9.082670633089028e-06,
|
| 700 |
+
"loss": 1.73472900390625,
|
| 701 |
+
"memory(GiB)": 56.25,
|
| 702 |
+
"step": 315,
|
| 703 |
+
"token_acc": 0.5575434638870522,
|
| 704 |
+
"train_speed(iter/s)": 0.068493
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.2399700037495313,
|
| 708 |
+
"grad_norm": 0.7043926428771533,
|
| 709 |
+
"learning_rate": 9.046543069073361e-06,
|
| 710 |
+
"loss": 1.7091964721679687,
|
| 711 |
+
"memory(GiB)": 56.25,
|
| 712 |
+
"step": 320,
|
| 713 |
+
"token_acc": 0.5550820254340313,
|
| 714 |
+
"train_speed(iter/s)": 0.068631
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.24371953505811775,
|
| 718 |
+
"grad_norm": 0.9370678193137794,
|
| 719 |
+
"learning_rate": 9.009792558287777e-06,
|
| 720 |
+
"loss": 1.6914459228515626,
|
| 721 |
+
"memory(GiB)": 56.25,
|
| 722 |
+
"step": 325,
|
| 723 |
+
"token_acc": 0.5512165093950109,
|
| 724 |
+
"train_speed(iter/s)": 0.068756
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.24746906636670415,
|
| 728 |
+
"grad_norm": 0.731053265622514,
|
| 729 |
+
"learning_rate": 8.972424758305073e-06,
|
| 730 |
+
"loss": 1.7122673034667968,
|
| 731 |
+
"memory(GiB)": 56.25,
|
| 732 |
+
"step": 330,
|
| 733 |
+
"token_acc": 0.5968278226389376,
|
| 734 |
+
"train_speed(iter/s)": 0.068874
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.2512185976752906,
|
| 738 |
+
"grad_norm": 0.7366718319888682,
|
| 739 |
+
"learning_rate": 8.934445421726888e-06,
|
| 740 |
+
"loss": 1.7182106018066405,
|
| 741 |
+
"memory(GiB)": 56.25,
|
| 742 |
+
"step": 335,
|
| 743 |
+
"token_acc": 0.5372630837386368,
|
| 744 |
+
"train_speed(iter/s)": 0.068995
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.254968128983877,
|
| 748 |
+
"grad_norm": 0.7982735213900777,
|
| 749 |
+
"learning_rate": 8.895860395298121e-06,
|
| 750 |
+
"loss": 1.6983676910400392,
|
| 751 |
+
"memory(GiB)": 56.25,
|
| 752 |
+
"step": 340,
|
| 753 |
+
"token_acc": 0.5665421173385775,
|
| 754 |
+
"train_speed(iter/s)": 0.06911
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.25871766029246346,
|
| 758 |
+
"grad_norm": 0.6678550800092121,
|
| 759 |
+
"learning_rate": 8.85667561900685e-06,
|
| 760 |
+
"loss": 1.706036376953125,
|
| 761 |
+
"memory(GiB)": 56.25,
|
| 762 |
+
"step": 345,
|
| 763 |
+
"token_acc": 0.5582484033156679,
|
| 764 |
+
"train_speed(iter/s)": 0.069229
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.26246719160104987,
|
| 768 |
+
"grad_norm": 0.9786171582688546,
|
| 769 |
+
"learning_rate": 8.816897125169894e-06,
|
| 770 |
+
"loss": 1.7054430007934571,
|
| 771 |
+
"memory(GiB)": 56.25,
|
| 772 |
+
"step": 350,
|
| 773 |
+
"token_acc": 0.5696195618783663,
|
| 774 |
+
"train_speed(iter/s)": 0.069341
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.26246719160104987,
|
| 778 |
+
"eval_loss": 1.7168406248092651,
|
| 779 |
+
"eval_runtime": 42.5781,
|
| 780 |
+
"eval_samples_per_second": 60.688,
|
| 781 |
+
"eval_steps_per_second": 1.268,
|
| 782 |
+
"eval_token_acc": 0.5604134336360598,
|
| 783 |
+
"step": 350
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.26621672290963627,
|
| 787 |
+
"grad_norm": 0.6683654189193357,
|
| 788 |
+
"learning_rate": 8.77653103750417e-06,
|
| 789 |
+
"loss": 1.7113515853881835,
|
| 790 |
+
"memory(GiB)": 56.25,
|
| 791 |
+
"step": 355,
|
| 792 |
+
"token_acc": 0.5511189438910106,
|
| 793 |
+
"train_speed(iter/s)": 0.068103
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.26996625421822273,
|
| 797 |
+
"grad_norm": 0.5865077122513812,
|
| 798 |
+
"learning_rate": 8.735583570183974e-06,
|
| 799 |
+
"loss": 1.692765235900879,
|
| 800 |
+
"memory(GiB)": 56.25,
|
| 801 |
+
"step": 360,
|
| 802 |
+
"token_acc": 0.5464648214223854,
|
| 803 |
+
"train_speed(iter/s)": 0.068223
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.27371578552680914,
|
| 807 |
+
"grad_norm": 0.7402904205476919,
|
| 808 |
+
"learning_rate": 8.694061026884336e-06,
|
| 809 |
+
"loss": 1.7002754211425781,
|
| 810 |
+
"memory(GiB)": 56.25,
|
| 811 |
+
"step": 365,
|
| 812 |
+
"token_acc": 0.5705252525252525,
|
| 813 |
+
"train_speed(iter/s)": 0.068345
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.27746531683539555,
|
| 817 |
+
"grad_norm": 0.7571500537288366,
|
| 818 |
+
"learning_rate": 8.6519697998106e-06,
|
| 819 |
+
"loss": 1.7284061431884765,
|
| 820 |
+
"memory(GiB)": 56.25,
|
| 821 |
+
"step": 370,
|
| 822 |
+
"token_acc": 0.5748253023425441,
|
| 823 |
+
"train_speed(iter/s)": 0.068462
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.281214848143982,
|
| 827 |
+
"grad_norm": 0.6730846815952083,
|
| 828 |
+
"learning_rate": 8.609316368714371e-06,
|
| 829 |
+
"loss": 1.702937126159668,
|
| 830 |
+
"memory(GiB)": 56.25,
|
| 831 |
+
"step": 375,
|
| 832 |
+
"token_acc": 0.5256013175105557,
|
| 833 |
+
"train_speed(iter/s)": 0.068577
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.2849643794525684,
|
| 837 |
+
"grad_norm": 0.8462516862337176,
|
| 838 |
+
"learning_rate": 8.566107299895988e-06,
|
| 839 |
+
"loss": 1.7094646453857423,
|
| 840 |
+
"memory(GiB)": 56.25,
|
| 841 |
+
"step": 380,
|
| 842 |
+
"token_acc": 0.5898579849946409,
|
| 843 |
+
"train_speed(iter/s)": 0.068693
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.2887139107611549,
|
| 847 |
+
"grad_norm": 0.7482611525034915,
|
| 848 |
+
"learning_rate": 8.52234924519367e-06,
|
| 849 |
+
"loss": 1.7032199859619142,
|
| 850 |
+
"memory(GiB)": 56.25,
|
| 851 |
+
"step": 385,
|
| 852 |
+
"token_acc": 0.5420495124984953,
|
| 853 |
+
"train_speed(iter/s)": 0.068801
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.2924634420697413,
|
| 857 |
+
"grad_norm": 0.696958027931327,
|
| 858 |
+
"learning_rate": 8.478048940959503e-06,
|
| 859 |
+
"loss": 1.6919610977172852,
|
| 860 |
+
"memory(GiB)": 56.25,
|
| 861 |
+
"step": 390,
|
| 862 |
+
"token_acc": 0.5534309086430637,
|
| 863 |
+
"train_speed(iter/s)": 0.068909
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.2962129733783277,
|
| 867 |
+
"grad_norm": 0.5696305911831858,
|
| 868 |
+
"learning_rate": 8.433213207022404e-06,
|
| 869 |
+
"loss": 1.6833906173706055,
|
| 870 |
+
"memory(GiB)": 56.25,
|
| 871 |
+
"step": 395,
|
| 872 |
+
"token_acc": 0.5494928069022507,
|
| 873 |
+
"train_speed(iter/s)": 0.069018
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.29996250468691416,
|
| 877 |
+
"grad_norm": 1.0078405970909443,
|
| 878 |
+
"learning_rate": 8.387848945638235e-06,
|
| 879 |
+
"loss": 1.6984685897827148,
|
| 880 |
+
"memory(GiB)": 56.25,
|
| 881 |
+
"step": 400,
|
| 882 |
+
"token_acc": 0.5672347570979136,
|
| 883 |
+
"train_speed(iter/s)": 0.069122
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.29996250468691416,
|
| 887 |
+
"eval_loss": 1.7021883726119995,
|
| 888 |
+
"eval_runtime": 42.5994,
|
| 889 |
+
"eval_samples_per_second": 60.658,
|
| 890 |
+
"eval_steps_per_second": 1.268,
|
| 891 |
+
"eval_token_acc": 0.5648435441261425,
|
| 892 |
+
"step": 400
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.30371203599550056,
|
| 896 |
+
"grad_norm": 0.8681641024485509,
|
| 897 |
+
"learning_rate": 8.341963140427242e-06,
|
| 898 |
+
"loss": 1.7013641357421876,
|
| 899 |
+
"memory(GiB)": 56.25,
|
| 900 |
+
"step": 405,
|
| 901 |
+
"token_acc": 0.5476675242907254,
|
| 902 |
+
"train_speed(iter/s)": 0.068148
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.30746156730408697,
|
| 906 |
+
"grad_norm": 0.8528961417036697,
|
| 907 |
+
"learning_rate": 8.295562855298954e-06,
|
| 908 |
+
"loss": 1.6895477294921875,
|
| 909 |
+
"memory(GiB)": 56.25,
|
| 910 |
+
"step": 410,
|
| 911 |
+
"token_acc": 0.5703600854040318,
|
| 912 |
+
"train_speed(iter/s)": 0.068257
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.31121109861267343,
|
| 916 |
+
"grad_norm": 0.7966281406635305,
|
| 917 |
+
"learning_rate": 8.248655233364724e-06,
|
| 918 |
+
"loss": 1.703116226196289,
|
| 919 |
+
"memory(GiB)": 56.25,
|
| 920 |
+
"step": 415,
|
| 921 |
+
"token_acc": 0.5499540331642208,
|
| 922 |
+
"train_speed(iter/s)": 0.068363
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.31496062992125984,
|
| 926 |
+
"grad_norm": 0.7388959481441792,
|
| 927 |
+
"learning_rate": 8.201247495838087e-06,
|
| 928 |
+
"loss": 1.7138626098632812,
|
| 929 |
+
"memory(GiB)": 56.25,
|
| 930 |
+
"step": 420,
|
| 931 |
+
"token_acc": 0.5509550893931054,
|
| 932 |
+
"train_speed(iter/s)": 0.06846
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.31871016122984624,
|
| 936 |
+
"grad_norm": 0.6668812721055195,
|
| 937 |
+
"learning_rate": 8.153346940923076e-06,
|
| 938 |
+
"loss": 1.7163211822509765,
|
| 939 |
+
"memory(GiB)": 56.25,
|
| 940 |
+
"step": 425,
|
| 941 |
+
"token_acc": 0.5339552948740659,
|
| 942 |
+
"train_speed(iter/s)": 0.068558
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.3224596925384327,
|
| 946 |
+
"grad_norm": 0.8897674249190526,
|
| 947 |
+
"learning_rate": 8.104960942690709e-06,
|
| 948 |
+
"loss": 1.6957286834716796,
|
| 949 |
+
"memory(GiB)": 56.25,
|
| 950 |
+
"step": 430,
|
| 951 |
+
"token_acc": 0.5618721676102363,
|
| 952 |
+
"train_speed(iter/s)": 0.068655
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.3262092238470191,
|
| 956 |
+
"grad_norm": 0.6998717792447529,
|
| 957 |
+
"learning_rate": 8.056096949943777e-06,
|
| 958 |
+
"loss": 1.7086776733398437,
|
| 959 |
+
"memory(GiB)": 56.25,
|
| 960 |
+
"step": 435,
|
| 961 |
+
"token_acc": 0.5755440200763176,
|
| 962 |
+
"train_speed(iter/s)": 0.068745
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.3299587551556056,
|
| 966 |
+
"grad_norm": 0.8395556848202133,
|
| 967 |
+
"learning_rate": 8.006762485070138e-06,
|
| 968 |
+
"loss": 1.6877761840820313,
|
| 969 |
+
"memory(GiB)": 56.25,
|
| 970 |
+
"step": 440,
|
| 971 |
+
"token_acc": 0.574201770399201,
|
| 972 |
+
"train_speed(iter/s)": 0.068841
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.333708286464192,
