Upload sft-017-qwen3-vl-30b-a3b-full-tuning-8k-ckpt-444/trainer_state.json with huggingface_hub
Browse files
sft-017-qwen3-vl-30b-a3b-full-tuning-8k-ckpt-444/trainer_state.json
ADDED
|
@@ -0,0 +1,3142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.8048780487804879,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 444,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0040650406504065045,
|
| 14 |
+
"grad_norm": 42.591163635253906,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 1.2665,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.008130081300813009,
|
| 21 |
+
"grad_norm": 33.325687408447266,
|
| 22 |
+
"learning_rate": 1.25e-06,
|
| 23 |
+
"loss": 1.0069,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.012195121951219513,
|
| 28 |
+
"grad_norm": 35.02531433105469,
|
| 29 |
+
"learning_rate": 2.5e-06,
|
| 30 |
+
"loss": 1.0689,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.016260162601626018,
|
| 35 |
+
"grad_norm": 30.56825065612793,
|
| 36 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 37 |
+
"loss": 0.9288,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02032520325203252,
|
| 42 |
+
"grad_norm": 15.84094524383545,
|
| 43 |
+
"learning_rate": 5e-06,
|
| 44 |
+
"loss": 0.752,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.024390243902439025,
|
| 49 |
+
"grad_norm": 10.455734252929688,
|
| 50 |
+
"learning_rate": 6.25e-06,
|
| 51 |
+
"loss": 0.5474,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.028455284552845527,
|
| 56 |
+
"grad_norm": 8.037217140197754,
|
| 57 |
+
"learning_rate": 7.500000000000001e-06,
|
| 58 |
+
"loss": 0.4786,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.032520325203252036,
|
| 63 |
+
"grad_norm": 6.807183265686035,
|
| 64 |
+
"learning_rate": 8.750000000000001e-06,
|
| 65 |
+
"loss": 0.4173,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.036585365853658534,
|
| 70 |
+
"grad_norm": 4.072251796722412,
|
| 71 |
+
"learning_rate": 1e-05,
|
| 72 |
+
"loss": 0.3302,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04065040650406504,
|
| 77 |
+
"grad_norm": 6.469480991363525,
|
| 78 |
+
"learning_rate": 9.999953698680951e-06,
|
| 79 |
+
"loss": 0.3781,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.044715447154471545,
|
| 84 |
+
"grad_norm": 5.010468482971191,
|
| 85 |
+
"learning_rate": 9.999814795581329e-06,
|
| 86 |
+
"loss": 0.3314,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.04878048780487805,
|
| 91 |
+
"grad_norm": 3.7235605716705322,
|
| 92 |
+
"learning_rate": 9.999583293273691e-06,
|
| 93 |
+
"loss": 0.3041,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.052845528455284556,
|
| 98 |
+
"grad_norm": 2.6193606853485107,
|
| 99 |
+
"learning_rate": 9.999259196045581e-06,
|
| 100 |
+
"loss": 0.2606,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.056910569105691054,
|
| 105 |
+
"grad_norm": 2.5086917877197266,
|
| 106 |
+
"learning_rate": 9.998842509899456e-06,
|
| 107 |
+
"loss": 0.2816,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.06097560975609756,
|
| 112 |
+
"grad_norm": 2.6058692932128906,
|
| 113 |
+
"learning_rate": 9.998333242552557e-06,
|
| 114 |
+
"loss": 0.2639,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06504065040650407,
|
| 119 |
+
"grad_norm": 2.3764359951019287,
|
| 120 |
+
"learning_rate": 9.997731403436787e-06,
|
| 121 |
+
"loss": 0.2792,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06910569105691057,
|
| 126 |
+
"grad_norm": 2.2673380374908447,
|
| 127 |
+
"learning_rate": 9.997037003698524e-06,
|
| 128 |
+
"loss": 0.2538,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.07317073170731707,
|
| 133 |
+
"grad_norm": 2.9037580490112305,
|
| 134 |
+
"learning_rate": 9.996250056198418e-06,
|
| 135 |
+
"loss": 0.2685,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.07723577235772358,
|
| 140 |
+
"grad_norm": 2.128932476043701,
|
| 141 |
+
"learning_rate": 9.995370575511151e-06,
|
| 142 |
+
"loss": 0.2501,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.08130081300813008,
|
| 147 |
+
"grad_norm": 1.7147555351257324,
|
| 148 |
+
"learning_rate": 9.994398577925169e-06,
|
| 149 |
+
"loss": 0.2122,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08536585365853659,
|
| 154 |
+
"grad_norm": 1.7391780614852905,
|
| 155 |
+
"learning_rate": 9.993334081442382e-06,
|
| 156 |
+
"loss": 0.2364,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.08943089430894309,
|
| 161 |
+
"grad_norm": 1.6465215682983398,
|
| 162 |
+
"learning_rate": 9.992177105777823e-06,
|
| 163 |
+
"loss": 0.2421,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.09349593495934959,
|
| 168 |
+
"grad_norm": 1.687512755393982,
|
| 169 |
+
"learning_rate": 9.990927672359294e-06,
|
| 170 |
+
"loss": 0.239,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.0975609756097561,
|
| 175 |
+
"grad_norm": 1.800450086593628,
|
| 176 |
+
"learning_rate": 9.989585804326963e-06,
|
| 177 |
+
"loss": 0.2244,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.1016260162601626,
|
| 182 |
+
"grad_norm": 1.5062322616577148,
|
| 183 |
+
"learning_rate": 9.98815152653293e-06,
|
| 184 |
+
"loss": 0.2274,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.10569105691056911,
|
| 189 |
+
"grad_norm": 1.8082964420318604,
|
| 190 |
+
"learning_rate": 9.986624865540779e-06,
|
| 191 |
+
"loss": 0.1961,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.10975609756097561,
|
| 196 |
+
"grad_norm": 1.6425156593322754,
|
| 197 |
+
"learning_rate": 9.985005849625075e-06,
|
| 198 |
+
"loss": 0.2158,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.11382113821138211,
|
| 203 |
+
"grad_norm": 1.8497402667999268,
|
| 204 |
+
"learning_rate": 9.983294508770851e-06,
|
| 205 |
+
"loss": 0.2114,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.11788617886178862,
|
| 210 |
+
"grad_norm": 1.188692331314087,
|
| 211 |
+
"learning_rate": 9.98149087467304e-06,
|
| 212 |
+
"loss": 0.2262,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.12195121951219512,
|
| 217 |
+
"grad_norm": 1.5211570262908936,
|
| 218 |
+
"learning_rate": 9.979594980735896e-06,
|
| 219 |
+
"loss": 0.2099,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.12601626016260162,
|
| 224 |
+
"grad_norm": 1.280957818031311,
|
| 225 |
+
"learning_rate": 9.977606862072377e-06,
|
| 226 |
+
"loss": 0.1964,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.13008130081300814,
|
| 231 |
+
"grad_norm": 1.432111382484436,
|
| 232 |
+
"learning_rate": 9.97552655550349e-06,
|
| 233 |
+
"loss": 0.2226,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.13414634146341464,
|
| 238 |
+
"grad_norm": 1.2207342386245728,
|
| 239 |
+
"learning_rate": 9.973354099557606e-06,
|
| 240 |
+
"loss": 0.2022,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.13821138211382114,
|
| 245 |
+
"grad_norm": 1.665354609489441,
|
| 246 |
+
"learning_rate": 9.97108953446976e-06,
|
| 247 |
+
"loss": 0.1951,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.14227642276422764,
|
| 252 |
+
"grad_norm": 1.4010629653930664,
|
| 253 |
+
"learning_rate": 9.968732902180892e-06,
|
| 254 |
+
"loss": 0.2026,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.14634146341463414,
|
| 259 |
+
"grad_norm": 1.3375614881515503,
|
| 260 |
+
"learning_rate": 9.966284246337072e-06,
|
| 261 |
+
"loss": 0.1979,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15040650406504066,
|
| 266 |
+
"grad_norm": 2.4575119018554688,
|
| 267 |
+
"learning_rate": 9.963743612288701e-06,
|
| 268 |
+
"loss": 0.2123,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.15447154471544716,
|
| 273 |
+
"grad_norm": 1.476641058921814,
|
| 274 |
+
"learning_rate": 9.961111047089662e-06,
|
| 275 |
+
"loss": 0.2032,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.15853658536585366,
|
| 280 |
+
"grad_norm": 1.4410269260406494,
|
| 281 |
+
"learning_rate": 9.95838659949645e-06,
|
| 282 |
+
"loss": 0.1878,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.16260162601626016,
|
| 287 |
+
"grad_norm": 1.4006271362304688,
|
| 288 |
+
"learning_rate": 9.955570319967273e-06,
|
| 289 |
+
"loss": 0.1885,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.16666666666666666,
|
| 294 |
+
"grad_norm": 2.408001661300659,
|
| 295 |
+
"learning_rate": 9.952662260661115e-06,
|
| 296 |
+
"loss": 0.1988,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.17073170731707318,
|
| 301 |
+
"grad_norm": 1.7849098443984985,
|
| 302 |
+
"learning_rate": 9.949662475436766e-06,
|
| 303 |
+
"loss": 0.2008,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.17479674796747968,
|
| 308 |
+
"grad_norm": 1.4735466241836548,
|
| 309 |
+
"learning_rate": 9.946571019851831e-06,
|
| 310 |
+
"loss": 0.2022,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.17886178861788618,
|
| 315 |
+
"grad_norm": 1.6618183851242065,
|
| 316 |
+
"learning_rate": 9.943387951161702e-06,
|
| 317 |
+
"loss": 0.2107,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.18292682926829268,
|
| 322 |
+
"grad_norm": 1.3834326267242432,
|
| 323 |
+
"learning_rate": 9.940113328318489e-06,
|
| 324 |
+
"loss": 0.2096,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.18699186991869918,
|
| 329 |
+
"grad_norm": 1.9319173097610474,
|
| 330 |
+
"learning_rate": 9.936747211969932e-06,
|
| 331 |
+
"loss": 0.1868,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.1910569105691057,
|
| 336 |
+
"grad_norm": 1.6279851198196411,
|
| 337 |
+
"learning_rate": 9.933289664458286e-06,
|
| 338 |
+
"loss": 0.183,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.1951219512195122,
|
| 343 |
+
"grad_norm": 1.516767978668213,
|
| 344 |
+
"learning_rate": 9.929740749819152e-06,
|
| 345 |
+
"loss": 0.2063,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.1991869918699187,
|
| 350 |
+
"grad_norm": 1.8240041732788086,
|
| 351 |
+
"learning_rate": 9.926100533780304e-06,
|
| 352 |
+
"loss": 0.1824,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.2032520325203252,
|
| 357 |
+
"grad_norm": 1.2722930908203125,
|
| 358 |
+
"learning_rate": 9.922369083760461e-06,
|
| 359 |
+
"loss": 0.1743,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2073170731707317,
|
| 364 |
+
"grad_norm": 1.3577343225479126,
|
| 365 |
+
"learning_rate": 9.91854646886805e-06,
|
| 366 |
+
"loss": 0.1948,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.21138211382113822,
|
| 371 |
+
"grad_norm": 1.2834316492080688,
|
| 372 |
+
"learning_rate": 9.914632759899912e-06,
|
| 373 |
+
"loss": 0.2084,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.21544715447154472,
|
| 378 |
+
"grad_norm": 1.4345507621765137,
|
| 379 |
+
"learning_rate": 9.910628029340004e-06,
|
| 380 |
+
"loss": 0.2076,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.21951219512195122,
|
| 385 |
+
"grad_norm": 1.0987663269042969,
|
| 386 |
+
"learning_rate": 9.906532351358047e-06,
|
| 387 |
+
"loss": 0.1828,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.22357723577235772,
|
| 392 |
+
"grad_norm": 0.8563470840454102,
|
| 393 |
+
"learning_rate": 9.902345801808162e-06,
|
| 394 |
+
"loss": 0.1926,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.22764227642276422,
|
| 399 |
+
"grad_norm": 1.6927144527435303,
|
| 400 |
+
"learning_rate": 9.898068458227452e-06,
|
| 401 |
+
"loss": 0.2053,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.23170731707317074,
|
| 406 |
+
"grad_norm": 1.7086968421936035,
|
| 407 |
+
"learning_rate": 9.893700399834577e-06,
|
| 408 |
+
"loss": 0.1933,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.23577235772357724,
|
| 413 |
+
"grad_norm": 1.612290382385254,
|
| 414 |
+
"learning_rate": 9.889241707528285e-06,
|
| 415 |
+
"loss": 0.2127,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.23983739837398374,
|
| 420 |
+
"grad_norm": 1.4562150239944458,
|
| 421 |
+
"learning_rate": 9.88469246388591e-06,
|
| 422 |
+
"loss": 0.1697,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.24390243902439024,
|
| 427 |
+
"grad_norm": 1.047518014907837,
|
| 428 |
+
"learning_rate": 9.880052753161841e-06,
|
| 429 |
+
"loss": 0.2,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.24796747967479674,
|
| 434 |
+
"grad_norm": 1.394024133682251,
|
| 435 |
+
"learning_rate": 9.875322661285976e-06,
|
| 436 |
+
"loss": 0.1938,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.25203252032520324,
|
| 441 |
+
"grad_norm": 1.3038023710250854,
|
| 442 |
+
"learning_rate": 9.870502275862104e-06,
|
| 443 |
+
"loss": 0.1891,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.25609756097560976,
|
| 448 |
+
"grad_norm": 1.553615689277649,
|
| 449 |
+
"learning_rate": 9.86559168616631e-06,
|
| 450 |
+
"loss": 0.1887,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2601626016260163,
|
| 455 |
+
"grad_norm": 1.077765703201294,
|
| 456 |
+
"learning_rate": 9.860590983145307e-06,
|
| 457 |
+
"loss": 0.1866,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.26422764227642276,
|
| 462 |
+
"grad_norm": 1.4067726135253906,
|
| 463 |
+
"learning_rate": 9.855500259414752e-06,
|
| 464 |
+
"loss": 0.1955,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.2682926829268293,
|
| 469 |
+
"grad_norm": 1.4341750144958496,
|
| 470 |
+
"learning_rate": 9.850319609257536e-06,
|
| 471 |
+
"loss": 0.1969,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.27235772357723576,
|
| 476 |
+
"grad_norm": 1.605159044265747,
|
| 477 |
+
"learning_rate": 9.845049128622031e-06,
|
| 478 |
+
"loss": 0.196,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2764227642276423,
|
| 483 |
+
"grad_norm": 1.032614827156067,
|
| 484 |
+
"learning_rate": 9.839688915120322e-06,
|
| 485 |
+
"loss": 0.1698,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.2804878048780488,
|
| 490 |
+
"grad_norm": 1.256981372833252,
|
| 491 |
+
"learning_rate": 9.834239068026388e-06,
|
| 492 |
+
"loss": 0.1821,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.2845528455284553,
|
| 497 |
+
"grad_norm": 1.1425936222076416,
|
| 498 |
+
"learning_rate": 9.828699688274275e-06,
|
| 499 |
+
"loss": 0.1794,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.2886178861788618,
|
| 504 |
+
"grad_norm": 1.3988679647445679,
|
| 505 |
+
"learning_rate": 9.823070878456218e-06,
|
| 506 |
+
"loss": 0.1796,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.2926829268292683,
|
| 511 |
+
"grad_norm": 1.4095934629440308,
|
| 512 |
+
"learning_rate": 9.817352742820745e-06,
|
| 513 |
+
"loss": 0.1796,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2967479674796748,
|
| 518 |
+
"grad_norm": 1.2526706457138062,
|
| 519 |
+
"learning_rate": 9.811545387270745e-06,
|
| 520 |
+
"loss": 0.1747,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.3008130081300813,
|
| 525 |
+
"grad_norm": 1.1797000169754028,
|
| 526 |
+
"learning_rate": 9.805648919361505e-06,
|
| 527 |
+
"loss": 0.1868,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.3048780487804878,
|
| 532 |
+
"grad_norm": 1.2842937707901,
|
| 533 |
+
"learning_rate": 9.799663448298724e-06,
|
| 534 |
+
"loss": 0.186,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.3089430894308943,
|
| 539 |
+
"grad_norm": 1.338019847869873,
|
| 540 |
+
"learning_rate": 9.793589084936483e-06,
|
| 541 |
+
"loss": 0.1834,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.3130081300813008,
|
| 546 |
+
"grad_norm": 1.3759115934371948,
|
| 547 |
+
"learning_rate": 9.787425941775197e-06,
|
| 548 |
+
"loss": 0.1794,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3170731707317073,
|
| 553 |
+
"grad_norm": 1.1694071292877197,
|
| 554 |
+
"learning_rate": 9.781174132959528e-06,
|
| 555 |
+
"loss": 0.1809,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.32113821138211385,
|
| 560 |
+
"grad_norm": 1.0123424530029297,
|
| 561 |
+
"learning_rate": 9.774833774276278e-06,
|
| 562 |
+
"loss": 0.1975,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.3252032520325203,
|
| 567 |
+
"grad_norm": 1.3919343948364258,
|
| 568 |
+
"learning_rate": 9.768404983152228e-06,