|
| 976 |
+
"grad_norm": 0.7641827410573238,
|
| 977 |
+
"learning_rate": 7.956965142884678e-06,
|
| 978 |
+
"loss": 1.6771770477294923,
|
| 979 |
+
"memory(GiB)": 56.25,
|
| 980 |
+
"step": 445,
|
| 981 |
+
"token_acc": 0.5614385821706016,
|
| 982 |
+
"train_speed(iter/s)": 0.068935
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3374578177727784,
|
| 986 |
+
"grad_norm": 0.7082113636629225,
|
| 987 |
+
"learning_rate": 7.906712589460124e-06,
|
| 988 |
+
"loss": 1.6755485534667969,
|
| 989 |
+
"memory(GiB)": 56.25,
|
| 990 |
+
"step": 450,
|
| 991 |
+
"token_acc": 0.5444774209358365,
|
| 992 |
+
"train_speed(iter/s)": 0.06903
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.3374578177727784,
|
| 996 |
+
"eval_loss": 1.6896531581878662,
|
| 997 |
+
"eval_runtime": 42.4554,
|
| 998 |
+
"eval_samples_per_second": 60.864,
|
| 999 |
+
"eval_steps_per_second": 1.272,
|
| 1000 |
+
"eval_token_acc": 0.5685763214385071,
|
| 1001 |
+
"step": 450
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.34120734908136485,
|
| 1005 |
+
"grad_norm": 0.7807331567305439,
|
| 1006 |
+
"learning_rate": 7.85601256094689e-06,
|
| 1007 |
+
"loss": 1.6793529510498046,
|
| 1008 |
+
"memory(GiB)": 56.25,
|
| 1009 |
+
"step": 455,
|
| 1010 |
+
"token_acc": 0.5562552637973148,
|
| 1011 |
+
"train_speed(iter/s)": 0.068158
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.34495688038995126,
|
| 1015 |
+
"grad_norm": 0.7642434347013445,
|
| 1016 |
+
"learning_rate": 7.804872862382132e-06,
|
| 1017 |
+
"loss": 1.7076757431030274,
|
| 1018 |
+
"memory(GiB)": 56.25,
|
| 1019 |
+
"step": 460,
|
| 1020 |
+
"token_acc": 0.5478764760999728,
|
| 1021 |
+
"train_speed(iter/s)": 0.068252
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.34870641169853767,
|
| 1025 |
+
"grad_norm": 0.7018356558770414,
|
| 1026 |
+
"learning_rate": 7.753301366488187e-06,
|
| 1027 |
+
"loss": 1.6917272567749024,
|
| 1028 |
+
"memory(GiB)": 56.25,
|
| 1029 |
+
"step": 465,
|
| 1030 |
+
"token_acc": 0.6022955543080534,
|
| 1031 |
+
"train_speed(iter/s)": 0.068345
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.35245594300712413,
|
| 1035 |
+
"grad_norm": 0.7600704177992731,
|
| 1036 |
+
"learning_rate": 7.701306012460627e-06,
|
| 1037 |
+
"loss": 1.6827001571655273,
|
| 1038 |
+
"memory(GiB)": 56.25,
|
| 1039 |
+
"step": 470,
|
| 1040 |
+
"token_acc": 0.5491566959780885,
|
| 1041 |
+
"train_speed(iter/s)": 0.068431
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.35620547431571054,
|
| 1045 |
+
"grad_norm": 0.6673098312724823,
|
| 1046 |
+
"learning_rate": 7.648894804746031e-06,
|
| 1047 |
+
"loss": 1.6575685501098634,
|
| 1048 |
+
"memory(GiB)": 56.25,
|
| 1049 |
+
"step": 475,
|
| 1050 |
+
"token_acc": 0.5520260966055696,
|
| 1051 |
+
"train_speed(iter/s)": 0.068525
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.35995500562429694,
|
| 1055 |
+
"grad_norm": 0.6492156502549102,
|
| 1056 |
+
"learning_rate": 7.596075811809753e-06,
|
| 1057 |
+
"loss": 1.6792295455932618,
|
| 1058 |
+
"memory(GiB)": 56.25,
|
| 1059 |
+
"step": 480,
|
| 1060 |
+
"token_acc": 0.5631662021022937,
|
| 1061 |
+
"train_speed(iter/s)": 0.068609
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.3637045369328834,
|
| 1065 |
+
"grad_norm": 0.6958194870105822,
|
| 1066 |
+
"learning_rate": 7.542857164893816e-06,
|
| 1067 |
+
"loss": 1.7015853881835938,
|
| 1068 |
+
"memory(GiB)": 56.25,
|
| 1069 |
+
"step": 485,
|
| 1070 |
+
"token_acc": 0.5768759371536606,
|
| 1071 |
+
"train_speed(iter/s)": 0.06869
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 0.3674540682414698,
|
| 1075 |
+
"grad_norm": 0.685535234176806,
|
| 1076 |
+
"learning_rate": 7.489247056765134e-06,
|
| 1077 |
+
"loss": 1.7032451629638672,
|
| 1078 |
+
"memory(GiB)": 56.25,
|
| 1079 |
+
"step": 490,
|
| 1080 |
+
"token_acc": 0.561515250858049,
|
| 1081 |
+
"train_speed(iter/s)": 0.068773
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.3712035995500562,
|
| 1085 |
+
"grad_norm": 0.7726913261926227,
|
| 1086 |
+
"learning_rate": 7.4352537404542935e-06,
|
| 1087 |
+
"loss": 1.6790489196777343,
|
| 1088 |
+
"memory(GiB)": 56.25,
|
| 1089 |
+
"step": 495,
|
| 1090 |
+
"token_acc": 0.5535088912920483,
|
| 1091 |
+
"train_speed(iter/s)": 0.068855
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.3749531308586427,
|
| 1095 |
+
"grad_norm": 0.6215971963792798,
|
| 1096 |
+
"learning_rate": 7.380885527985016e-06,
|
| 1097 |
+
"loss": 1.6753128051757813,
|
| 1098 |
+
"memory(GiB)": 56.25,
|
| 1099 |
+
"step": 500,
|
| 1100 |
+
"token_acc": 0.5804309499145921,
|
| 1101 |
+
"train_speed(iter/s)": 0.068939
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3749531308586427,
|
| 1105 |
+
"eval_loss": 1.6785035133361816,
|
| 1106 |
+
"eval_runtime": 42.5761,
|
| 1107 |
+
"eval_samples_per_second": 60.691,
|
| 1108 |
+
"eval_steps_per_second": 1.268,
|
| 1109 |
+
"eval_token_acc": 0.5717880388170251,
|
| 1110 |
+
"step": 500
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.3787026621672291,
|
| 1114 |
+
"grad_norm": 0.7571930333892343,
|
| 1115 |
+
"learning_rate": 7.326150789094571e-06,
|
| 1116 |
+
"loss": 1.6797908782958983,
|
| 1117 |
+
"memory(GiB)": 56.25,
|
| 1118 |
+
"step": 505,
|
| 1119 |
+
"token_acc": 0.5666724708340588,
|
| 1120 |
+
"train_speed(iter/s)": 0.068153
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.38245219347581555,
|
| 1124 |
+
"grad_norm": 0.5273195202991797,
|
| 1125 |
+
"learning_rate": 7.271057949945297e-06,
|
| 1126 |
+
"loss": 1.6877239227294922,
|
| 1127 |
+
"memory(GiB)": 56.25,
|
| 1128 |
+
"step": 510,
|
| 1129 |
+
"token_acc": 0.5795585438375612,
|
| 1130 |
+
"train_speed(iter/s)": 0.068239
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.38620172478440196,
|
| 1134 |
+
"grad_norm": 0.644647069123782,
|
| 1135 |
+
"learning_rate": 7.2156154918274194e-06,
|
| 1136 |
+
"loss": 1.6607694625854492,
|
| 1137 |
+
"memory(GiB)": 56.25,
|
| 1138 |
+
"step": 515,
|
| 1139 |
+
"token_acc": 0.5639819575190945,
|
| 1140 |
+
"train_speed(iter/s)": 0.068326
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.38995125609298836,
|
| 1144 |
+
"grad_norm": 0.8427621851278584,
|
| 1145 |
+
"learning_rate": 7.159831949853409e-06,
|
| 1146 |
+
"loss": 1.6399917602539062,
|
| 1147 |
+
"memory(GiB)": 56.25,
|
| 1148 |
+
"step": 520,
|
| 1149 |
+
"token_acc": 0.5831872255951778,
|
| 1150 |
+
"train_speed(iter/s)": 0.068407
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3937007874015748,
|
| 1154 |
+
"grad_norm": 0.7860592513933189,
|
| 1155 |
+
"learning_rate": 7.103715911644029e-06,
|
| 1156 |
+
"loss": 1.6666389465332032,
|
| 1157 |
+
"memory(GiB)": 56.25,
|
| 1158 |
+
"step": 525,
|
| 1159 |
+
"token_acc": 0.6023049439476434,
|
| 1160 |
+
"train_speed(iter/s)": 0.068489
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.39745031871016123,
|
| 1164 |
+
"grad_norm": 0.8223721237706454,
|
| 1165 |
+
"learning_rate": 7.047276016006318e-06,
|
| 1166 |
+
"loss": 1.695573616027832,
|
| 1167 |
+
"memory(GiB)": 56.25,
|
| 1168 |
+
"step": 530,
|
| 1169 |
+
"token_acc": 0.5720906143643139,
|
| 1170 |
+
"train_speed(iter/s)": 0.068571
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.40119985001874764,
|
| 1174 |
+
"grad_norm": 0.7877770309362587,
|
| 1175 |
+
"learning_rate": 6.990520951603682e-06,
|
| 1176 |
+
"loss": 1.6714542388916016,
|
| 1177 |
+
"memory(GiB)": 56.25,
|
| 1178 |
+
"step": 535,
|
| 1179 |
+
"token_acc": 0.5545731394354149,
|
| 1180 |
+
"train_speed(iter/s)": 0.068649
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.4049493813273341,
|
| 1184 |
+
"grad_norm": 0.5849943208055685,
|
| 1185 |
+
"learning_rate": 6.933459455618312e-06,
|
| 1186 |
+
"loss": 1.663173294067383,
|
| 1187 |
+
"memory(GiB)": 56.25,
|
| 1188 |
+
"step": 540,
|
| 1189 |
+
"token_acc": 0.5916160291243944,
|
| 1190 |
+
"train_speed(iter/s)": 0.068729
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.4086989126359205,
|
| 1194 |
+
"grad_norm": 0.6975030189627052,
|
| 1195 |
+
"learning_rate": 6.876100312406141e-06,
|
| 1196 |
+
"loss": 1.672979736328125,
|
| 1197 |
+
"memory(GiB)": 56.25,
|
| 1198 |
+
"step": 545,
|
| 1199 |
+
"token_acc": 0.5678786183750525,
|
| 1200 |
+
"train_speed(iter/s)": 0.068809
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.4124484439445069,
|
| 1204 |
+
"grad_norm": 0.8832443309131802,
|
| 1205 |
+
"learning_rate": 6.818452352144527e-06,
|
| 1206 |
+
"loss": 1.663263702392578,
|
| 1207 |
+
"memory(GiB)": 56.25,
|
| 1208 |
+
"step": 550,
|
| 1209 |
+
"token_acc": 0.5777180525291963,
|
| 1210 |
+
"train_speed(iter/s)": 0.068885
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.4124484439445069,
|
| 1214 |
+
"eval_loss": 1.669090747833252,
|
| 1215 |
+
"eval_runtime": 42.3533,
|
| 1216 |
+
"eval_samples_per_second": 61.011,
|
| 1217 |
+
"eval_steps_per_second": 1.275,
|
| 1218 |
+
"eval_token_acc": 0.5746502507096283,
|
| 1219 |
+
"step": 550
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.4161979752530934,
|
| 1223 |
+
"grad_norm": 0.7101210397707838,
|
| 1224 |
+
"learning_rate": 6.760524449472889e-06,
|
| 1225 |
+
"loss": 1.653615951538086,
|
| 1226 |
+
"memory(GiB)": 56.25,
|
| 1227 |
+
"step": 555,
|
| 1228 |
+
"token_acc": 0.563550813623823,
|
| 1229 |
+
"train_speed(iter/s)": 0.068165
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.4199475065616798,
|
| 1233 |
+
"grad_norm": 0.7092912767199613,
|
| 1234 |
+
"learning_rate": 6.702325522126503e-06,
|
| 1235 |
+
"loss": 1.663838768005371,
|
| 1236 |
+
"memory(GiB)": 56.25,
|
| 1237 |
+
"step": 560,
|
| 1238 |
+
"token_acc": 0.5644253522152244,
|
| 1239 |
+
"train_speed(iter/s)": 0.068242
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.4236970378702662,
|
| 1243 |
+
"grad_norm": 0.8492548795119265,
|
| 1244 |
+
"learning_rate": 6.643864529563644e-06,
|
| 1245 |
+