|
| 569 |
+
"loss": 0.1687,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.32926829268292684,
|
| 574 |
+
"grad_norm": 1.314193844795227,
|
| 575 |
+
"learning_rate": 9.761887878651987e-06,
|
| 576 |
+
"loss": 0.1893,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3333333333333333,
|
| 581 |
+
"grad_norm": 1.2398736476898193,
|
| 582 |
+
"learning_rate": 9.755282581475769e-06,
|
| 583 |
+
"loss": 0.183,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.33739837398373984,
|
| 588 |
+
"grad_norm": 1.213040828704834,
|
| 589 |
+
"learning_rate": 9.74858921395716e-06,
|
| 590 |
+
"loss": 0.2065,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.34146341463414637,
|
| 595 |
+
"grad_norm": 1.3436555862426758,
|
| 596 |
+
"learning_rate": 9.741807900060858e-06,
|
| 597 |
+
"loss": 0.1686,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.34552845528455284,
|
| 602 |
+
"grad_norm": 1.178867220878601,
|
| 603 |
+
"learning_rate": 9.734938765380377e-06,
|
| 604 |
+
"loss": 0.1546,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.34959349593495936,
|
| 609 |
+
"grad_norm": 1.6577900648117065,
|
| 610 |
+
"learning_rate": 9.727981937135713e-06,
|
| 611 |
+
"loss": 0.1847,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.35365853658536583,
|
| 616 |
+
"grad_norm": 1.3156465291976929,
|
| 617 |
+
"learning_rate": 9.720937544171e-06,
|
| 618 |
+
"loss": 0.1635,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.35772357723577236,
|
| 623 |
+
"grad_norm": 1.3383092880249023,
|
| 624 |
+
"learning_rate": 9.713805716952106e-06,
|
| 625 |
+
"loss": 0.1809,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.3617886178861789,
|
| 630 |
+
"grad_norm": 1.229560136795044,
|
| 631 |
+
"learning_rate": 9.706586587564236e-06,
|
| 632 |
+
"loss": 0.1916,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.36585365853658536,
|
| 637 |
+
"grad_norm": 1.2169535160064697,
|
| 638 |
+
"learning_rate": 9.699280289709479e-06,
|
| 639 |
+
"loss": 0.1688,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.3699186991869919,
|
| 644 |
+
"grad_norm": 1.8572520017623901,
|
| 645 |
+
"learning_rate": 9.691886958704322e-06,
|
| 646 |
+
"loss": 0.1837,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.37398373983739835,
|
| 651 |
+
"grad_norm": 1.5430294275283813,
|
| 652 |
+
"learning_rate": 9.684406731477158e-06,
|
| 653 |
+
"loss": 0.1705,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3780487804878049,
|
| 658 |
+
"grad_norm": 1.603841781616211,
|
| 659 |
+
"learning_rate": 9.676839746565743e-06,
|
| 660 |
+
"loss": 0.1826,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3821138211382114,
|
| 665 |
+
"grad_norm": 1.1137135028839111,
|
| 666 |
+
"learning_rate": 9.669186144114627e-06,
|
| 667 |
+
"loss": 0.1874,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.3861788617886179,
|
| 672 |
+
"grad_norm": 1.2633637189865112,
|
| 673 |
+
"learning_rate": 9.661446065872568e-06,
|
| 674 |
+
"loss": 0.1623,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3902439024390244,
|
| 679 |
+
"grad_norm": 1.3832820653915405,
|
| 680 |
+
"learning_rate": 9.653619655189897e-06,
|
| 681 |
+
"loss": 0.187,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.3943089430894309,
|
| 686 |
+
"grad_norm": 1.438998818397522,
|
| 687 |
+
"learning_rate": 9.645707057015872e-06,
|
| 688 |
+
"loss": 0.175,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.3983739837398374,
|
| 693 |
+
"grad_norm": 1.2887821197509766,
|
| 694 |
+
"learning_rate": 9.637708417895984e-06,
|
| 695 |
+
"loss": 0.1958,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.4024390243902439,
|
| 700 |
+
"grad_norm": 1.3028004169464111,
|
| 701 |
+
"learning_rate": 9.62962388596925e-06,
|
| 702 |
+
"loss": 0.1811,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4065040650406504,
|
| 707 |
+
"grad_norm": 1.0944734811782837,
|
| 708 |
+
"learning_rate": 9.621453610965467e-06,
|
| 709 |
+
"loss": 0.1692,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.4105691056910569,
|
| 714 |
+
"grad_norm": 1.1278656721115112,
|
| 715 |
+
"learning_rate": 9.613197744202437e-06,
|
| 716 |
+
"loss": 0.1743,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4146341463414634,
|
| 721 |
+
"grad_norm": 1.009210467338562,
|
| 722 |
+
"learning_rate": 9.604856438583174e-06,
|
| 723 |
+
"loss": 0.1749,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.4186991869918699,
|
| 728 |
+
"grad_norm": 1.2928873300552368,
|
| 729 |
+
"learning_rate": 9.596429848593053e-06,
|
| 730 |
+
"loss": 0.178,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.42276422764227645,
|
| 735 |
+
"grad_norm": 1.3745077848434448,
|
| 736 |
+
"learning_rate": 9.587918130296969e-06,
|
| 737 |
+
"loss": 0.1825,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4268292682926829,
|
| 742 |
+
"grad_norm": 1.1097923517227173,
|
| 743 |
+
"learning_rate": 9.579321441336436e-06,
|
| 744 |
+
"loss": 0.1899,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.43089430894308944,
|
| 749 |
+
"grad_norm": 1.1704317331314087,
|
| 750 |
+
"learning_rate": 9.57063994092667e-06,
|
| 751 |
+
"loss": 0.194,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.4349593495934959,
|
| 756 |
+
"grad_norm": 1.0622445344924927,
|
| 757 |
+
"learning_rate": 9.561873789853638e-06,
|
| 758 |
+
"loss": 0.1815,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.43902439024390244,
|
| 763 |
+
"grad_norm": 1.6588221788406372,
|
| 764 |
+
"learning_rate": 9.553023150471083e-06,
|
| 765 |
+
"loss": 0.1818,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.44308943089430897,
|
| 770 |
+
"grad_norm": 1.7097984552383423,
|
| 771 |
+
"learning_rate": 9.544088186697515e-06,
|
| 772 |
+
"loss": 0.1899,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.44715447154471544,
|
| 777 |
+
"grad_norm": 1.154319405555725,
|
| 778 |
+
"learning_rate": 9.53506906401318e-06,
|
| 779 |
+
"loss": 0.1625,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.45121951219512196,
|
| 784 |
+
"grad_norm": 1.3818111419677734,
|
| 785 |
+
"learning_rate": 9.525965949456989e-06,
|
| 786 |
+
"loss": 0.1903,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.45528455284552843,
|
| 791 |
+
"grad_norm": 1.0541447401046753,
|
| 792 |
+
"learning_rate": 9.516779011623422e-06,
|
| 793 |
+
"loss": 0.1731,
|
| 794 |
+
"step": 112
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.45934959349593496,
|
| 798 |
+
"grad_norm": 1.1937304735183716,
|
| 799 |
+
"learning_rate": 9.507508420659422e-06,
|
| 800 |
+
"loss": 0.1664,
|
| 801 |
+
"step": 113
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4634146341463415,
|
| 805 |
+
"grad_norm": 1.2838006019592285,
|
| 806 |
+
"learning_rate": 9.498154348261217e-06,
|
| 807 |
+
"loss": 0.1866,
|
| 808 |
+
"step": 114
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.46747967479674796,
|
| 812 |
+
"grad_norm": 1.0903745889663696,
|
| 813 |
+
"learning_rate": 9.488716967671169e-06,
|
| 814 |
+
"loss": 0.1852,
|
| 815 |
+
"step": 115
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.4715447154471545,
|
| 819 |
+
"grad_norm": 1.1739411354064941,
|
| 820 |
+
"learning_rate": 9.479196453674544e-06,
|
| 821 |
+
"loss": 0.1814,
|
| 822 |
+
"step": 116
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.47560975609756095,
|
| 826 |
+
"grad_norm": 1.060652732849121,
|
| 827 |
+
"learning_rate": 9.469592982596286e-06,
|
| 828 |
+
"loss": 0.1874,
|
| 829 |
+
"step": 117
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.4796747967479675,
|
| 833 |
+
"grad_norm": 1.3752566576004028,
|
| 834 |
+
"learning_rate": 9.459906732297744e-06,
|
| 835 |
+
"loss": 0.1723,
|
| 836 |
+
"step": 118
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.483739837398374,
|
| 840 |
+
"grad_norm": 1.035359501838684,
|
| 841 |
+
"learning_rate": 9.450137882173385e-06,
|
| 842 |
+
"loss": 0.1894,
|
| 843 |
+
"step": 119
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.4878048780487805,
|
| 847 |
+
"grad_norm": 0.8948262333869934,
|
| 848 |
+
"learning_rate": 9.440286613147467e-06,
|
| 849 |
+
"loss": 0.171,
|
| 850 |
+
"step": 120
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.491869918699187,
|
| 854 |
+
"grad_norm": 0.9894183278083801,
|
| 855 |
+
"learning_rate": 9.430353107670691e-06,
|
| 856 |
+
"loss": 0.163,
|
| 857 |
+
"step": 121
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.4959349593495935,
|
| 861 |
+
"grad_norm": 1.2397547960281372,
|
| 862 |
+
"learning_rate": 9.420337549716819e-06,
|
| 863 |
+
"loss": 0.1622,
|
| 864 |
+
"step": 122
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5,
|
| 868 |
+
"grad_norm": 0.9623039364814758,
|
| 869 |
+
"learning_rate": 9.410240124779269e-06,
|
| 870 |
+
"loss": 0.1743,
|
| 871 |
+
"step": 123
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5040650406504065,
|
| 875 |
+
"grad_norm": 0.978670597076416,
|
| 876 |
+
"learning_rate": 9.40006101986768e-06,
|
| 877 |
+
"loss": 0.1673,
|
| 878 |
+
"step": 124
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.508130081300813,
|
| 882 |
+
"grad_norm": 0.943098783493042,
|
| 883 |
+
"learning_rate": 9.389800423504442e-06,
|
| 884 |
+
"loss": 0.1795,
|
| 885 |
+
"step": 125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5121951219512195,
|
| 889 |
+
"grad_norm": 1.4346143007278442,
|
| 890 |
+
"learning_rate": 9.379458525721216e-06,
|
| 891 |
+
"loss": 0.1846,
|
| 892 |
+
"step": 126
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.516260162601626,
|
| 896 |
+
"grad_norm": 1.3881056308746338,
|
| 897 |
+
"learning_rate": 9.369035518055405e-06,
|
| 898 |
+
"loss": 0.1826,
|
| 899 |
+
"step": 127
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.5203252032520326,
|
| 903 |
+
"grad_norm": 1.1909414529800415,
|
| 904 |
+
"learning_rate": 9.35853159354661e-06,
|
| 905 |
+
"loss": 0.1658,
|
| 906 |
+
"step": 128
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.524390243902439,
|
| 910 |
+
"grad_norm": 1.1166255474090576,
|
| 911 |
+
"learning_rate": 9.347946946733055e-06,
|
| 912 |
+
"loss": 0.1708,
|
| 913 |
+
"step": 129
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5284552845528455,
|
| 917 |
+
"grad_norm": 1.2461472749710083,
|
| 918 |
+
"learning_rate": 9.337281773647986e-06,
|
| 919 |
+
"loss": 0.1867,
|
| 920 |
+
"step": 130
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.532520325203252,
|
| 924 |
+
"grad_norm": 1.0086560249328613,
|
| 925 |
+
"learning_rate": 9.326536271816032e-06,
|
| 926 |
+
"loss": 0.1542,
|
| 927 |
+
"step": 131
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5365853658536586,
|
| 931 |
+
"grad_norm": 1.1569541692733765,
|
| 932 |
+
"learning_rate": 9.315710640249557e-06,
|
| 933 |
+
"loss": 0.1673,
|
| 934 |
+
"step": 132
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.540650406504065,
|
| 938 |
+
"grad_norm": 1.0628113746643066,
|
| 939 |
+
"learning_rate": 9.304805079444972e-06,
|
| 940 |
+
"loss": 0.1627,
|
| 941 |
+
"step": 133
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5447154471544715,
|
| 945 |
+
"grad_norm": 1.1958539485931396,
|
| 946 |
+
"learning_rate": 9.293819791379016e-06,
|
| 947 |
+
"loss": 0.173,
|
| 948 |
+
"step": 134
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5487804878048781,
|
| 952 |
+
"grad_norm": 1.3623948097229004,
|
| 953 |
+
"learning_rate": 9.282754979505019e-06,
|
| 954 |
+
"loss": 0.1742,
|
| 955 |
+
"step": 135
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5528455284552846,
|
| 959 |
+
"grad_norm": 1.2932652235031128,
|
| 960 |
+
"learning_rate": 9.271610848749134e-06,
|
| 961 |
+
"loss": 0.1706,
|
| 962 |
+
"step": 136
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.556910569105691,
|
| 966 |
+
"grad_norm": 1.0769895315170288,
|
| 967 |
+
"learning_rate": 9.260387605506547e-06,
|
| 968 |
+
"loss": 0.1856,
|
| 969 |
+
"step": 137
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5609756097560976,
|
| 973 |
+
"grad_norm": 1.0445054769515991,
|
| 974 |
+
"learning_rate": 9.24908545763764e-06,
|
| 975 |
+
"loss": 0.166,
|
| 976 |
+
"step": 138
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5650406504065041,
|
| 980 |
+
"grad_norm": 1.2573658227920532,
|
| 981 |
+
"learning_rate": 9.237704614464157e-06,
|
| 982 |
+
"loss": 0.1752,
|
| 983 |
+
"step": 139
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5691056910569106,
|
| 987 |
+
"grad_norm": 1.2782193422317505,
|
| 988 |
+
"learning_rate": 9.226245286765316e-06,
|
| 989 |
+
"loss": 0.1682,
|
| 990 |
+
"step": 140
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.573170731707317,
|
| 994 |
+
"grad_norm": 1.1732444763183594,
|
| 995 |
+
"learning_rate": 9.214707686773915e-06,
|
| 996 |
+
"loss": 0.1732,
|
| 997 |
+
"step": 141
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5772357723577236,
|
| 1001 |
+
"grad_norm": 0.9451704025268555,
|
| 1002 |
+
"learning_rate": 9.203092028172392e-06,
|
| 1003 |
+
"loss": 0.1974,
|
| 1004 |
+
"step": 142
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.5813008130081301,
|
| 1008 |
+
"grad_norm": 1.0854642391204834,
|
| 1009 |
+
"learning_rate": 9.191398526088873e-06,
|
| 1010 |
+
"loss": 0.1721,
|
| 1011 |
+
"step": 143
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.5853658536585366,
|
| 1015 |
+
"grad_norm": 1.0151511430740356,
|
| 1016 |
+
"learning_rate": 9.179627397093184e-06,
|
| 1017 |
+
"loss": 0.1742,
|
| 1018 |
+
"step": 144
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.5894308943089431,
|
| 1022 |
+
"grad_norm": 1.2080364227294922,
|
| 1023 |
+
"learning_rate": 9.167778859192851e-06,
|
| 1024 |
+
"loss": 0.1891,
|
| 1025 |
+
"step": 145
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.5934959349593496,
|
| 1029 |
+
"grad_norm": 1.5080487728118896,
|
| 1030 |
+
"learning_rate": 9.15585313182904e-06,
|
| 1031 |
+
"loss": 0.1952,
|
| 1032 |
+
"step": 146
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.5975609756097561,
|
| 1036 |
+
"grad_norm": 1.42489492893219,
|
| 1037 |
+
"learning_rate": 9.143850435872518e-06,
|
| 1038 |
+
"loss": 0.1729,
|
| 1039 |
+
"step": 147
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6016260162601627,
|
| 1043 |
+
"grad_norm": 1.0646756887435913,
|
| 1044 |
+
"learning_rate": 9.131770993619548e-06,
|
| 1045 |
+
"loss": 0.1683,
|
| 1046 |
+
"step": 148
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6056910569105691,
|
| 1050 |
+
"grad_norm": 1.00276780128479,
|
| 1051 |
+
"learning_rate": 9.119615028787771e-06,
|
| 1052 |
+
"loss": 0.1732,
|
| 1053 |
+
"step": 149
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6097560975609756,
|
| 1057 |
+
"grad_norm": 1.4372562170028687,
|
| 1058 |
+
"learning_rate": 9.107382766512072e-06,
|
| 1059 |
+
"loss": 0.1692,
|
| 1060 |
+
"step": 150
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6138211382113821,
|
| 1064 |
+
"grad_norm": 1.1330622434616089,
|
| 1065 |
+
"learning_rate": 9.0950744333404e-06,
|
| 1066 |
+
"loss": 0.1689,
|
| 1067 |
+
"step": 151
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6178861788617886,
|
| 1071 |
+
"grad_norm": 1.3941912651062012,
|
| 1072 |
+
"learning_rate": 9.08269025722958e-06,
|
| 1073 |
+
"loss": 0.1817,
|
| 1074 |
+
"step": 152
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6219512195121951,
|
| 1078 |
+
"grad_norm": 1.3748692274093628,
|
| 1079 |
+
"learning_rate": 9.07023046754109e-06,
|
| 1080 |
+
"loss": 0.1886,
|
| 1081 |
+
"step": 153
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6260162601626016,
|
| 1085 |
+
"grad_norm": 1.2595335245132446,
|
| 1086 |
+
"learning_rate": 9.057695295036806e-06,
|
| 1087 |
+
"loss": 0.1875,
|
| 1088 |
+
"step": 154
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6300813008130082,
|
| 1092 |
+
"grad_norm": 1.4108858108520508,
|
| 1093 |
+
"learning_rate": 9.045084971874738e-06,
|
| 1094 |
+
"loss": 0.1725,
|
| 1095 |
+
"step": 155
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6341463414634146,