"loss": 1.670602798461914,
|
| 1246 |
+
"memory(GiB)": 56.25,
|
| 1247 |
+
"step": 565,
|
| 1248 |
+
"token_acc": 0.5895219375017935,
|
| 1249 |
+
"train_speed(iter/s)": 0.068317
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.42744656917885265,
|
| 1253 |
+
"grad_norm": 0.8995573558948677,
|
| 1254 |
+
"learning_rate": 6.5851504715863345e-06,
|
| 1255 |
+
"loss": 1.661231231689453,
|
| 1256 |
+
"memory(GiB)": 56.25,
|
| 1257 |
+
"step": 570,
|
| 1258 |
+
"token_acc": 0.5656868865305689,
|
| 1259 |
+
"train_speed(iter/s)": 0.068394
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.43119610048743906,
|
| 1263 |
+
"grad_norm": 0.8124537569507534,
|
| 1264 |
+
"learning_rate": 6.526192386954853e-06,
|
| 1265 |
+
"loss": 1.6499761581420898,
|
| 1266 |
+
"memory(GiB)": 56.25,
|
| 1267 |
+
"step": 575,
|
| 1268 |
+
"token_acc": 0.5715577630764308,
|
| 1269 |
+
"train_speed(iter/s)": 0.06847
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4349456317960255,
|
| 1273 |
+
"grad_norm": 0.7462194182878145,
|
| 1274 |
+
"learning_rate": 6.466999351996266e-06,
|
| 1275 |
+
"loss": 1.6548486709594727,
|
| 1276 |
+
"memory(GiB)": 56.25,
|
| 1277 |
+
"step": 580,
|
| 1278 |
+
"token_acc": 0.5821864123070383,
|
| 1279 |
+
"train_speed(iter/s)": 0.068539
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 0.43869516310461193,
|
| 1283 |
+
"grad_norm": 0.6235784183842846,
|
| 1284 |
+
"learning_rate": 6.407580479207166e-06,
|
| 1285 |
+
"loss": 1.6605405807495117,
|
| 1286 |
+
"memory(GiB)": 56.25,
|
| 1287 |
+
"step": 585,
|
| 1288 |
+
"token_acc": 0.5868372560515956,
|
| 1289 |
+
"train_speed(iter/s)": 0.068608
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.44244469441319834,
|
| 1293 |
+
"grad_norm": 0.6346754736055541,
|
| 1294 |
+
"learning_rate": 6.347944915850846e-06,
|
| 1295 |
+
"loss": 1.6546539306640624,
|
| 1296 |
+
"memory(GiB)": 56.25,
|
| 1297 |
+
"step": 590,
|
| 1298 |
+
"token_acc": 0.5951824777399143,
|
| 1299 |
+
"train_speed(iter/s)": 0.068679
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.4461942257217848,
|
| 1303 |
+
"grad_norm": 0.5320819638022591,
|
| 1304 |
+
"learning_rate": 6.288101842549117e-06,
|
| 1305 |
+
"loss": 1.6418937683105468,
|
| 1306 |
+
"memory(GiB)": 56.25,
|
| 1307 |
+
"step": 595,
|
| 1308 |
+
"token_acc": 0.595732223476298,
|
| 1309 |
+
"train_speed(iter/s)": 0.068751
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 0.4499437570303712,
|
| 1313 |
+
"grad_norm": 0.5547338870237839,
|
| 1314 |
+
"learning_rate": 6.228060471868998e-06,
|
| 1315 |
+
"loss": 1.6649009704589843,
|
| 1316 |
+
"memory(GiB)": 56.25,
|
| 1317 |
+
"step": 600,
|
| 1318 |
+
"token_acc": 0.5625491462592676,
|
| 1319 |
+
"train_speed(iter/s)": 0.068817
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.4499437570303712,
|
| 1323 |
+
"eval_loss": 1.6608483791351318,
|
| 1324 |
+
"eval_runtime": 42.6065,
|
| 1325 |
+
"eval_samples_per_second": 60.648,
|
| 1326 |
+
"eval_steps_per_second": 1.267,
|
| 1327 |
+
"eval_token_acc": 0.5769941290838164,
|
| 1328 |
+
"step": 600
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.4536932883389576,
|
| 1332 |
+
"grad_norm": 0.5482512110483113,
|
| 1333 |
+
"learning_rate": 6.167830046904481e-06,
|
| 1334 |
+
"loss": 1.656534767150879,
|
| 1335 |
+
"memory(GiB)": 56.25,
|
| 1336 |
+
"step": 605,
|
| 1337 |
+
"token_acc": 0.5782699227350105,
|
| 1338 |
+
"train_speed(iter/s)": 0.068151
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.4574428196475441,
|
| 1342 |
+
"grad_norm": 0.7099660977474908,
|
| 1343 |
+
"learning_rate": 6.1074198398535964e-06,
|
| 1344 |
+
"loss": 1.6757530212402343,
|
| 1345 |
+
"memory(GiB)": 56.25,
|
| 1346 |
+
"step": 610,
|
| 1347 |
+
"token_acc": 0.5806772839007541,
|
| 1348 |
+
"train_speed(iter/s)": 0.068224
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.4611923509561305,
|
| 1352 |
+
"grad_norm": 0.6849024601166154,
|
| 1353 |
+
"learning_rate": 6.0468391505910064e-06,
|
| 1354 |
+
"loss": 1.6539661407470703,
|
| 1355 |
+
"memory(GiB)": 56.25,
|
| 1356 |
+
"step": 615,
|
| 1357 |
+
"token_acc": 0.5897343123691958,
|
| 1358 |
+
"train_speed(iter/s)": 0.068293
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.4649418822647169,
|
| 1362 |
+
"grad_norm": 0.6387929116637958,
|
| 1363 |
+
"learning_rate": 5.986097305236327e-06,
|
| 1364 |
+
"loss": 1.6756690979003905,
|
| 1365 |
+
"memory(GiB)": 56.25,
|
| 1366 |
+
"step": 620,
|
| 1367 |
+
"token_acc": 0.566821454125962,
|
| 1368 |
+
"train_speed(iter/s)": 0.068364
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.46869141357330335,
|
| 1372 |
+
"grad_norm": 0.5972563894220153,
|
| 1373 |
+
"learning_rate": 5.925203654718416e-06,
|
| 1374 |
+
"loss": 1.647392463684082,
|
| 1375 |
+
"memory(GiB)": 56.25,
|
| 1376 |
+
"step": 625,
|
| 1377 |
+
"token_acc": 0.5551161839909385,
|
| 1378 |
+
"train_speed(iter/s)": 0.068434
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.47244094488188976,
|
| 1382 |
+
"grad_norm": 0.7595259975476277,
|
| 1383 |
+
"learning_rate": 5.8641675733358415e-06,
|
| 1384 |
+
"loss": 1.6608802795410156,
|
| 1385 |
+
"memory(GiB)": 56.25,
|
| 1386 |
+
"step": 630,
|
| 1387 |
+
"token_acc": 0.5813017902089991,
|
| 1388 |
+
"train_speed(iter/s)": 0.068499
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.47619047619047616,
|
| 1392 |
+
"grad_norm": 0.6016960158110866,
|
| 1393 |
+
"learning_rate": 5.8029984573137545e-06,
|
| 1394 |
+
"loss": 1.6570241928100586,
|
| 1395 |
+
"memory(GiB)": 56.25,
|
| 1396 |
+
"step": 635,
|
| 1397 |
+
"token_acc": 0.5774235481033102,
|
| 1398 |
+
"train_speed(iter/s)": 0.068565
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 0.4799400074990626,
|
| 1402 |
+
"grad_norm": 0.6354804378746023,
|
| 1403 |
+
"learning_rate": 5.741705723357372e-06,
|
| 1404 |
+
"loss": 1.6393739700317382,
|
| 1405 |
+
"memory(GiB)": 56.25,
|
| 1406 |
+
"step": 640,
|
| 1407 |
+
"token_acc": 0.5658296440587923,
|
| 1408 |
+
"train_speed(iter/s)": 0.068634
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.48368953880764903,
|
| 1412 |
+
"grad_norm": 0.6139385718585026,
|
| 1413 |
+
"learning_rate": 5.680298807202332e-06,
|
| 1414 |
+
"loss": 1.6520938873291016,
|
| 1415 |
+
"memory(GiB)": 56.25,
|
| 1416 |
+
"step": 645,
|
| 1417 |
+
"token_acc": 0.5701600067018514,
|
| 1418 |
+
"train_speed(iter/s)": 0.068699
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.4874390701162355,
|
| 1422 |
+
"grad_norm": 0.5387776161432817,
|
| 1423 |
+
"learning_rate": 5.618787162162093e-06,
|
| 1424 |
+
"loss": 1.6615371704101562,
|
| 1425 |
+
"memory(GiB)": 56.25,
|
| 1426 |
+
"step": 650,
|
| 1427 |
+
"token_acc": 0.58879061422444,
|
| 1428 |
+
"train_speed(iter/s)": 0.068765
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.4874390701162355,
|
| 1432 |
+
"eval_loss": 1.6536719799041748,
|
| 1433 |
+
"eval_runtime": 42.5747,
|
| 1434 |
+
"eval_samples_per_second": 60.693,
|
| 1435 |
+
"eval_steps_per_second": 1.268,
|
| 1436 |
+
"eval_token_acc": 0.5791460048943351,
|
| 1437 |
+
"step": 650
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.4911886014248219,
|
| 1441 |
+
"grad_norm": 0.7530435416345947,
|
| 1442 |
+
"learning_rate": 5.557180257672651e-06,
|
| 1443 |
+
"loss": 1.6289875030517578,
|
| 1444 |
+
"memory(GiB)": 56.25,
|
| 1445 |
+
"step": 655,
|
| 1446 |
+
"token_acc": 0.5682923514329584,
|
| 1447 |
+
"train_speed(iter/s)": 0.06816
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 0.4949381327334083,
|
| 1451 |
+
"grad_norm": 0.7717324275253389,
|
| 1452 |
+
"learning_rate": 5.495487577834758e-06,
|
| 1453 |
+
"loss": 1.6309438705444337,
|
| 1454 |
+
"memory(GiB)": 56.25,
|
| 1455 |
+
"step": 660,
|
| 1456 |
+
"token_acc": 0.587423404922088,
|
| 1457 |
+
"train_speed(iter/s)": 0.068227
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.49868766404199477,
|
| 1461 |
+
"grad_norm": 0.5895405199877466,
|
| 1462 |
+
"learning_rate": 5.433718619953883e-06,
|
| 1463 |
+
"loss": 1.6529617309570312,
|
| 1464 |
+
"memory(GiB)": 56.25,
|
| 1465 |
+
"step": 665,
|
| 1466 |
+
"token_acc": 0.6041080816725761,
|
| 1467 |
+
"train_speed(iter/s)": 0.068294
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.5024371953505812,
|
| 1471 |
+
"grad_norm": 0.6454461142803576,
|
| 1472 |
+
"learning_rate": 5.3718828930781564e-06,
|
| 1473 |
+
"loss": 1.6488149642944336,
|
| 1474 |
+
"memory(GiB)": 56.25,
|
| 1475 |
+
"step": 670,
|
| 1476 |
+
"token_acc": 0.5847953625785856,
|
| 1477 |
+
"train_speed(iter/s)": 0.068355
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 0.5061867266591676,
|
| 1481 |
+
"grad_norm": 0.673823253803827,
|
| 1482 |
+
"learning_rate": 5.309989916534482e-06,
|
| 1483 |
+
"loss": 1.6447179794311524,
|
| 1484 |
+
"memory(GiB)": 56.25,
|
| 1485 |
+
"step": 675,
|
| 1486 |
+
"token_acc": 0.5982730626303525,
|
| 1487 |
+
"train_speed(iter/s)": 0.068416
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.509936257967754,
|
| 1491 |
+
"grad_norm": 0.5749920984370196,
|
| 1492 |
+
"learning_rate": 5.2480492184630975e-06,
|
| 1493 |
+
"loss": 1.6585128784179688,
|
| 1494 |
+
"memory(GiB)": 56.25,
|
| 1495 |
+
"step": 680,
|
| 1496 |
+
"token_acc": 0.5631895577350786,
|
| 1497 |
+
"train_speed(iter/s)": 0.068475
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.5136857892763405,
|
| 1501 |
+
"grad_norm": 0.6387955325138472,
|
| 1502 |
+
"learning_rate": 5.1860703343507415e-06,
|
| 1503 |
+
"loss": 1.6461997985839845,
|
| 1504 |
+
"memory(GiB)": 56.25,
|
| 1505 |
+
"step": 685,
|
| 1506 |
+
"token_acc": 0.5752287735058492,
|
| 1507 |
+
"train_speed(iter/s)": 0.068538
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5174353205849269,
|
| 1511 |
+
"grad_norm": 0.4715376939278375,
|
| 1512 |
+
"learning_rate": 5.124062805562725e-06,
|
| 1513 |
+
"loss": 1.6515190124511718,
|
| 1514 |
+
"memory(GiB)": 56.25,
|