|
| 1099 |
+
"grad_norm": 1.327467918395996,
|
| 1100 |
+
"learning_rate": 9.032399731604724e-06,
|
| 1101 |
+
"loss": 0.1631,
|
| 1102 |
+
"step": 156
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6382113821138211,
|
| 1106 |
+
"grad_norm": 1.1732194423675537,
|
| 1107 |
+
"learning_rate": 9.019639809164107e-06,
|
| 1108 |
+
"loss": 0.1741,
|
| 1109 |
+
"step": 157
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6422764227642277,
|
| 1113 |
+
"grad_norm": 1.121265172958374,
|
| 1114 |
+
"learning_rate": 9.006805440873383e-06,
|
| 1115 |
+
"loss": 0.1679,
|
| 1116 |
+
"step": 158
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6463414634146342,
|
| 1120 |
+
"grad_norm": 0.9814187288284302,
|
| 1121 |
+
"learning_rate": 8.993896864431825e-06,
|
| 1122 |
+
"loss": 0.1765,
|
| 1123 |
+
"step": 159
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6504065040650406,
|
| 1127 |
+
"grad_norm": 1.479628562927246,
|
| 1128 |
+
"learning_rate": 8.980914318913079e-06,
|
| 1129 |
+
"loss": 0.1633,
|
| 1130 |
+
"step": 160
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6544715447154471,
|
| 1134 |
+
"grad_norm": 1.3863518238067627,
|
| 1135 |
+
"learning_rate": 8.967858044760737e-06,
|
| 1136 |
+
"loss": 0.1583,
|
| 1137 |
+
"step": 161
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6585365853658537,
|
| 1141 |
+
"grad_norm": 1.066834807395935,
|
| 1142 |
+
"learning_rate": 8.954728283783887e-06,
|
| 1143 |
+
"loss": 0.1726,
|
| 1144 |
+
"step": 162
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.6626016260162602,
|
| 1148 |
+
"grad_norm": 1.0615098476409912,
|
| 1149 |
+
"learning_rate": 8.941525279152627e-06,
|
| 1150 |
+
"loss": 0.1648,
|
| 1151 |
+
"step": 163
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.6666666666666666,
|
| 1155 |
+
"grad_norm": 0.9500841498374939,
|
| 1156 |
+
"learning_rate": 8.928249275393572e-06,
|
| 1157 |
+
"loss": 0.1692,
|
| 1158 |
+
"step": 164
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.6707317073170732,
|
| 1162 |
+
"grad_norm": 1.0013989210128784,
|
| 1163 |
+
"learning_rate": 8.914900518385314e-06,
|
| 1164 |
+
"loss": 0.1588,
|
| 1165 |
+
"step": 165
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.6747967479674797,
|
| 1169 |
+
"grad_norm": 0.9967426061630249,
|
| 1170 |
+
"learning_rate": 8.90147925535388e-06,
|
| 1171 |
+
"loss": 0.1555,
|
| 1172 |
+
"step": 166
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.6788617886178862,
|
| 1176 |
+
"grad_norm": 1.2548542022705078,
|
| 1177 |
+
"learning_rate": 8.887985734868135e-06,
|
| 1178 |
+
"loss": 0.1636,
|
| 1179 |
+
"step": 167
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.6829268292682927,
|
| 1183 |
+
"grad_norm": 1.8094799518585205,
|
| 1184 |
+
"learning_rate": 8.874420206835204e-06,
|
| 1185 |
+
"loss": 0.1882,
|
| 1186 |
+
"step": 168
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.6869918699186992,
|
| 1190 |
+
"grad_norm": 0.8959202766418457,
|
| 1191 |
+
"learning_rate": 8.860782922495821e-06,
|
| 1192 |
+
"loss": 0.1848,
|
| 1193 |
+
"step": 169
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.6910569105691057,
|
| 1197 |
+
"grad_norm": 1.0040369033813477,
|
| 1198 |
+
"learning_rate": 8.847074134419691e-06,
|
| 1199 |
+
"loss": 0.1658,
|
| 1200 |
+
"step": 170
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.6951219512195121,
|
| 1204 |
+
"grad_norm": 1.1602360010147095,
|
| 1205 |
+
"learning_rate": 8.833294096500798e-06,
|
| 1206 |
+
"loss": 0.1918,
|
| 1207 |
+
"step": 171
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.6991869918699187,
|
| 1211 |
+
"grad_norm": 1.5341572761535645,
|
| 1212 |
+
"learning_rate": 8.819443063952714e-06,
|
| 1213 |
+
"loss": 0.1566,
|
| 1214 |
+
"step": 172
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7032520325203252,
|
| 1218 |
+
"grad_norm": 1.1187114715576172,
|
| 1219 |
+
"learning_rate": 8.805521293303874e-06,
|
| 1220 |
+
"loss": 0.1522,
|
| 1221 |
+
"step": 173
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7073170731707317,
|
| 1225 |
+
"grad_norm": 0.9167439341545105,
|
| 1226 |
+
"learning_rate": 8.791529042392813e-06,
|
| 1227 |
+
"loss": 0.1723,
|
| 1228 |
+
"step": 174
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7113821138211383,
|
| 1232 |
+
"grad_norm": 1.347064733505249,
|
| 1233 |
+
"learning_rate": 8.777466570363402e-06,
|
| 1234 |
+
"loss": 0.1669,
|
| 1235 |
+
"step": 175
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7154471544715447,
|
| 1239 |
+
"grad_norm": 1.2650213241577148,
|
| 1240 |
+
"learning_rate": 8.763334137660043e-06,
|
| 1241 |
+
"loss": 0.1929,
|
| 1242 |
+
"step": 176
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7195121951219512,
|
| 1246 |
+
"grad_norm": 1.1395199298858643,
|
| 1247 |
+
"learning_rate": 8.749132006022843e-06,
|
| 1248 |
+
"loss": 0.1615,
|
| 1249 |
+
"step": 177
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7235772357723578,
|
| 1253 |
+
"grad_norm": 1.1134856939315796,
|
| 1254 |
+
"learning_rate": 8.734860438482777e-06,
|
| 1255 |
+
"loss": 0.1707,
|
| 1256 |
+
"step": 178
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7276422764227642,
|
| 1260 |
+
"grad_norm": 1.2230736017227173,
|
| 1261 |
+
"learning_rate": 8.720519699356804e-06,
|
| 1262 |
+
"loss": 0.1772,
|
| 1263 |
+
"step": 179
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7317073170731707,
|
| 1267 |
+
"grad_norm": 1.051064133644104,
|
| 1268 |
+
"learning_rate": 8.70611005424298e-06,
|
| 1269 |
+
"loss": 0.1707,
|
| 1270 |
+
"step": 180
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.7357723577235772,
|
| 1274 |
+
"grad_norm": 1.5673556327819824,
|
| 1275 |
+
"learning_rate": 8.691631770015534e-06,
|
| 1276 |
+
"loss": 0.1481,
|
| 1277 |
+
"step": 181
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7398373983739838,
|
| 1281 |
+
"grad_norm": 0.9244242906570435,
|
| 1282 |
+
"learning_rate": 8.677085114819928e-06,
|
| 1283 |
+
"loss": 0.1564,
|
| 1284 |
+
"step": 182
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.7439024390243902,
|
| 1288 |
+
"grad_norm": 1.3698534965515137,
|
| 1289 |
+
"learning_rate": 8.662470358067895e-06,
|
| 1290 |
+
"loss": 0.1684,
|
| 1291 |
+
"step": 183
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7479674796747967,
|
| 1295 |
+
"grad_norm": 0.9856629371643066,
|
| 1296 |
+
"learning_rate": 8.647787770432439e-06,
|
| 1297 |
+
"loss": 0.1562,
|
| 1298 |
+
"step": 184
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7520325203252033,
|
| 1302 |
+
"grad_norm": 1.1614383459091187,
|
| 1303 |
+
"learning_rate": 8.633037623842828e-06,
|
| 1304 |
+
"loss": 0.1782,
|
| 1305 |
+
"step": 185
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7560975609756098,
|
| 1309 |
+
"grad_norm": 0.9977948665618896,
|
| 1310 |
+
"learning_rate": 8.618220191479563e-06,
|
| 1311 |
+
"loss": 0.1672,
|
| 1312 |
+
"step": 186
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7601626016260162,
|
| 1316 |
+
"grad_norm": 0.8170651197433472,
|
| 1317 |
+
"learning_rate": 8.603335747769305e-06,
|
| 1318 |
+
"loss": 0.1429,
|
| 1319 |
+
"step": 187
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7642276422764228,
|
| 1323 |
+
"grad_norm": 0.9341660737991333,
|
| 1324 |
+
"learning_rate": 8.588384568379811e-06,
|
| 1325 |
+
"loss": 0.1676,
|
| 1326 |
+
"step": 188
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.7682926829268293,
|
| 1330 |
+
"grad_norm": 0.9494506120681763,
|
| 1331 |
+
"learning_rate": 8.573366930214807e-06,
|
| 1332 |
+
"loss": 0.1315,
|
| 1333 |
+
"step": 189
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.7723577235772358,
|
| 1337 |
+
"grad_norm": 1.4293547868728638,
|
| 1338 |
+
"learning_rate": 8.558283111408874e-06,
|
| 1339 |
+
"loss": 0.1612,
|
| 1340 |
+
"step": 190
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.7764227642276422,
|
| 1344 |
+
"grad_norm": 1.1602962017059326,
|
| 1345 |
+