| 1515 |
+
"step": 690,
|
| 1516 |
+
"token_acc": 0.5873850771494762,
|
| 1517 |
+
"train_speed(iter/s)": 0.068601
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 0.5211848518935133,
|
| 1521 |
+
"grad_norm": 0.625438137972778,
|
| 1522 |
+
"learning_rate": 5.062036177874075e-06,
|
| 1523 |
+
"loss": 1.6295310974121093,
|
| 1524 |
+
"memory(GiB)": 56.25,
|
| 1525 |
+
"step": 695,
|
| 1526 |
+
"token_acc": 0.5652246926848928,
|
| 1527 |
+
"train_speed(iter/s)": 0.06866
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.5249343832020997,
|
| 1531 |
+
"grad_norm": 0.5483253705980401,
|
| 1532 |
+
"learning_rate": 5e-06,
|
| 1533 |
+
"loss": 1.6582225799560546,
|
| 1534 |
+
"memory(GiB)": 56.25,
|
| 1535 |
+
"step": 700,
|
| 1536 |
+
"token_acc": 0.5789502069653619,
|
| 1537 |
+
"train_speed(iter/s)": 0.068716
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.5249343832020997,
|
| 1541 |
+
"eval_loss": 1.646588921546936,
|
| 1542 |
+
"eval_runtime": 42.7729,
|
| 1543 |
+
"eval_samples_per_second": 60.412,
|
| 1544 |
+
"eval_steps_per_second": 1.262,
|
| 1545 |
+
"eval_token_acc": 0.581152436927787,
|
| 1546 |
+
"step": 700
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.5286839145106862,
|
| 1550 |
+
"grad_norm": 0.6796052373678554,
|
| 1551 |
+
"learning_rate": 4.937963822125928e-06,
|
| 1552 |
+
"loss": 1.6299430847167968,
|
| 1553 |
+
"memory(GiB)": 56.25,
|
| 1554 |
+
"step": 705,
|
| 1555 |
+
"token_acc": 0.5749007372754903,
|
| 1556 |
+
"train_speed(iter/s)": 0.068145
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5324334458192725,
|
| 1560 |
+
"grad_norm": 0.5623730859974818,
|
| 1561 |
+
"learning_rate": 4.875937194437275e-06,
|
| 1562 |
+
"loss": 1.6273029327392579,
|
| 1563 |
+
"memory(GiB)": 56.25,
|
| 1564 |
+
"step": 710,
|
| 1565 |
+
"token_acc": 0.5774565153542408,
|
| 1566 |
+
"train_speed(iter/s)": 0.068206
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 0.536182977127859,
|
| 1570 |
+
"grad_norm": 0.5556142061160076,
|
| 1571 |
+
"learning_rate": 4.813929665649261e-06,
|
| 1572 |
+
"loss": 1.6527215957641601,
|
| 1573 |
+
"memory(GiB)": 56.25,
|
| 1574 |
+
"step": 715,
|
| 1575 |
+
"token_acc": 0.5842528134716451,
|
| 1576 |
+
"train_speed(iter/s)": 0.068268
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.5399325084364455,
|
| 1580 |
+
"grad_norm": 0.6937818396456759,
|
| 1581 |
+
"learning_rate": 4.751950781536905e-06,
|
| 1582 |
+
"loss": 1.6325241088867188,
|
| 1583 |
+
"memory(GiB)": 56.25,
|
| 1584 |
+
"step": 720,
|
| 1585 |
+
"token_acc": 0.550934829059829,
|
| 1586 |
+
"train_speed(iter/s)": 0.068328
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.5436820397450318,
|
| 1590 |
+
"grad_norm": 0.6616141973881087,
|
| 1591 |
+
"learning_rate": 4.6900100834655185e-06,
|
| 1592 |
+
"loss": 1.6385433197021484,
|
| 1593 |
+
"memory(GiB)": 56.25,
|
| 1594 |
+
"step": 725,
|
| 1595 |
+
"token_acc": 0.6094119122781343,
|
| 1596 |
+
"train_speed(iter/s)": 0.068386
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 0.5474315710536183,
|
| 1600 |
+
"grad_norm": 0.6804973836917435,
|
| 1601 |
+
"learning_rate": 4.628117106921845e-06,
|
| 1602 |
+
"loss": 1.64478759765625,
|
| 1603 |
+
"memory(GiB)": 56.25,
|
| 1604 |
+
"step": 730,
|
| 1605 |
+
"token_acc": 0.5699465347192243,
|
| 1606 |
+
"train_speed(iter/s)": 0.068446
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.5511811023622047,
|
| 1610 |
+
"grad_norm": 0.644058141393178,
|
| 1611 |
+
"learning_rate": 4.566281380046117e-06,
|
| 1612 |
+
"loss": 1.6421289443969727,
|
| 1613 |
+
"memory(GiB)": 56.25,
|
| 1614 |
+
"step": 735,
|
| 1615 |
+
"token_acc": 0.6001181374556146,
|
| 1616 |
+
"train_speed(iter/s)": 0.068503
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 0.5549306336707911,
|
| 1620 |
+
"grad_norm": 0.656894451589833,
|
| 1621 |
+
"learning_rate": 4.5045124221652445e-06,
|
| 1622 |
+
"loss": 1.6506340026855468,
|
| 1623 |
+
"memory(GiB)": 56.25,
|
| 1624 |
+
"step": 740,
|
| 1625 |
+
"token_acc": 0.5980861889787159,
|
| 1626 |
+
"train_speed(iter/s)": 0.068564
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5586801649793776,
|
| 1630 |
+
"grad_norm": 0.6955052749545766,
|
| 1631 |
+
"learning_rate": 4.44281974232735e-06,
|
| 1632 |
+
"loss": 1.615028190612793,
|
| 1633 |
+
"memory(GiB)": 56.25,
|
| 1634 |
+
"step": 745,
|
| 1635 |
+
"token_acc": 0.5978904615718982,
|
| 1636 |
+
"train_speed(iter/s)": 0.06862
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 0.562429696287964,
|
| 1640 |
+
"grad_norm": 0.5828454100208457,
|
| 1641 |
+
"learning_rate": 4.381212837837909e-06,
|
| 1642 |
+
"loss": 1.6183147430419922,
|
| 1643 |
+
"memory(GiB)": 56.25,
|
| 1644 |
+
"step": 750,
|
| 1645 |
+
"token_acc": 0.5896384248921636,
|
| 1646 |
+
"train_speed(iter/s)": 0.068677
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.562429696287964,
|
| 1650 |
+
"eval_loss": 1.6402919292449951,
|
| 1651 |
+
"eval_runtime": 42.4068,
|
| 1652 |
+
"eval_samples_per_second": 60.934,
|
| 1653 |
+
"eval_steps_per_second": 1.273,
|
| 1654 |
+
"eval_token_acc": 0.5830287350554422,
|
| 1655 |
+
"step": 750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.5661792275965505,
|
| 1659 |
+
"grad_norm": 0.6333529913255437,
|
| 1660 |
+
"learning_rate": 4.319701192797671e-06,
|
| 1661 |
+
"loss": 1.6344381332397462,
|
| 1662 |
+
"memory(GiB)": 56.25,
|
| 1663 |
+
"step": 755,
|
| 1664 |
+
"token_acc": 0.5781321414957826,
|
| 1665 |
+
"train_speed(iter/s)": 0.068158
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 0.5699287589051368,
|
| 1669 |
+
"grad_norm": 0.7581894334078688,
|
| 1670 |
+
"learning_rate": 4.258294276642629e-06,
|
| 1671 |
+
"loss": 1.6357366561889648,
|
| 1672 |
+
"memory(GiB)": 56.25,
|
| 1673 |
+
"step": 760,
|
| 1674 |
+
"token_acc": 0.5599813650128116,
|
| 1675 |
+
"train_speed(iter/s)": 0.068217
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.5736782902137233,
|
| 1679 |
+
"grad_norm": 0.6702527130448565,
|
| 1680 |
+
"learning_rate": 4.197001542686248e-06,
|
| 1681 |
+
"loss": 1.623959732055664,
|
| 1682 |
+
"memory(GiB)": 56.25,
|
| 1683 |
+
"step": 765,
|
| 1684 |
+
"token_acc": 0.600753268699251,
|
| 1685 |
+
"train_speed(iter/s)": 0.068276
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 0.5774278215223098,
|
| 1689 |
+
"grad_norm": 0.5903205080818366,
|
| 1690 |
+
"learning_rate": 4.135832426664159e-06,
|
| 1691 |
+
"loss": 1.644424819946289,
|
| 1692 |
+
"memory(GiB)": 56.25,
|
| 1693 |
+
"step": 770,
|
| 1694 |
+
"token_acc": 0.5540505934872132,
|
| 1695 |
+
"train_speed(iter/s)": 0.068329
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 0.5811773528308961,
|
| 1699 |
+
"grad_norm": 0.6763632334448669,
|
| 1700 |
+
"learning_rate": 4.074796345281586e-06,
|
| 1701 |
+
"loss": 1.637106704711914,
|
| 1702 |
+
"memory(GiB)": 56.25,
|
| 1703 |
+
"step": 775,
|
| 1704 |
+
"token_acc": 0.5886795020069392,
|
| 1705 |
+
"train_speed(iter/s)": 0.068385
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.5849268841394826,
|
| 1709 |
+
"grad_norm": 0.5776530336437159,
|
| 1710 |
+
"learning_rate": 4.013902694763675e-06,
|
| 1711 |
+
"loss": 1.6276361465454101,
|
| 1712 |
+
"memory(GiB)": 56.25,
|
| 1713 |
+
"step": 780,
|
| 1714 |
+
"token_acc": 0.5987051157048562,
|
| 1715 |
+
"train_speed(iter/s)": 0.068441
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 0.588676415448069,
|
| 1719 |
+
"grad_norm": 0.6034625719784629,
|
| 1720 |
+
"learning_rate": 3.953160849408996e-06,
|
| 1721 |
+
"loss": 1.629465103149414,
|
| 1722 |
+
"memory(GiB)": 56.25,
|
| 1723 |
+
"step": 785,
|
| 1724 |
+
"token_acc": 0.618713946271952,
|
| 1725 |
+
"train_speed(iter/s)": 0.068491
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.5924259467566554,
|
| 1729 |
+
"grad_norm": 0.5419553921726837,
|
| 1730 |
+
"learning_rate": 3.892580160146406e-06,
|
| 1731 |
+
"loss": 1.6470441818237305,
|
| 1732 |
+
"memory(GiB)": 56.25,
|
| 1733 |
+
"step": 790,
|
| 1734 |
+
"token_acc": 0.5608493116499604,
|
| 1735 |
+
"train_speed(iter/s)": 0.068545
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 0.5961754780652418,
|
| 1739 |
+
"grad_norm": 0.5829879329157514,
|
| 1740 |
+
"learning_rate": 3.832169953095521e-06,
|
| 1741 |
+
"loss": 1.625351333618164,
|
| 1742 |
+
"memory(GiB)": 56.25,
|
| 1743 |
+
"step": 795,
|
| 1744 |
+
"token_acc": 0.5870836712285511,
|
| 1745 |
+
"train_speed(iter/s)": 0.068598
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5999250093738283,
|
| 1749 |
+
"grad_norm": 0.6022627590387872,
|
| 1750 |
+
"learning_rate": 3.771939528131002e-06,
|
| 1751 |
+
"loss": 1.6552854537963868,
|
| 1752 |
+
"memory(GiB)": 56.25,
|
| 1753 |
+
"step": 800,
|
| 1754 |
+
"token_acc": 0.5696196443784868,
|
| 1755 |
+
"train_speed(iter/s)": 0.068653
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 0.5999250093738283,
|
| 1759 |
+
"eval_loss": 1.6350337266921997,
|
| 1760 |
+
"eval_runtime": 42.2859,
|
| 1761 |
+
"eval_samples_per_second": 61.108,
|
| 1762 |
+
"eval_steps_per_second": 1.277,
|
| 1763 |
+
"eval_token_acc": 0.5844979583867213,
|
| 1764 |
+
"step": 800
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 0.6036745406824147,
|
| 1768 |
+
"grad_norm": 0.5942705099349358,
|
| 1769 |
+
"learning_rate": 3.7118981574508845e-06,
|
| 1770 |
+
"loss": 1.6126163482666016,
|
| 1771 |
+
"memory(GiB)": 69.47,
|
| 1772 |
+
"step": 805,
|
| 1773 |
+
"token_acc": 0.581555841193403,
|
| 1774 |
+
"train_speed(iter/s)": 0.068172
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.6074240719910011,
|
| 1778 |
+
"grad_norm": 0.6117465912856315,
|
| 1779 |
+
"learning_rate": 3.652055084149155e-06,
|
| 1780 |
+
"loss": 1.6182636260986327,
|
| 1781 |
+
"memory(GiB)": 69.47,
|
| 1782 |
+
"step": 810,
|
| 1783 |
+
"token_acc": 0.5813004785607525,
|
| 1784 |
+
"train_speed(iter/s)": 0.068228
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 0.6111736032995876,
|
| 1788 |
+
"grad_norm": 0.575721344187492,
|
| 1789 |
+
"learning_rate": 3.5924195207928353e-06,
|
| 1790 |
+
"loss": 1.6351320266723632,
|
| 1791 |
+
"memory(GiB)": 69.47,
|
| 1792 |
+
"step": 815,
|
| 1793 |
+
"token_acc": 0.5948800021886178,
|
| 1794 |
+
"train_speed(iter/s)": 0.068282
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6149231346081739,
|
| 1798 |
+
"grad_norm": 0.6604622423054451,
|
| 1799 |
+
"learning_rate": 3.5330006480037347e-06,
|
| 1800 |
+
"loss": 1.624082374572754,
|
| 1801 |
+
"memory(GiB)": 69.47,
|
| 1802 |
+
"step": 820,
|
| 1803 |
+
"token_acc": 0.592462272265228,
|
| 1804 |
+
"train_speed(iter/s)": 0.068336
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 0.6186726659167604,
|
| 1808 |
+
"grad_norm": 0.5815560175144889,
|
| 1809 |
+
"learning_rate": 3.4738076130451486e-06,
|
| 1810 |
+
"loss": 1.611598587036133,
|
| 1811 |
+
"memory(GiB)": 69.47,
|
| 1812 |
+
"step": 825,
|
| 1813 |
+
"token_acc": 0.6089666369873775,
|
| 1814 |
+
"train_speed(iter/s)": 0.068389
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.6224221972253469,
|
| 1818 |
+
"grad_norm": 0.6364044991983494,
|
| 1819 |
+
"learning_rate": 3.4148495284136667e-06,
|
| 1820 |
+
"loss": 1.6444446563720703,
|
| 1821 |
+
"memory(GiB)": 69.47,
|
| 1822 |
+
"step": 830,
|
| 1823 |
+
"token_acc": 0.5723734347579862,
|
| 1824 |
+
"train_speed(iter/s)": 0.068442
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.6261717285339332,
|
| 1828 |
+
"grad_norm": 0.6154495413241989,
|
| 1829 |
+
"learning_rate": 3.3561354704363563e-06,
|
| 1830 |
+
"loss": 1.6572345733642577,
|
| 1831 |
+
"memory(GiB)": 69.47,
|
| 1832 |
+
"step": 835,
|
| 1833 |
+
"token_acc": 0.5823670492160711,
|
| 1834 |
+
"train_speed(iter/s)": 0.068497
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 0.6299212598425197,
|
| 1838 |
+
"grad_norm": 0.5623546818589049,
|
| 1839 |
+
"learning_rate": 3.2976744778734988e-06,
|
| 1840 |
+
"loss": 1.6390718460083007,
|
| 1841 |
+
"memory(GiB)": 69.47,
|
| 1842 |
+
"step": 840,
|
| 1843 |
+
"token_acc": 0.5890854687950369,
|
| 1844 |
+
"train_speed(iter/s)": 0.068547
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.6336707911511061,
|
| 1848 |
+
"grad_norm": 0.5272938620213984,
|
| 1849 |
+
"learning_rate": 3.2394755505271125e-06,
|
| 1850 |
+
"loss": 1.6225774765014649,
|
| 1851 |
+
"memory(GiB)": 69.47,
|
| 1852 |
+
"step": 845,
|
| 1853 |
+
"token_acc": 0.5662987770510283,
|
| 1854 |
+
"train_speed(iter/s)": 0.068596
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 0.6374203224596925,
|
| 1858 |
+
"grad_norm": 0.6526804036833902,
|
| 1859 |
+
"learning_rate": 3.181547647855475e-06,
|
| 1860 |
+
"loss": 1.613861083984375,
|
| 1861 |
+
"memory(GiB)": 69.47,
|
| 1862 |
+
"step": 850,
|
| 1863 |
+
"token_acc": 0.5893846343231525,
|
| 1864 |
+
"train_speed(iter/s)": 0.068648
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6374203224596925,
|
| 1868 |
+
"eval_loss": 1.6293845176696777,
|
| 1869 |
+
"eval_runtime": 42.4837,
|
| 1870 |
+
"eval_samples_per_second": 60.823,
|
| 1871 |
+
"eval_steps_per_second": 1.271,
|
| 1872 |
+
"eval_token_acc": 0.5861385264416682,
|
| 1873 |
+
"step": 850
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.641169853768279,
|
| 1877 |
+
"grad_norm": 0.6043970243871736,
|
| 1878 |
+
"learning_rate": 3.1238996875938604e-06,
|
| 1879 |
+
"loss": 1.611886215209961,
|
| 1880 |
+
"memory(GiB)": 69.47,
|
| 1881 |
+
"step": 855,
|
| 1882 |
+
"token_acc": 0.5842907996411311,
|
| 1883 |
+
"train_speed(iter/s)": 0.068192
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 0.6449193850768654,
|
| 1887 |
+
"grad_norm": 0.5603396498660076,
|
| 1888 |
+
"learning_rate": 3.0665405443816886e-06,
|
| 1889 |
+
"loss": 1.6631004333496093,
|
| 1890 |
+
"memory(GiB)": 69.47,
|
| 1891 |
+
"step": 860,
|
| 1892 |
+
"token_acc": 0.5783650290912643,
|
| 1893 |
+
"train_speed(iter/s)": 0.068245
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.6486689163854518,
|
| 1897 |
+
"grad_norm": 0.6300453949234538,
|
| 1898 |
+
"learning_rate": 3.009479048396321e-06,
|
| 1899 |
+
"loss": 1.6396780014038086,
|
| 1900 |
+
"memory(GiB)": 69.47,
|
| 1901 |
+
"step": 865,
|
| 1902 |
+
"token_acc": 0.6025950351187603,
|
| 1903 |
+
"train_speed(iter/s)": 0.068295
|
| 1904 |
+
},
|
| 1905 |
+
{
|
| 1906 |
+
"epoch": 0.6524184476940382,
|
| 1907 |
+
"grad_norm": 0.594302672746407,
|
| 1908 |
+
"learning_rate": 2.952723983993684e-06,
|
| 1909 |
+
"loss": 1.6129501342773438,
|
| 1910 |
+
"memory(GiB)": 69.47,
|
| 1911 |
+
"step": 870,
|
| 1912 |
+
"token_acc": 0.5875128078883285,
|
| 1913 |
+
"train_speed(iter/s)": 0.068345
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6561679790026247,
|
| 1917 |
+
"grad_norm": 0.5632989526424523,
|
| 1918 |
+
"learning_rate": 2.8962840883559724e-06,
|
| 1919 |
+
"loss": 1.6319787979125977,
|
| 1920 |
+
"memory(GiB)": 69.47,
|
| 1921 |
+
"step": 875,
|
| 1922 |
+
"token_acc": 0.5844086299001264,
|
| 1923 |
+
"train_speed(iter/s)": 0.068395
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 0.6599175103112112,
|
| 1927 |
+
"grad_norm": 0.6424604810798464,
|
| 1928 |
+
"learning_rate": 2.840168050146591e-06,
|
| 1929 |
+
"loss": 1.6522933959960937,
|
| 1930 |
+
"memory(GiB)": 69.47,
|
| 1931 |
+
"step": 880,
|
| 1932 |
+
"token_acc": 0.608855717389555,
|
| 1933 |
+
"train_speed(iter/s)": 0.068443
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.6636670416197975,
|
| 1937 |
+
"grad_norm": 0.6469172755932601,
|
| 1938 |
+
"learning_rate": 2.7843845081725814e-06,
|
| 1939 |
+
"loss": 1.643354606628418,
|
| 1940 |
+
"memory(GiB)": 69.47,
|
| 1941 |
+
"step": 885,
|
| 1942 |
+
"token_acc": 0.5810390809667215,
|
| 1943 |
+
"train_speed(iter/s)": 0.06849
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.667416572928384,
|
| 1947 |
+
"grad_norm": 0.5183485601959567,
|
| 1948 |
+
"learning_rate": 2.728942050054705e-06,
|
| 1949 |
+
"loss": 1.6205764770507813,
|
| 1950 |
+
"memory(GiB)": 69.47,
|
| 1951 |
+
"step": 890,
|
| 1952 |
+
"token_acc": 0.5878430401103352,
|
| 1953 |
+
"train_speed(iter/s)": 0.068532
|
| 1954 |
+
},
|
| 1955 |
+
{
|
| 1956 |
+
"epoch": 0.6711661042369704,
|
| 1957 |
+
"grad_norm": 0.6352153807462149,
|
| 1958 |
+
"learning_rate": 2.6738492109054305e-06,
|
| 1959 |
+
"loss": 1.6091579437255858,
|
| 1960 |
+
"memory(GiB)": 69.47,
|
| 1961 |
+
"step": 895,
|
| 1962 |
+
"token_acc": 0.5991015059266862,
|
| 1963 |
+
"train_speed(iter/s)": 0.068581
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.6749156355455568,
|
| 1967 |
+
"grad_norm": 0.5703799504436093,
|
| 1968 |
+
"learning_rate": 2.6191144720149853e-06,
|
| 1969 |
+
"loss": 1.6384305953979492,
|
| 1970 |
+
"memory(GiB)": 69.47,
|
| 1971 |
+
"step": 900,
|
| 1972 |
+
"token_acc": 0.5845274359009385,
|
| 1973 |
+
"train_speed(iter/s)": 0.068623
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 0.6749156355455568,
|
| 1977 |
+
"eval_loss": 1.6254901885986328,
|
| 1978 |
+
"eval_runtime": 42.4366,
|
| 1979 |
+
"eval_samples_per_second": 60.891,
|
| 1980 |
+
"eval_steps_per_second": 1.272,
|
| 1981 |
+
"eval_token_acc": 0.5872697791215511,
|
| 1982 |
+
"step": 900
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.6786651668541432,
|
| 1986 |
+
"grad_norm": 0.5425064005079052,
|
| 1987 |
+
"learning_rate": 2.5647462595457073e-06,
|
| 1988 |
+
"loss": 1.5991607666015626,
|
| 1989 |
+
"memory(GiB)": 69.47,
|
| 1990 |
+
"step": 905,
|
| 1991 |
+
"token_acc": 0.5766468975192909,
|
| 1992 |
+
"train_speed(iter/s)": 0.0682
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.6824146981627297,
|
| 1996 |
+
"grad_norm": 0.6083526490876717,
|
| 1997 |
+
"learning_rate": 2.5107529432348664e-06,
|
| 1998 |
+
"loss": 1.6035924911499024,
|
| 1999 |
+
"memory(GiB)": 69.47,
|
| 2000 |
+
"step": 910,
|
| 2001 |
+
"token_acc": 0.5805827292561053,
|
| 2002 |
+
"train_speed(iter/s)": 0.06825
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 0.6861642294713161,
|
| 2006 |
+
"grad_norm": 0.6382000976807077,
|
| 2007 |
+
"learning_rate": 2.4571428351061872e-06,
|
| 2008 |
+
"loss": 1.610848617553711,
|
| 2009 |
+
"memory(GiB)": 69.47,
|
| 2010 |
+
"step": 915,
|
| 2011 |
+
"token_acc": 0.6075299795955721,
|
| 2012 |
+
"train_speed(iter/s)": 0.068299
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.6899137607799025,
|
| 2016 |
+
"grad_norm": 0.6269897687508513,
|
| 2017 |
+
"learning_rate": 2.403924188190247e-06,
|
| 2018 |
+
"loss": 1.6187686920166016,
|
| 2019 |
+
"memory(GiB)": 69.47,
|
| 2020 |
+
"step": 920,
|
| 2021 |
+
"token_acc": 0.5857751413329086,
|
| 2022 |
+
"train_speed(iter/s)": 0.068347
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 0.693663292088489,
|
| 2026 |
+
"grad_norm": 0.5561791127288412,
|
| 2027 |
+
"learning_rate": 2.3511051952539703e-06,
|
| 2028 |
+
"loss": 1.6138473510742188,
|
| 2029 |
+
"memory(GiB)": 69.47,
|
| 2030 |
+
"step": 925,
|
| 2031 |
+
"token_acc": 0.5752952655172179,
|
| 2032 |
+
"train_speed(iter/s)": 0.068393
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6974128233970753,
|
| 2036 |
+
"grad_norm": 0.5157680455833342,
|
| 2037 |
+
"learning_rate": 2.2986939875393753e-06,
|
| 2038 |
+
"loss": 1.6303401947021485,
|
| 2039 |
+
"memory(GiB)": 69.47,
|
| 2040 |
+
"step": 930,
|
| 2041 |
+
"token_acc": 0.5684129567444703,
|
| 2042 |
+
"train_speed(iter/s)": 0.068439
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 0.7011623547056618,
|
| 2046 |
+
"grad_norm": 0.4944152577486627,
|
| 2047 |
+
"learning_rate": 2.246698633511813e-06,
|
| 2048 |
+
"loss": 1.6404861450195312,
|
| 2049 |
+
"memory(GiB)": 69.47,
|
| 2050 |
+
"step": 935,
|
| 2051 |
+
"token_acc": 0.6276472253680634,
|
| 2052 |
+
"train_speed(iter/s)": 0.068486
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.7049118860142483,