"learning_rate": 8.5431333913223e-06,
|
| 1346 |
+
"loss": 0.131,
|
| 1347 |
+
"step": 191
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.7804878048780488,
|
| 1351 |
+
"grad_norm": 1.2388046979904175,
|
| 1352 |
+
"learning_rate": 8.52791805053589e-06,
|
| 1353 |
+
"loss": 0.1555,
|
| 1354 |
+
"step": 192
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.7845528455284553,
|
| 1358 |
+
"grad_norm": 0.9719280004501343,
|
| 1359 |
+
"learning_rate": 8.512637370845785e-06,
|
| 1360 |
+
"loss": 0.1573,
|
| 1361 |
+
"step": 193
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.7886178861788617,
|
| 1365 |
+
"grad_norm": 1.2786000967025757,
|
| 1366 |
+
"learning_rate": 8.497291635258235e-06,
|
| 1367 |
+
"loss": 0.1504,
|
| 1368 |
+
"step": 194
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.7926829268292683,
|
| 1372 |
+
"grad_norm": 1.314968466758728,
|
| 1373 |
+
"learning_rate": 8.481881127984361e-06,
|
| 1374 |
+
"loss": 0.1734,
|
| 1375 |
+
"step": 195
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.7967479674796748,
|
| 1379 |
+
"grad_norm": 0.7489177584648132,
|
| 1380 |
+
"learning_rate": 8.466406134434888e-06,
|
| 1381 |
+
"loss": 0.1545,
|
| 1382 |
+
"step": 196
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.8008130081300813,
|
| 1386 |
+
"grad_norm": 1.1359909772872925,
|
| 1387 |
+
"learning_rate": 8.45086694121486e-06,
|
| 1388 |
+
"loss": 0.1692,
|
| 1389 |
+
"step": 197
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8048780487804879,
|
| 1393 |
+
"grad_norm": 1.418720006942749,
|
| 1394 |
+
"learning_rate": 8.435263836118336e-06,
|
| 1395 |
+
"loss": 0.1755,
|
| 1396 |
+
"step": 198
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8089430894308943,
|
| 1400 |
+
"grad_norm": 1.4378105401992798,
|
| 1401 |
+
"learning_rate": 8.419597108123054e-06,
|
| 1402 |
+
"loss": 0.1934,
|
| 1403 |
+
"step": 199
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.8130081300813008,
|
| 1407 |
+
"grad_norm": 1.1287977695465088,
|
| 1408 |
+
"learning_rate": 8.403867047385082e-06,
|
| 1409 |
+
"loss": 0.1487,
|
| 1410 |
+
"step": 200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.8170731707317073,
|
| 1414 |
+
"grad_norm": 1.0692404508590698,
|
| 1415 |
+
"learning_rate": 8.388073945233446e-06,
|
| 1416 |
+
"loss": 0.1723,
|
| 1417 |
+
"step": 201
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8211382113821138,
|
| 1421 |
+
"grad_norm": 1.3414398431777954,
|
| 1422 |
+
"learning_rate": 8.372218094164727e-06,
|
| 1423 |
+
"loss": 0.1732,
|
| 1424 |
+
"step": 202
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.8252032520325203,
|
| 1428 |
+
"grad_norm": 1.2793030738830566,
|
| 1429 |
+
"learning_rate": 8.356299787837659e-06,
|
| 1430 |
+
"loss": 0.1862,
|
| 1431 |
+
"step": 203
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8292682926829268,
|
| 1435 |
+
"grad_norm": 1.4274896383285522,
|
| 1436 |
+
"learning_rate": 8.340319321067668e-06,
|
| 1437 |
+
"loss": 0.1742,
|
| 1438 |
+
"step": 204
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8333333333333334,
|
| 1442 |
+
"grad_norm": 0.9525288343429565,
|
| 1443 |
+
"learning_rate": 8.324276989821433e-06,
|
| 1444 |
+
"loss": 0.1743,
|
| 1445 |
+
"step": 205
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8373983739837398,
|
| 1449 |
+
"grad_norm": 0.9827443361282349,
|
| 1450 |
+
"learning_rate": 8.308173091211392e-06,
|
| 1451 |
+
"loss": 0.1685,
|
| 1452 |
+
"step": 206
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8414634146341463,
|
| 1456 |
+
"grad_norm": 1.1109260320663452,
|
| 1457 |
+
"learning_rate": 8.292007923490244e-06,
|
| 1458 |
+
"loss": 0.1632,
|
| 1459 |
+
"step": 207
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8455284552845529,
|
| 1463 |
+
"grad_norm": 1.3498234748840332,
|
| 1464 |
+
"learning_rate": 8.275781786045427e-06,
|
| 1465 |
+
"loss": 0.161,
|
| 1466 |
+
"step": 208
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8495934959349594,
|
| 1470 |
+
"grad_norm": 1.0240503549575806,
|
| 1471 |
+
"learning_rate": 8.259494979393563e-06,
|
| 1472 |
+
"loss": 0.1788,
|
| 1473 |
+
"step": 209
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8536585365853658,
|
| 1477 |
+
"grad_norm": 0.9341413378715515,
|
| 1478 |
+
"learning_rate": 8.243147805174908e-06,
|
| 1479 |
+
"loss": 0.1712,
|
| 1480 |
+
"step": 210
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8577235772357723,
|
| 1484 |
+
"grad_norm": 0.9897603392601013,
|
| 1485 |
+
"learning_rate": 8.226740566147753e-06,
|
| 1486 |
+
"loss": 0.1404,
|
| 1487 |
+
"step": 211
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8617886178861789,
|
| 1491 |
+
"grad_norm": 1.1501319408416748,
|
| 1492 |
+
"learning_rate": 8.210273566182819e-06,
|
| 1493 |
+
"loss": 0.1848,
|
| 1494 |
+
"step": 212
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8658536585365854,
|
| 1498 |
+
"grad_norm": 0.8344603180885315,
|
| 1499 |
+
"learning_rate": 8.193747110257638e-06,
|
| 1500 |
+
"loss": 0.1699,
|
| 1501 |
+
"step": 213
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.8699186991869918,
|
| 1505 |
+
"grad_norm": 1.102390170097351,
|
| 1506 |
+
"learning_rate": 8.177161504450887e-06,
|
| 1507 |
+
"loss": 0.1884,
|
| 1508 |
+
"step": 214
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.8739837398373984,
|
| 1512 |
+
"grad_norm": 1.0625027418136597,
|
| 1513 |
+
"learning_rate": 8.160517055936744e-06,
|
| 1514 |
+
"loss": 0.1697,
|
| 1515 |
+
"step": 215
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.8780487804878049,
|
| 1519 |
+
"grad_norm": 1.1806248426437378,
|
| 1520 |
+
"learning_rate": 8.143814072979173e-06,
|
| 1521 |
+
"loss": 0.1796,
|
| 1522 |
+
"step": 216
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.8821138211382114,
|
| 1526 |
+
"grad_norm": 0.9685717821121216,
|
| 1527 |
+
"learning_rate": 8.127052864926233e-06,
|
| 1528 |
+
"loss": 0.1634,
|
| 1529 |
+
"step": 217
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.8861788617886179,
|
| 1533 |
+
"grad_norm": 1.3439109325408936,
|
| 1534 |
+
"learning_rate": 8.110233742204339e-06,
|
| 1535 |
+
"loss": 0.1571,
|
| 1536 |
+
"step": 218
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.8902439024390244,
|
| 1540 |
+
"grad_norm": 0.9250830411911011,
|
| 1541 |
+
"learning_rate": 8.093357016312518e-06,
|
| 1542 |
+
"loss": 0.1694,
|
| 1543 |
+
"step": 219
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.8943089430894309,
|
| 1547 |
+
"grad_norm": 0.908725917339325,
|
| 1548 |
+
"learning_rate": 8.07642299981664e-06,
|
| 1549 |
+
"loss": 0.1833,
|
| 1550 |
+
"step": 220
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.8983739837398373,
|
| 1554 |
+
"grad_norm": 0.862690806388855,
|
| 1555 |
+
"learning_rate": 8.059432006343623e-06,
|
| 1556 |
+
"loss": 0.177,
|
| 1557 |
+
"step": 221
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.9024390243902439,
|
| 1561 |
+
"grad_norm": 1.4902915954589844,
|
| 1562 |
+
"learning_rate": 8.042384350575631e-06,
|
| 1563 |
+
"loss": 0.1678,
|
| 1564 |
+
"step": 222
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.9065040650406504,
|
| 1568 |
+
"grad_norm": 1.0128511190414429,
|
| 1569 |
+
"learning_rate": 8.025280348244247e-06,
|
| 1570 |
+
"loss": 0.1487,
|
| 1571 |
+
"step": 223
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.9105691056910569,
|
| 1575 |
+
"grad_norm": 0.8974586725234985,
|
| 1576 |
+
"learning_rate": 8.008120316124612e-06,
|
| 1577 |
+
"loss": 0.1651,
|
| 1578 |
+
"step": 224
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.9146341463414634,
|
| 1582 |
+
"grad_norm": 1.0842063426971436,
|
| 1583 |
+
"learning_rate": 7.990904572029584e-06,
|
| 1584 |
+
"loss": 0.1681,
|
| 1585 |
+
"step": 225
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9186991869918699,
|
| 1589 |
+
"grad_norm": 1.0749802589416504,
|
| 1590 |
+
"learning_rate": 7.973633434803817e-06,
|
| 1591 |
+
"loss": 0.1523,
|
| 1592 |
+
"step": 226
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.9227642276422764,
|
| 1596 |
+
"grad_norm": 1.2847234010696411,
|
| 1597 |
+
"learning_rate": 7.956307224317892e-06,
|
| 1598 |
+
"loss": 0.1441,
|
| 1599 |
+
"step": 227
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.926829268292683,
|
| 1603 |
+
"grad_norm": 1.2760282754898071,
|
| 1604 |
+
"learning_rate": 7.938926261462366e-06,
|
| 1605 |
+
"loss": 0.1619,
|
| 1606 |
+
"step": 228
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9308943089430894,
|
| 1610 |
+
"grad_norm": 1.1432336568832397,
|
| 1611 |
+
"learning_rate": 7.921490868141843e-06,
|
| 1612 |
+
"loss": 0.1936,
|
| 1613 |
+
"step": 229
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9349593495934959,
|
| 1617 |
+
"grad_norm": 0.8965000510215759,
|
| 1618 |
+
"learning_rate": 7.904001367269005e-06,
|
| 1619 |
+
"loss": 0.1487,
|
| 1620 |
+
"step": 230
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.9390243902439024,
|
| 1624 |
+
"grad_norm": 1.2623180150985718,
|
| 1625 |
+
"learning_rate": 7.886458082758637e-06,
|
| 1626 |
+
"loss": 0.1593,
|
| 1627 |
+
"step": 231
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.943089430894309,
|
| 1631 |
+
"grad_norm": 1.0198849439620972,
|
| 1632 |
+
"learning_rate": 7.868861339521626e-06,
|
| 1633 |
+
"loss": 0.1574,
|
| 1634 |
+
"step": 232
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.9471544715447154,
|
| 1638 |
+
"grad_norm": 1.1471402645111084,
|
| 1639 |
+
"learning_rate": 7.851211463458936e-06,
|
| 1640 |
+
"loss": 0.1707,
|
| 1641 |
+
"step": 233
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9512195121951219,
|
| 1645 |
+
"grad_norm": 1.4508056640625,
|
| 1646 |
+
"learning_rate": 7.83350878145559e-06,
|
| 1647 |
+
"loss": 0.1812,
|
| 1648 |
+
"step": 234
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9552845528455285,
|
| 1652 |
+
"grad_norm": 0.9366198778152466,
|
| 1653 |
+
"learning_rate": 7.815753621374594e-06,
|
| 1654 |
+
"loss": 0.1722,
|
| 1655 |
+
"step": 235
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.959349593495935,
|
| 1659 |
+
"grad_norm": 0.9615839719772339,
|
| 1660 |
+
"learning_rate": 7.797946312050884e-06,
|
| 1661 |
+
"loss": 0.1713,
|
| 1662 |
+
"step": 236
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9634146341463414,
|
| 1666 |
+
"grad_norm": 0.8765515685081482,
|
| 1667 |
+
"learning_rate": 7.780087183285224e-06,
|
| 1668 |
+
"loss": 0.1569,
|
| 1669 |
+
"step": 237
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.967479674796748,
|
| 1673 |
+
"grad_norm": 1.1343492269515991,
|
| 1674 |
+
"learning_rate": 7.7621765658381e-06,
|
| 1675 |
+
"loss": 0.1741,
|
| 1676 |
+
"step": 238
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.9715447154471545,
|
| 1680 |
+
"grad_norm": 1.2616182565689087,
|
| 1681 |
+
"learning_rate": 7.744214791423597e-06,
|
| 1682 |
+
"loss": 0.1706,
|
| 1683 |
+
"step": 239
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.975609756097561,
|
| 1687 |
+
"grad_norm": 1.5109883546829224,
|
| 1688 |
+
"learning_rate": 7.726202192703256e-06,
|
| 1689 |
+
"loss": 0.1682,
|
| 1690 |
+
"step": 240
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.9796747967479674,
|
| 1694 |
+
"grad_norm": 0.9209504127502441,
|
| 1695 |
+
"learning_rate": 7.708139103279908e-06,
|
| 1696 |
+
"loss": 0.1674,
|
| 1697 |
+
"step": 241
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.983739837398374,
|
| 1701 |
+
"grad_norm": 1.3614517450332642,
|
| 1702 |
+
"learning_rate": 7.6900258576915e-06,
|
| 1703 |
+
"loss": 0.1367,
|
| 1704 |
+
"step": 242
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.9878048780487805,
|
| 1708 |
+
"grad_norm": 0.8787021040916443,
|
| 1709 |
+
"learning_rate": 7.671862791404897e-06,
|
| 1710 |
+
"loss": 0.1546,
|
| 1711 |
+
"step": 243
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.991869918699187,
|
| 1715 |
+
"grad_norm": 1.0463536977767944,
|
| 1716 |
+
"learning_rate": 7.653650240809667e-06,
|
| 1717 |
+
"loss": 0.1782,
|
| 1718 |
+
"step": 244
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.9959349593495935,
|
| 1722 |
+
"grad_norm": 1.0385479927062988,
|
| 1723 |
+
"learning_rate": 7.635388543211862e-06,
|
| 1724 |
+
"loss": 0.1816,
|
| 1725 |
+
"step": 245
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.0,
|
| 1729 |
+
"grad_norm": 1.118327021598816,
|
| 1730 |
+
"learning_rate": 7.617078036827752e-06,
|
| 1731 |
+
"loss": 0.1757,
|
| 1732 |
+
"step": 246
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.0040650406504066,
|
| 1736 |
+
"grad_norm": 1.1916835308074951,
|
| 1737 |
+
"learning_rate": 7.5987190607775795e-06,
|
| 1738 |
+
"loss": 0.1399,
|
| 1739 |
+
"step": 247
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.008130081300813,
|
| 1743 |
+
"grad_norm": 0.9139823317527771,
|
| 1744 |
+
"learning_rate": 7.580311955079265e-06,
|
| 1745 |
+
"loss": 0.1317,
|
| 1746 |
+
"step": 248
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.0121951219512195,
|
| 1750 |
+
"grad_norm": 1.271535038948059,
|
| 1751 |
+
"learning_rate": 7.56185706064212e-06,
|
| 1752 |
+
"loss": 0.1537,
|
| 1753 |
+
"step": 249
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.016260162601626,
|
| 1757 |
+
"grad_norm": 1.03314208984375,
|
| 1758 |
+
"learning_rate": 7.543354719260523e-06,
|
| 1759 |
+
"loss": 0.1613,
|
| 1760 |
+
"step": 250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.0203252032520325,
|
| 1764 |
+
"grad_norm": 0.9401463270187378,
|
| 1765 |
+
"learning_rate": 7.5248052736076025e-06,
|
| 1766 |
+
"loss": 0.1654,
|
| 1767 |
+
"step": 251
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.024390243902439,
|
| 1771 |
+
"grad_norm": 0.8938135504722595,
|
| 1772 |
+
"learning_rate": 7.506209067228878e-06,
|
| 1773 |
+
"loss": 0.1333,
|
| 1774 |
+
"step": 252
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.0284552845528456,
|
| 1778 |
+
"grad_norm": 1.4358232021331787,
|
| 1779 |
+
"learning_rate": 7.487566444535904e-06,
|
| 1780 |
+
"loss": 0.1434,
|
| 1781 |
+
"step": 253
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.032520325203252,
|
| 1785 |
+
"grad_norm": 1.056515097618103,
|
| 1786 |
+
"learning_rate": 7.468877750799887e-06,
|
| 1787 |
+
"loss": 0.1369,
|
| 1788 |
+
"step": 254
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.0365853658536586,
|
| 1792 |
+
"grad_norm": 1.0800681114196777,
|
| 1793 |
+
"learning_rate": 7.450143332145297e-06,
|
| 1794 |
+
"loss": 0.1607,
|
| 1795 |
+
"step": 255
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.040650406504065,
|
| 1799 |
+
"grad_norm": 1.109963059425354,
|
| 1800 |
+
"learning_rate": 7.431363535543451e-06,
|
| 1801 |
+
"loss": 0.1515,
|
| 1802 |
+
"step": 256
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.0447154471544715,
|
| 1806 |
+
"grad_norm": 1.0541678667068481,
|
| 1807 |
+
"learning_rate": 7.412538708806093e-06,
|
| 1808 |
+
"loss": 0.1583,
|
| 1809 |
+
"step": 257
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.048780487804878,
|
| 1813 |
+
"grad_norm": 0.8683096170425415,
|
| 1814 |
+
"learning_rate": 7.393669200578944e-06,
|
| 1815 |
+
"loss": 0.144,
|
| 1816 |
+
"step": 258
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.0528455284552845,
|
| 1820 |
+
"grad_norm": 0.9488138556480408,
|
| 1821 |
+
"learning_rate": 7.374755360335253e-06,
|
| 1822 |
+
"loss": 0.1578,
|
| 1823 |
+
"step": 259
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.056910569105691,
|
| 1827 |
+
"grad_norm": 1.3625383377075195,
|
| 1828 |
+
"learning_rate": 7.35579753836932e-06,
|
| 1829 |
+
"loss": 0.1356,
|
| 1830 |
+
"step": 260
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.0609756097560976,
|
| 1834 |
+
"grad_norm": 1.2058069705963135,
|
| 1835 |
+
"learning_rate": 7.336796085790012e-06,
|
| 1836 |
+
"loss": 0.1491,
|
| 1837 |
+
"step": 261
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.065040650406504,
|
| 1841 |
+