|
| 2056 |
+
"grad_norm": 0.5307892695648406,
|
| 2057 |
+
"learning_rate": 2.1951271376178708e-06,
|
| 2058 |
+
"loss": 1.6242366790771485,
|
| 2059 |
+
"memory(GiB)": 69.47,
|
| 2060 |
+
"step": 940,
|
| 2061 |
+
"token_acc": 0.5880176396748096,
|
| 2062 |
+
"train_speed(iter/s)": 0.068532
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.7086614173228346,
|
| 2066 |
+
"grad_norm": 0.5381438614462183,
|
| 2067 |
+
"learning_rate": 2.143987439053111e-06,
|
| 2068 |
+
"loss": 1.6123714447021484,
|
| 2069 |
+
"memory(GiB)": 69.47,
|
| 2070 |
+
"step": 945,
|
| 2071 |
+
"token_acc": 0.6045954904448173,
|
| 2072 |
+
"train_speed(iter/s)": 0.068578
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 0.7124109486314211,
|
| 2076 |
+
"grad_norm": 0.6061036399917892,
|
| 2077 |
+
"learning_rate": 2.0932874105398774e-06,
|
| 2078 |
+
"loss": 1.6138986587524413,
|
| 2079 |
+
"memory(GiB)": 69.47,
|
| 2080 |
+
"step": 950,
|
| 2081 |
+
"token_acc": 0.6026358079705822,
|
| 2082 |
+
"train_speed(iter/s)": 0.068622
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.7124109486314211,
|
| 2086 |
+
"eval_loss": 1.6221652030944824,
|
| 2087 |
+
"eval_runtime": 42.632,
|
| 2088 |
+
"eval_samples_per_second": 60.612,
|
| 2089 |
+
"eval_steps_per_second": 1.267,
|
| 2090 |
+
"eval_token_acc": 0.58834860073544,
|
| 2091 |
+
"step": 950
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 0.7161604799400075,
|
| 2095 |
+
"grad_norm": 0.5014848788091244,
|
| 2096 |
+
"learning_rate": 2.043034857115323e-06,
|
| 2097 |
+
"loss": 1.6557624816894532,
|
| 2098 |
+
"memory(GiB)": 69.47,
|
| 2099 |
+
"step": 955,
|
| 2100 |
+
"token_acc": 0.5814458131233236,
|
| 2101 |
+
"train_speed(iter/s)": 0.068219
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.7199100112485939,
|
| 2105 |
+
"grad_norm": 0.5129920621567414,
|
| 2106 |
+
"learning_rate": 1.9932375149298628e-06,
|
| 2107 |
+
"loss": 1.621377944946289,
|
| 2108 |
+
"memory(GiB)": 69.47,
|
| 2109 |
+
"step": 960,
|
| 2110 |
+
"token_acc": 0.6076489727869205,
|
| 2111 |
+
"train_speed(iter/s)": 0.068261
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.7236595425571803,
|
| 2115 |
+
"grad_norm": 0.506502880183358,
|
| 2116 |
+
"learning_rate": 1.9439030500562243e-06,
|
| 2117 |
+
"loss": 1.6196178436279296,
|
| 2118 |
+
"memory(GiB)": 69.47,
|
| 2119 |
+
"step": 965,
|
| 2120 |
+
"token_acc": 0.5846923847966841,
|
| 2121 |
+
"train_speed(iter/s)": 0.068305
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 0.7274090738657668,
|
| 2125 |
+
"grad_norm": 0.5526154345489055,
|
| 2126 |
+
"learning_rate": 1.895039057309293e-06,
|
| 2127 |
+
"loss": 1.6083795547485351,
|
| 2128 |
+
"memory(GiB)": 69.47,
|
| 2129 |
+
"step": 970,
|
| 2130 |
+
"token_acc": 0.5995856662175794,
|
| 2131 |
+
"train_speed(iter/s)": 0.068352
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.7311586051743532,
|
| 2135 |
+
"grad_norm": 0.5152057399852353,
|
| 2136 |
+
"learning_rate": 1.846653059076925e-06,
|
| 2137 |
+
"loss": 1.6139591217041016,
|
| 2138 |
+
"memory(GiB)": 69.47,
|
| 2139 |
+
"step": 975,
|
| 2140 |
+
"token_acc": 0.5797222490647304,
|
| 2141 |
+
"train_speed(iter/s)": 0.068398
|
| 2142 |
+
},
|
| 2143 |
+
{
|
| 2144 |
+
"epoch": 0.7349081364829396,
|
| 2145 |
+
"grad_norm": 0.4824927120599927,
|
| 2146 |
+
"learning_rate": 1.7987525041619147e-06,
|
| 2147 |
+
"loss": 1.6110733032226563,
|
| 2148 |
+
"memory(GiB)": 69.47,
|
| 2149 |
+
"step": 980,
|
| 2150 |
+
"token_acc": 0.5872091501557306,
|
| 2151 |
+
"train_speed(iter/s)": 0.068442
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7386576677915261,
|
| 2155 |
+
"grad_norm": 0.5019512033941552,
|
| 2156 |
+
"learning_rate": 1.7513447666352752e-06,
|
| 2157 |
+
"loss": 1.6114360809326171,
|
| 2158 |
+
"memory(GiB)": 69.47,
|
| 2159 |
+
"step": 985,
|
| 2160 |
+
"token_acc": 0.5956581172217651,
|
| 2161 |
+
"train_speed(iter/s)": 0.068486
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 0.7424071991001124,
|
| 2165 |
+
"grad_norm": 0.47021109811395784,
|
| 2166 |
+
"learning_rate": 1.7044371447010483e-06,
|
| 2167 |
+
"loss": 1.6211065292358398,
|
| 2168 |
+
"memory(GiB)": 69.47,
|
| 2169 |
+
"step": 990,
|
| 2170 |
+
"token_acc": 0.569245869458531,
|
| 2171 |
+
"train_speed(iter/s)": 0.068527
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.7461567304086989,
|
| 2175 |
+
"grad_norm": 0.5121836319300844,
|
| 2176 |
+
"learning_rate": 1.6580368595727586e-06,
|
| 2177 |
+
"loss": 1.6363491058349608,
|
| 2178 |
+
"memory(GiB)": 69.47,
|
| 2179 |
+
"step": 995,
|
| 2180 |
+
"token_acc": 0.5638876766133871,
|
| 2181 |
+
"train_speed(iter/s)": 0.06857
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 0.7499062617172854,
|
| 2185 |
+
"grad_norm": 0.47784184287569526,
|
| 2186 |
+
"learning_rate": 1.6121510543617668e-06,
|
| 2187 |
+
"loss": 1.642209243774414,
|
| 2188 |
+
"memory(GiB)": 69.47,
|
| 2189 |
+
"step": 1000,
|
| 2190 |
+
"token_acc": 0.5950520124176243,
|
| 2191 |
+
"train_speed(iter/s)": 0.068612
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 0.7499062617172854,
|
| 2195 |
+
"eval_loss": 1.6193331480026245,
|
| 2196 |
+
"eval_runtime": 42.5704,
|
| 2197 |
+
"eval_samples_per_second": 60.7,
|
| 2198 |
+
"eval_steps_per_second": 1.268,
|
| 2199 |
+
"eval_token_acc": 0.589054532607981,
|
| 2200 |
+
"step": 1000
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7536557930258717,
|
| 2204 |
+
"grad_norm": 0.5585585189121498,
|
| 2205 |
+
"learning_rate": 1.566786792977597e-06,
|
| 2206 |
+
"loss": 1.6204872131347656,
|
| 2207 |
+
"memory(GiB)": 69.47,
|
| 2208 |
+
"step": 1005,
|
| 2209 |
+
"token_acc": 0.5842172093073306,
|
| 2210 |
+
"train_speed(iter/s)": 0.068226
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 0.7574053243344582,
|
| 2214 |
+
"grad_norm": 0.48263411156016095,
|
| 2215 |
+
"learning_rate": 1.5219510590404973e-06,
|
| 2216 |
+
"loss": 1.6084114074707032,
|
| 2217 |
+
"memory(GiB)": 69.47,
|
| 2218 |
+
"step": 1010,
|
| 2219 |
+
"token_acc": 0.6060496626072535,
|
| 2220 |
+
"train_speed(iter/s)": 0.068269
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 0.7611548556430446,
|
| 2224 |
+
"grad_norm": 0.4497762254615072,
|
| 2225 |
+
"learning_rate": 1.4776507548063319e-06,
|
| 2226 |
+
"loss": 1.599934959411621,
|
| 2227 |
+
"memory(GiB)": 69.47,
|
| 2228 |
+
"step": 1015,
|
| 2229 |
+
"token_acc": 0.5598463390701008,
|
| 2230 |
+
"train_speed(iter/s)": 0.068312
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.7649043869516311,
|
| 2234 |
+
"grad_norm": 0.5283692793491782,
|
| 2235 |
+
"learning_rate": 1.4338927001040154e-06,
|
| 2236 |
+
"loss": 1.6040872573852538,
|
| 2237 |
+
"memory(GiB)": 69.47,
|
| 2238 |
+
"step": 1020,
|
| 2239 |
+
"token_acc": 0.5726066007732008,
|
| 2240 |
+
"train_speed(iter/s)": 0.068356
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 0.7686539182602175,
|
| 2244 |
+
"grad_norm": 0.5067077833706963,
|
| 2245 |
+
"learning_rate": 1.3906836312856304e-06,
|
| 2246 |
+
"loss": 1.6202468872070312,
|
| 2247 |
+
"memory(GiB)": 69.47,
|
| 2248 |
+
"step": 1025,
|
| 2249 |
+
"token_acc": 0.5568129330254041,
|
| 2250 |
+
"train_speed(iter/s)": 0.068396
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.7724034495688039,
|
| 2254 |
+
"grad_norm": 0.5090445248471341,
|
| 2255 |
+
"learning_rate": 1.3480302001894007e-06,
|
| 2256 |
+
"loss": 1.6604093551635741,
|
| 2257 |
+
"memory(GiB)": 69.47,
|
| 2258 |
+
"step": 1030,
|
| 2259 |
+
"token_acc": 0.5570990237099024,
|
| 2260 |
+
"train_speed(iter/s)": 0.06844
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 0.7761529808773904,
|
| 2264 |
+
"grad_norm": 0.5108687306008184,
|
| 2265 |
+
"learning_rate": 1.3059389731156635e-06,
|
| 2266 |
+
"loss": 1.6176353454589845,
|
| 2267 |
+
"memory(GiB)": 69.47,
|
| 2268 |
+
"step": 1035,
|
| 2269 |
+
"token_acc": 0.6020968399214611,
|
| 2270 |
+
"train_speed(iter/s)": 0.06848
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7799025121859767,
|
| 2274 |
+
"grad_norm": 0.5073039466614888,
|
| 2275 |
+
"learning_rate": 1.2644164298160278e-06,
|
| 2276 |
+
"loss": 1.6030658721923827,
|
| 2277 |
+
"memory(GiB)": 69.47,
|
| 2278 |
+
"step": 1040,
|
| 2279 |
+
"token_acc": 0.5653390050744758,
|
| 2280 |
+
"train_speed(iter/s)": 0.068521
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 0.7836520434945632,
|
| 2284 |
+
"grad_norm": 0.5102888558915185,
|
| 2285 |
+
"learning_rate": 1.2234689624958307e-06,
|
| 2286 |
+
"loss": 1.6093769073486328,
|
| 2287 |
+
"memory(GiB)": 69.47,
|
| 2288 |
+
"step": 1045,
|
| 2289 |
+
"token_acc": 0.5926654563459317,
|
| 2290 |
+
"train_speed(iter/s)": 0.06856
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 0.7874015748031497,
|
| 2294 |
+
"grad_norm": 0.5211478805411307,
|
| 2295 |
+
"learning_rate": 1.1831028748301071e-06,
|
| 2296 |
+
"loss": 1.6387947082519532,
|
| 2297 |
+
"memory(GiB)": 69.47,
|
| 2298 |
+
"step": 1050,
|
| 2299 |
+
"token_acc": 0.5662679083094556,
|
| 2300 |
+
"train_speed(iter/s)": 0.068597
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.7874015748031497,
|
| 2304 |
+
"eval_loss": 1.6168898344039917,
|
| 2305 |
+
"eval_runtime": 43.2438,
|
| 2306 |
+
"eval_samples_per_second": 59.754,
|
| 2307 |
+
"eval_steps_per_second": 1.249,
|
| 2308 |
+
"eval_token_acc": 0.5897800736992036,
|
| 2309 |
+
"step": 1050
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 0.791151106111736,
|
| 2313 |
+
"grad_norm": 0.49447345448344293,
|
| 2314 |
+
"learning_rate": 1.14332438099315e-06,
|
| 2315 |
+
"loss": 1.603018569946289,
|
| 2316 |
+
"memory(GiB)": 69.47,
|
| 2317 |
+
"step": 1055,
|
| 2318 |
+
"token_acc": 0.5800278558567661,
|
| 2319 |
+
"train_speed(iter/s)": 0.06823
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7949006374203225,
|
| 2323 |
+
"grad_norm": 0.4741061460173634,
|
| 2324 |
+
"learning_rate": 1.1041396047018793e-06,
|
| 2325 |
+
"loss": 1.6282894134521484,
|
| 2326 |
+
"memory(GiB)": 69.47,
|
| 2327 |
+
"step": 1060,
|
| 2328 |
+
"token_acc": 0.5936924273646574,
|
| 2329 |
+
"train_speed(iter/s)": 0.068272
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 0.7986501687289089,
|
| 2333 |
+
"grad_norm": 0.5034300959507061,
|
| 2334 |
+
"learning_rate": 1.065554578273113e-06,
|
| 2335 |
+
"loss": 1.606418991088867,
|
| 2336 |
+
"memory(GiB)": 69.47,
|
| 2337 |
+
"step": 1065,
|
| 2338 |
+
"token_acc": 0.5734419887017269,
|
| 2339 |
+
"train_speed(iter/s)": 0.068313
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 0.8023997000374953,
|
| 2343 |
+
"grad_norm": 0.45462008885394545,
|
| 2344 |
+
"learning_rate": 1.0275752416949291e-06,
|
| 2345 |
+
"loss": 1.6223371505737305,
|
| 2346 |
+
"memory(GiB)": 69.47,
|
| 2347 |
+
"step": 1070,
|
| 2348 |
+
"token_acc": 0.591089095797518,
|
| 2349 |
+
"train_speed(iter/s)": 0.068354
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 0.8061492313460817,
|
| 2353 |
+
"grad_norm": 0.5314761160291909,
|
| 2354 |
+
"learning_rate": 9.902074417122233e-07,
|
| 2355 |
+
"loss": 1.603017807006836,
|
| 2356 |
+
"memory(GiB)": 69.47,
|
| 2357 |
+
"step": 1075,
|
| 2358 |
+
"token_acc": 0.5886109380502208,
|
| 2359 |
+
"train_speed(iter/s)": 0.068393
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 0.8098987626546682,
|
| 2363 |
+
"grad_norm": 0.5248943174736966,
|
| 2364 |
+
"learning_rate": 9.534569309266395e-07,
|
| 2365 |
+
"loss": 1.5781710624694825,
|
| 2366 |
+
"memory(GiB)": 69.47,
|
| 2367 |
+
"step": 1080,
|
| 2368 |
+
"token_acc": 0.5995606920789402,
|
| 2369 |
+
"train_speed(iter/s)": 0.068431
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.8136482939632546,
|
| 2373 |
+
"grad_norm": 0.515744567059207,
|
| 2374 |
+
"learning_rate": 9.173293669109728e-07,
|
| 2375 |
+
"loss": 1.6221351623535156,
|
| 2376 |
+
"memory(GiB)": 69.47,
|
| 2377 |
+
"step": 1085,
|
| 2378 |
+
"token_acc": 0.5510625879767913,
|
| 2379 |
+
"train_speed(iter/s)": 0.068471
|
| 2380 |
+
},
|
| 2381 |
+
{
|
| 2382 |
+
"epoch": 0.817397825271841,
|
| 2383 |
+
"grad_norm": 0.4846549490970001,
|
| 2384 |
+
"learning_rate": 8.818303113382176e-07,
|
| 2385 |
+
"loss": 1.5834909439086915,
|
| 2386 |
+
"memory(GiB)": 69.47,
|
| 2387 |
+
"step": 1090,
|
| 2388 |
+
"token_acc": 0.5919089498189343,
|
| 2389 |
+
"train_speed(iter/s)": 0.06851
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8211473565804275,
|
| 2393 |
+
"grad_norm": 0.4513353191128082,
|
| 2394 |
+
"learning_rate": 8.46965229125371e-07,
|
| 2395 |
+
"loss": 1.6127058029174806,
|
| 2396 |
+
"memory(GiB)": 69.47,
|
| 2397 |
+
"step": 1095,
|
| 2398 |
+
"token_acc": 0.6122484158559205,
|
| 2399 |
+
"train_speed(iter/s)": 0.068551
|
| 2400 |
+
},
|
| 2401 |
+
{
|
| 2402 |
+
"epoch": 0.8248968878890138,
|
| 2403 |
+
"grad_norm": 0.4544425755483677,
|
| 2404 |
+
"learning_rate": 8.127394875921401e-07,
|
| 2405 |
+
"loss": 1.607330322265625,
|
| 2406 |
+
"memory(GiB)": 69.47,
|
| 2407 |
+
"step": 1100,
|
| 2408 |
+
"token_acc": 0.6195696090184728,
|
| 2409 |
+
"train_speed(iter/s)": 0.068591
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 0.8248968878890138,
|
| 2413 |
+
"eval_loss": 1.6152210235595703,
|
| 2414 |
+
"eval_runtime": 42.4585,
|
| 2415 |
+
"eval_samples_per_second": 60.859,
|
| 2416 |
+
"eval_steps_per_second": 1.272,
|
| 2417 |
+
"eval_token_acc": 0.5902661096809664,
|
| 2418 |
+
"step": 1100
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.8286464191976003,
|
| 2422 |
+
"grad_norm": 0.4699375444459336,
|
| 2423 |
+
"learning_rate": 7.791583556346577e-07,
|
| 2424 |
+
"loss": 1.6486019134521483,
|
| 2425 |
+
"memory(GiB)": 69.47,
|
| 2426 |
+
"step": 1105,
|
| 2427 |
+
"token_acc": 0.5841665130220646,
|
| 2428 |
+
"train_speed(iter/s)": 0.068244
|
| 2429 |
+
},
|
| 2430 |
+
{
|
| 2431 |
+
"epoch": 0.8323959505061868,
|
| 2432 |
+
"grad_norm": 0.4655325816892184,
|
| 2433 |
+
"learning_rate": 7.46227002914367e-07,
|
| 2434 |
+
"loss": 1.6185653686523438,
|
| 2435 |
+
"memory(GiB)": 69.47,
|
| 2436 |
+
"step": 1110,
|
| 2437 |
+
"token_acc": 0.5508876719131502,
|
| 2438 |
+
"train_speed(iter/s)": 0.068285
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8361454818147731,
|
| 2442 |
+
"grad_norm": 0.443815549611747,
|
| 2443 |
+
"learning_rate": 7.139504990621754e-07,
|
| 2444 |
+
"loss": 1.6039731979370118,
|
| 2445 |
+
"memory(GiB)": 69.47,
|
| 2446 |
+
"step": 1115,
|
| 2447 |
+
"token_acc": 0.5742345033607169,
|
| 2448 |
+
"train_speed(iter/s)": 0.068324
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 0.8398950131233596,
|
| 2452 |
+
"grad_norm": 0.47193117910563975,
|
| 2453 |
+
"learning_rate": 6.82333812898005e-07,
|
| 2454 |
+
"loss": 1.6046607971191407,
|
| 2455 |
+
"memory(GiB)": 69.47,
|
| 2456 |
+
"step": 1120,
|
| 2457 |
+
"token_acc": 0.5756778822186418,
|
| 2458 |
+
"train_speed(iter/s)": 0.068363
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 0.843644544431946,
|
| 2462 |
+
"grad_norm": 0.5016745684818081,
|
| 2463 |
+
"learning_rate": 6.513818116658671e-07,
|
| 2464 |
+
"loss": 1.6025604248046874,
|
| 2465 |
+
"memory(GiB)": 69.47,
|
| 2466 |
+
"step": 1125,
|
| 2467 |
+
"token_acc": 0.5735500039494754,
|
| 2468 |
+
"train_speed(iter/s)": 0.0684
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 0.8473940757405324,
|
| 2472 |
+
"grad_norm": 0.5220705510390201,
|
| 2473 |
+
"learning_rate": 6.210992602845722e-07,
|
| 2474 |
+
"loss": 1.5999751091003418,
|
| 2475 |
+
"memory(GiB)": 69.47,
|
| 2476 |
+
"step": 1130,
|
| 2477 |
+
"token_acc": 0.5666828322017459,
|
| 2478 |
+
"train_speed(iter/s)": 0.068438
|
| 2479 |
+
},
|
| 2480 |
+
{
|
| 2481 |
+
"epoch": 0.8511436070491188,
|
| 2482 |
+
"grad_norm": 0.4658872880073119,
|
| 2483 |
+
"learning_rate": 5.914908206141956e-07,
|
| 2484 |
+
"loss": 1.6399084091186524,
|
| 2485 |
+
"memory(GiB)": 69.47,
|
| 2486 |
+
"step": 1135,
|
| 2487 |
+
"token_acc": 0.5555669822893787,
|
| 2488 |
+
"train_speed(iter/s)": 0.068473
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.8548931383577053,
|
| 2492 |
+
"grad_norm": 0.49112601151308977,
|
| 2493 |
+
"learning_rate": 5.625610507383988e-07,
|
| 2494 |
+
"loss": 1.6437812805175782,
|
| 2495 |
+
"memory(GiB)": 69.47,
|
| 2496 |
+
"step": 1140,
|
| 2497 |
+
"token_acc": 0.571141520165108,
|
| 2498 |
+
"train_speed(iter/s)": 0.068511
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 0.8586426696662918,
|
| 2502 |
+
"grad_norm": 0.4822913125455344,
|
| 2503 |
+
"learning_rate": 5.343144042627391e-07,
|
| 2504 |
+
"loss": 1.621147918701172,
|
| 2505 |
+
"memory(GiB)": 69.47,
|
| 2506 |
+
"step": 1145,
|
| 2507 |
+
"token_acc": 0.5925554632407973,
|
| 2508 |
+
"train_speed(iter/s)": 0.068547
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8623922009748781,
|
| 2512 |
+
"grad_norm": 0.435848921973967,
|
| 2513 |
+
"learning_rate": 5.06755229629054e-07,
|
| 2514 |
+
"loss": 1.5979487419128418,
|
| 2515 |
+
"memory(GiB)": 69.47,
|
| 2516 |
+
"step": 1150,
|
| 2517 |
+
"token_acc": 0.6014443218658418,
|
| 2518 |
+
"train_speed(iter/s)": 0.068584
|
| 2519 |
+
},
|
| 2520 |
+
{
|
| 2521 |
+
"epoch": 0.8623922009748781,
|
| 2522 |
+
"eval_loss": 1.613835096359253,
|
| 2523 |
+
"eval_runtime": 42.8701,
|
| 2524 |
+
"eval_samples_per_second": 60.275,
|
| 2525 |
+
"eval_steps_per_second": 1.26,
|
| 2526 |
+
"eval_token_acc": 0.5905893996338843,
|
| 2527 |
+
"step": 1150
|
| 2528 |
+
},
|
| 2529 |
+
{
|
| 2530 |
+
"epoch": 0.8661417322834646,
|
| 2531 |
+
"grad_norm": 0.46089404385561655,
|
| 2532 |
+
"learning_rate": 4.798877694460424e-07,
|
| 2533 |
+
"loss": 1.6086740493774414,
|
| 2534 |
+
"memory(GiB)": 69.47,
|
| 2535 |
+
"step": 1155,
|
| 2536 |
+
"token_acc": 0.5848029286936108,
|
| 2537 |
+
"train_speed(iter/s)": 0.068246
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.869891263592051,
|
| 2541 |
+
"grad_norm": 0.42872964695512505,
|
| 2542 |
+
"learning_rate": 4.5371615983612947e-07,
|
| 2543 |
+
"loss": 1.6226417541503906,
|
| 2544 |
+
"memory(GiB)": 69.47,
|
| 2545 |
+
"step": 1160,
|
| 2546 |
+
"token_acc": 0.590493341547794,
|
| 2547 |
+
"train_speed(iter/s)": 0.068282
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"epoch": 0.8736407949006374,
|
| 2551 |
+
"grad_norm": 0.43701593339026695,
|
| 2552 |
+
"learning_rate": 4.282444297987359e-07,
|
| 2553 |
+
"loss": 1.61004638671875,
|
| 2554 |
+
"memory(GiB)": 69.47,
|
| 2555 |
+
"step": 1165,
|
| 2556 |
+
"token_acc": 0.5865000985543102,
|
| 2557 |
+
"train_speed(iter/s)": 0.06832
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8773903262092239,
|
| 2561 |
+
"grad_norm": 0.47705938510535956,
|
| 2562 |
+
"learning_rate": 4.03476500590021e-07,
|
| 2563 |
+
"loss": 1.5912567138671876,
|
| 2564 |
+
"memory(GiB)": 69.47,
|
| 2565 |
+
"step": 1170,
|
| 2566 |
+
"token_acc": 0.612401185770751,
|
| 2567 |
+
"train_speed(iter/s)": 0.068355
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 0.8811398575178103,
|
| 2571 |
+
"grad_norm": 0.4892608214978031,
|
| 2572 |
+
"learning_rate": 3.794161851192374e-07,
|
| 2573 |
+
"loss": 1.61513671875,
|
| 2574 |
+
"memory(GiB)": 69.47,
|
| 2575 |
+
"step": 1175,
|
| 2576 |
+
"token_acc": 0.5986275418415744,
|
| 2577 |
+
"train_speed(iter/s)": 0.068391
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 0.8848893888263967,
|
| 2581 |
+
"grad_norm": 0.4619591737564222,
|
| 2582 |
+
"learning_rate": 3.560671873617405e-07,
|
| 2583 |
+
"loss": 1.605311965942383,
|
| 2584 |
+
"memory(GiB)": 69.47,
|
| 2585 |
+
"step": 1180,
|
| 2586 |
+
"token_acc": 0.5807709544290082,
|
| 2587 |
+
"train_speed(iter/s)": 0.068428
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 0.8886389201349831,
|
| 2591 |
+
"grad_norm": 0.4929784580597144,
|
| 2592 |
+
"learning_rate": 3.334331017887837e-07,
|
| 2593 |
+
"loss": 1.6141300201416016,
|
| 2594 |
+
"memory(GiB)": 69.47,
|
| 2595 |
+
"step": 1185,
|
| 2596 |
+
"token_acc": 0.5794582531848332,
|
| 2597 |
+
"train_speed(iter/s)": 0.068462
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"epoch": 0.8923884514435696,
|
| 2601 |
+
"grad_norm": 0.4359896715792382,
|
| 2602 |
+
"learning_rate": 3.1151741281416236e-07,
|
| 2603 |
+
"loss": 1.6315603256225586,
|
| 2604 |
+
"memory(GiB)": 69.47,
|
| 2605 |
+
"step": 1190,
|
| 2606 |
+
"token_acc": 0.6129624092848794,
|
| 2607 |
+
"train_speed(iter/s)": 0.068499
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.896137982752156,
|
| 2611 |
+
"grad_norm": 0.4199860900705256,
|
| 2612 |
+
"learning_rate": 2.903234942578081e-07,
|
| 2613 |
+
"loss": 1.6342578887939454,
|
| 2614 |
+
"memory(GiB)": 69.47,
|
| 2615 |
+
"step": 1195,
|
| 2616 |
+
"token_acc": 0.5895935653783188,
|
| 2617 |
+
"train_speed(iter/s)": 0.068535
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 0.8998875140607424,
|
| 2621 |
+
"grad_norm": 0.42147511115613456,
|
| 2622 |
+
"learning_rate": 2.698546088264009e-07,
|
| 2623 |
+
"loss": 1.5956247329711915,
|
| 2624 |
+
"memory(GiB)": 69.47,
|
| 2625 |
+
"step": 1200,
|
| 2626 |
+
"token_acc": 0.6071841830830988,
|
| 2627 |
+
"train_speed(iter/s)": 0.068572
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8998875140607424,
|
| 2631 |
+
"eval_loss": 1.6130276918411255,
|
| 2632 |
+
"eval_runtime": 42.5246,
|
| 2633 |
+
"eval_samples_per_second": 60.765,
|
| 2634 |
+
"eval_steps_per_second": 1.27,
|
| 2635 |
+
"eval_token_acc": 0.5908747294950227,
|
| 2636 |
+
"step": 1200
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 0.9036370453693289,
|
| 2640 |
+
"grad_norm": 0.42547933706404784,
|
| 2641 |
+
"learning_rate": 2.5011390761109424e-07,
|
| 2642 |
+
"loss": 1.6252113342285157,
|
| 2643 |
+
"memory(GiB)": 69.47,
|
| 2644 |
+
"step": 1205,
|
| 2645 |
+
"token_acc": 0.591078917410982,
|
| 2646 |
+
"train_speed(iter/s)": 0.068254
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 0.9073865766779152,
|
| 2650 |
+
"grad_norm": 0.4405736679459605,
|
| 2651 |
+
"learning_rate": 2.3110442960241507e-07,
|
| 2652 |
+
"loss": 1.6035770416259765,
|
| 2653 |
+
"memory(GiB)": 69.47,
|
| 2654 |
+
"step": 1210,
|
| 2655 |
+
"token_acc": 0.5825880077861605,
|
| 2656 |
+
"train_speed(iter/s)": 0.068289
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.9111361079865017,
|
| 2660 |
+
"grad_norm": 0.44012672892851296,
|
| 2661 |
+
"learning_rate": 2.1282910122243038e-07,
|
| 2662 |
+
"loss": 1.59132719039917,
|
| 2663 |
+
"memory(GiB)": 69.47,
|
| 2664 |
+
"step": 1215,
|
| 2665 |
+
"token_acc": 0.5708482310549674,
|
| 2666 |
+
"train_speed(iter/s)": 0.068324
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 0.9148856392950881,
|
| 2670 |
+
"grad_norm": 0.436144469439609,
|
| 2671 |
+
"learning_rate": 1.9529073587423008e-07,
|
| 2672 |
+
"loss": 1.6214092254638672,
|
| 2673 |
+
"memory(GiB)": 69.47,
|
| 2674 |
+
"step": 1220,
|
| 2675 |
+
"token_acc": 0.5917138980903125,
|
| 2676 |
+
"train_speed(iter/s)": 0.06836
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9186351706036745,
|
| 2680 |
+
"grad_norm": 0.45595845007022057,
|
| 2681 |
+
"learning_rate": 1.7849203350882415e-07,
|
| 2682 |
+
"loss": 1.595216941833496,
|
| 2683 |
+
"memory(GiB)": 69.47,
|
| 2684 |
+
"step": 1225,
|
| 2685 |
+
"token_acc": 0.5995736316066336,
|
| 2686 |
+
"train_speed(iter/s)": 0.068394
|
| 2687 |
+
},
|
| 2688 |
+
{
|
| 2689 |
+
"epoch": 0.922384701912261,
|
| 2690 |
+
"grad_norm": 0.4354484890078837,
|
| 2691 |
+
"learning_rate": 1.6243558020949345e-07,
|
| 2692 |
+
"loss": 1.6196565628051758,
|
| 2693 |
+
"memory(GiB)": 69.47,
|
| 2694 |
+
"step": 1230,
|
| 2695 |
+
"token_acc": 0.5604222422568806,
|
| 2696 |
+
"train_speed(iter/s)": 0.068427
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 0.9261342332208474,
|
| 2700 |
+
"grad_norm": 0.43528308145508676,
|
| 2701 |
+
"learning_rate": 1.471238477936765e-07,
|
| 2702 |
+
"loss": 1.5957447052001954,
|
| 2703 |
+
"memory(GiB)": 69.47,
|
| 2704 |
+
"step": 1235,
|
| 2705 |
+
"token_acc": 0.6027920480891877,
|
| 2706 |
+
"train_speed(iter/s)": 0.068463
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 0.9298837645294338,
|
| 2710 |
+
"grad_norm": 0.4143657458995663,
|
| 2711 |
+
"learning_rate": 1.3255919343244105e-07,
|
| 2712 |
+
"loss": 1.5905607223510743,
|
| 2713 |
+
"memory(GiB)": 69.47,
|
| 2714 |
+
"step": 1240,
|
| 2715 |
+
"token_acc": 0.5880818473790528,
|
| 2716 |
+
"train_speed(iter/s)": 0.068497
|
| 2717 |
+
},
|
| 2718 |
+
{
|
| 2719 |
+
"epoch": 0.9336332958380202,
|
| 2720 |
+
"grad_norm": 0.4299846489431254,
|
| 2721 |
+
"learning_rate": 1.1874385928761112e-07,
|
| 2722 |
+
"loss": 1.598089599609375,
|
| 2723 |
+
"memory(GiB)": 69.47,
|
| 2724 |
+
"step": 1245,
|
| 2725 |
+
"token_acc": 0.5968487409559421,
|
| 2726 |
+
"train_speed(iter/s)": 0.068531
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.9373828271466067,
|
| 2730 |
+
"grad_norm": 0.428143089845641,
|
| 2731 |
+
"learning_rate": 1.0567997216659576e-07,
|
| 2732 |
+
"loss": 1.5996816635131836,
|
| 2733 |
+
"memory(GiB)": 69.47,
|
| 2734 |
+
"step": 1250,
|
| 2735 |
+
"token_acc": 0.5944731717144583,
|
| 2736 |
+
"train_speed(iter/s)": 0.068566
|
| 2737 |
+
},
|
| 2738 |
+
{
|
| 2739 |
+
"epoch": 0.9373828271466067,
|
| 2740 |
+
"eval_loss": 1.6125571727752686,
|
| 2741 |
+
"eval_runtime": 42.1918,
|
| 2742 |
+
"eval_samples_per_second": 61.244,
|
| 2743 |
+
"eval_steps_per_second": 1.28,
|
| 2744 |
+
"eval_token_acc": 0.5909382759470073,
|
| 2745 |
+
"step": 1250
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 0.941132358455193,
|
| 2749 |
+
"grad_norm": 0.4502688717392391,
|
| 2750 |
+
"learning_rate": 9.336954319497716e-08,
|
| 2751 |
+
"loss": 1.5971858978271485,
|
| 2752 |
+
"memory(GiB)": 69.47,
|
| 2753 |
+
"step": 1255,
|
| 2754 |
+
"token_acc": 0.5895391353585584,
|
| 2755 |
+
"train_speed(iter/s)": 0.068262
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 0.9448818897637795,
|
| 2759 |
+
"grad_norm": 0.4344596223428801,
|
| 2760 |
+
"learning_rate": 8.181446750690658e-08,
|
| 2761 |
+
"loss": 1.6137876510620117,
|
| 2762 |
+
"memory(GiB)": 69.47,
|
| 2763 |
+
"step": 1260,
|
| 2764 |
+
"token_acc": 0.5668742586002372,
|
| 2765 |
+
"train_speed(iter/s)": 0.068299
|
| 2766 |
+
},
|
| 2767 |
+
{
|
| 2768 |
+
"epoch": 0.948631421072366,
|
| 2769 |
+
"grad_norm": 0.45590977674430777,
|
| 2770 |
+
"learning_rate": 7.101652395335779e-08,
|
| 2771 |
+
"loss": 1.6049671173095703,
|
| 2772 |
+
"memory(GiB)": 69.47,
|
| 2773 |
+
"step": 1265,
|
| 2774 |
+
"token_acc": 0.6246026139173437,
|
| 2775 |
+
"train_speed(iter/s)": 0.068333
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.9523809523809523,
|
| 2779 |
+
"grad_norm": 0.4295730317443299,
|
| 2780 |
+
"learning_rate": 6.097737482827826e-08,
|
| 2781 |
+
"loss": 1.60078125,
|
| 2782 |
+
"memory(GiB)": 69.47,
|
| 2783 |
+
"step": 1270,
|
| 2784 |
+
"token_acc": 0.6323105986580678,
|
| 2785 |
+
"train_speed(iter/s)": 0.068365
|
| 2786 |
+
},
|
| 2787 |
+
{
|
| 2788 |
+
"epoch": 0.9561304836895388,
|
| 2789 |
+
"grad_norm": 0.4265511315819114,
|
| 2790 |
+
"learning_rate": 5.169856561269171e-08,
|
| 2791 |
+
"loss": 1.6046745300292968,
|
| 2792 |
+
"memory(GiB)": 69.47,
|
| 2793 |
+
"step": 1275,
|
| 2794 |
+
"token_acc": 0.5845875477479601,
|
| 2795 |
+
"train_speed(iter/s)": 0.0684
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9598800149981253,
|
| 2799 |
+
"grad_norm": 0.44161345011158165,
|
| 2800 |
+
"learning_rate": 4.3181524736773394e-08,
|
| 2801 |
+
"loss": 1.6101186752319336,
|
| 2802 |
+
"memory(GiB)": 69.47,
|
| 2803 |
+
"step": 1280,
|
| 2804 |
+
"token_acc": 0.6039562611800572,
|
| 2805 |
+
"train_speed(iter/s)": 0.068433
|
| 2806 |
+
},
|
| 2807 |
+
{
|
| 2808 |
+
"epoch": 0.9636295463067117,
|
| 2809 |
+
"grad_norm": 0.4040990948400036,
|
| 2810 |
+
"learning_rate": 3.54275633599549e-08,
|
| 2811 |
+
"loss": 1.5941335678100585,
|
| 2812 |
+
"memory(GiB)": 69.47,
|
| 2813 |
+
"step": 1285,
|
| 2814 |
+
"token_acc": 0.5708919115077552,
|
| 2815 |
+
"train_speed(iter/s)": 0.068466
|
| 2816 |
+
},
|
| 2817 |
+
{
|
| 2818 |
+
"epoch": 0.9673790776152981,
|
| 2819 |
+
"grad_norm": 0.4137779647306755,
|
| 2820 |
+
"learning_rate": 2.8437875169070595e-08,
|
| 2821 |
+
"loss": 1.5992546081542969,
|
| 2822 |
+
"memory(GiB)": 69.47,
|
| 2823 |
+
"step": 1290,
|
| 2824 |
+
"token_acc": 0.5816298270170738,
|
| 2825 |
+
"train_speed(iter/s)": 0.068501
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 0.9711286089238845,
|
| 2829 |
+
"grad_norm": 0.414717051427377,
|
| 2830 |
+
"learning_rate": 2.2213536194601314e-08,
|
| 2831 |
+
"loss": 1.6082378387451173,
|
| 2832 |
+
"memory(GiB)": 69.47,
|
| 2833 |
+
"step": 1295,
|
| 2834 |
+
"token_acc": 0.5427460601272402,
|
| 2835 |
+
"train_speed(iter/s)": 0.068535
|
| 2836 |
+
},
|
| 2837 |
+
{
|
| 2838 |
+
"epoch": 0.974878140232471,
|
| 2839 |
+
"grad_norm": 0.4389048585868221,
|
| 2840 |
+
"learning_rate": 1.6755504645021292e-08,
|
| 2841 |
+
"loss": 1.5763816833496094,
|
| 2842 |
+
"memory(GiB)": 69.47,
|
| 2843 |
+
"step": 1300,
|
| 2844 |
+
"token_acc": 0.5868564953189289,
|
| 2845 |
+
"train_speed(iter/s)": 0.068567
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.974878140232471,
|
| 2849 |
+
"eval_loss": 1.6123632192611694,
|
| 2850 |
+
"eval_runtime": 42.645,
|
| 2851 |
+
"eval_samples_per_second": 60.593,
|
| 2852 |
+
"eval_steps_per_second": 1.266,
|
| 2853 |
+
"eval_token_acc": 0.5910751210292512,
|
| 2854 |
+
"step": 1300
|
| 2855 |
+
}
|
| 2856 |
+
],
|
| 2857 |
+
"logging_steps": 5,
|
| 2858 |
+
"max_steps": 1333,
|
| 2859 |
+
"num_input_tokens_seen": 0,
|
| 2860 |
+
"num_train_epochs": 1,
|
| 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 |
+
}
|