"grad_norm": 1.1666299104690552,
|
| 1842 |
+
"learning_rate": 7.317751354514256e-06,
|
| 1843 |
+
"loss": 0.1603,
|
| 1844 |
+
"step": 262
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.0691056910569106,
|
| 1848 |
+
"grad_norm": 1.248290777206421,
|
| 1849 |
+
"learning_rate": 7.298663697260522e-06,
|
| 1850 |
+
"loss": 0.149,
|
| 1851 |
+
"step": 263
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.0731707317073171,
|
| 1855 |
+
"grad_norm": 1.0428169965744019,
|
| 1856 |
+
"learning_rate": 7.279533467542295e-06,
|
| 1857 |
+
"loss": 0.1643,
|
| 1858 |
+
"step": 264
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.0772357723577235,
|
| 1862 |
+
"grad_norm": 1.268149733543396,
|
| 1863 |
+
"learning_rate": 7.260361019661523e-06,
|
| 1864 |
+
"loss": 0.1467,
|
| 1865 |
+
"step": 265
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.08130081300813,
|
| 1869 |
+
"grad_norm": 0.9332938194274902,
|
| 1870 |
+
"learning_rate": 7.2411467087020536e-06,
|
| 1871 |
+
"loss": 0.1261,
|
| 1872 |
+
"step": 266
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.0853658536585367,
|
| 1876 |
+
"grad_norm": 1.2290884256362915,
|
| 1877 |
+
"learning_rate": 7.2218908905230674e-06,
|
| 1878 |
+
"loss": 0.1638,
|
| 1879 |
+
"step": 267
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.089430894308943,
|
| 1883 |
+
"grad_norm": 1.237654447555542,
|
| 1884 |
+
"learning_rate": 7.202593921752475e-06,
|
| 1885 |
+
"loss": 0.1252,
|
| 1886 |
+
"step": 268
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.0934959349593496,
|
| 1890 |
+
"grad_norm": 1.038710594177246,
|
| 1891 |
+
"learning_rate": 7.183256159780321e-06,
|
| 1892 |
+
"loss": 0.1663,
|
| 1893 |
+
"step": 269
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.0975609756097562,
|
| 1897 |
+
"grad_norm": 1.0794121026992798,
|
| 1898 |
+
"learning_rate": 7.163877962752157e-06,
|
| 1899 |
+
"loss": 0.1487,
|
| 1900 |
+
"step": 270
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.1016260162601625,
|
| 1904 |
+
"grad_norm": 1.0183168649673462,
|
| 1905 |
+
"learning_rate": 7.144459689562419e-06,
|
| 1906 |
+
"loss": 0.1678,
|
| 1907 |
+
"step": 271
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.1056910569105691,
|
| 1911 |
+
"grad_norm": 1.2545634508132935,
|
| 1912 |
+
"learning_rate": 7.12500169984777e-06,
|
| 1913 |
+
"loss": 0.1503,
|
| 1914 |
+
"step": 272
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.1097560975609757,
|
| 1918 |
+
"grad_norm": 0.8490943312644958,
|
| 1919 |
+
"learning_rate": 7.105504353980448e-06,
|
| 1920 |
+
"loss": 0.1613,
|
| 1921 |
+
"step": 273
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.113821138211382,
|
| 1925 |
+
"grad_norm": 1.0997651815414429,
|
| 1926 |
+
"learning_rate": 7.085968013061585e-06,
|
| 1927 |
+
"loss": 0.1737,
|
| 1928 |
+
"step": 274
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.1178861788617886,
|
| 1932 |
+
"grad_norm": 1.3434139490127563,
|
| 1933 |
+
"learning_rate": 7.066393038914522e-06,
|
| 1934 |
+
"loss": 0.1452,
|
| 1935 |
+
"step": 275
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.1219512195121952,
|
| 1939 |
+
"grad_norm": 1.0135364532470703,
|
| 1940 |
+
"learning_rate": 7.046779794078108e-06,
|
| 1941 |
+
"loss": 0.1635,
|
| 1942 |
+
"step": 276
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.1260162601626016,
|
| 1946 |
+
"grad_norm": 1.1503894329071045,
|
| 1947 |
+
"learning_rate": 7.027128641799987e-06,
|
| 1948 |
+
"loss": 0.176,
|
| 1949 |
+
"step": 277
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.1300813008130082,
|
| 1953 |
+
"grad_norm": 1.0549263954162598,
|
| 1954 |
+
"learning_rate": 7.007439946029866e-06,
|
| 1955 |
+
"loss": 0.1601,
|
| 1956 |
+
"step": 278
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.1341463414634148,
|
| 1960 |
+
"grad_norm": 1.084801197052002,
|
| 1961 |
+
"learning_rate": 6.987714071412781e-06,
|
| 1962 |
+
"loss": 0.1394,
|
| 1963 |
+
"step": 279
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.1382113821138211,
|
| 1967 |
+
"grad_norm": 1.3217436075210571,
|
| 1968 |
+
"learning_rate": 6.967951383282334e-06,
|
| 1969 |
+
"loss": 0.1603,
|
| 1970 |
+
"step": 280
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.1422764227642277,
|
| 1974 |
+
"grad_norm": 1.0782599449157715,
|
| 1975 |
+
"learning_rate": 6.948152247653941e-06,
|
| 1976 |
+
"loss": 0.1545,
|
| 1977 |
+
"step": 281
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.146341463414634,
|
| 1981 |
+
"grad_norm": 0.7805387377738953,
|
| 1982 |
+
"learning_rate": 6.928317031218036e-06,
|
| 1983 |
+
"loss": 0.1493,
|
| 1984 |
+
"step": 282
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.1504065040650406,
|
| 1988 |
+
"grad_norm": 1.0574467182159424,
|
| 1989 |
+
"learning_rate": 6.908446101333296e-06,
|
| 1990 |
+
"loss": 0.1516,
|
| 1991 |
+
"step": 283
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.1544715447154472,
|
| 1995 |
+
"grad_norm": 1.575653314590454,
|
| 1996 |
+
"learning_rate": 6.888539826019824e-06,
|
| 1997 |
+
"loss": 0.1678,
|
| 1998 |
+
"step": 284
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.1585365853658536,
|
| 2002 |
+
"grad_norm": 1.1213170289993286,
|
| 2003 |
+
"learning_rate": 6.868598573952345e-06,
|
| 2004 |
+
"loss": 0.1495,
|
| 2005 |
+
"step": 285
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.1626016260162602,
|
| 2009 |
+
"grad_norm": 1.0076125860214233,
|
| 2010 |
+
"learning_rate": 6.848622714453367e-06,
|
| 2011 |
+
"loss": 0.1499,
|
| 2012 |
+
"step": 286
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.1666666666666667,
|
| 2016 |
+
"grad_norm": 1.002481460571289,
|
| 2017 |
+
"learning_rate": 6.828612617486347e-06,
|
| 2018 |
+
"loss": 0.1637,
|
| 2019 |
+
"step": 287
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.170731707317073,
|
| 2023 |
+
"grad_norm": 1.576324224472046,
|
| 2024 |
+
"learning_rate": 6.808568653648838e-06,
|
| 2025 |
+
"loss": 0.1609,
|
| 2026 |
+
"step": 288
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.1747967479674797,
|
| 2030 |
+
"grad_norm": 1.2018259763717651,
|
| 2031 |
+
"learning_rate": 6.788491194165629e-06,
|
| 2032 |
+
"loss": 0.1488,
|
| 2033 |
+
"step": 289
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.1788617886178863,
|
| 2037 |
+
"grad_norm": 1.11052405834198,
|
| 2038 |
+
"learning_rate": 6.76838061088186e-06,
|
| 2039 |
+
"loss": 0.1597,
|
| 2040 |
+
"step": 290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.1829268292682926,
|
| 2044 |
+
"grad_norm": 1.0367604494094849,
|
| 2045 |
+
"learning_rate": 6.748237276256143e-06,
|
| 2046 |
+
"loss": 0.1477,
|
| 2047 |
+
"step": 291
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.1869918699186992,
|
| 2051 |
+
"grad_norm": 1.0717558860778809,
|
| 2052 |
+
"learning_rate": 6.728061563353667e-06,
|
| 2053 |
+
"loss": 0.1634,
|
| 2054 |
+
"step": 292
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.1910569105691058,
|
| 2058 |
+
"grad_norm": 0.9212750196456909,
|
| 2059 |
+
"learning_rate": 6.707853845839278e-06,
|
| 2060 |
+
"loss": 0.1486,
|
| 2061 |
+
"step": 293
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.1951219512195121,
|
| 2065 |
+
"grad_norm": 1.1082451343536377,
|
| 2066 |
+
"learning_rate": 6.687614497970567e-06,
|
| 2067 |
+
"loss": 0.1454,
|
| 2068 |
+
"step": 294
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.1991869918699187,
|
| 2072 |
+
"grad_norm": 1.2879371643066406,
|
| 2073 |
+
"learning_rate": 6.667343894590935e-06,
|
| 2074 |
+
"loss": 0.1461,
|
| 2075 |
+
"step": 295
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.203252032520325,
|
| 2079 |
+
"grad_norm": 1.3221591711044312,
|
| 2080 |
+
"learning_rate": 6.647042411122652e-06,
|
| 2081 |
+
"loss": 0.1376,
|
| 2082 |
+
"step": 296
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.2073170731707317,
|
| 2086 |
+
"grad_norm": 1.0754611492156982,
|
| 2087 |
+
"learning_rate": 6.626710423559902e-06,
|
| 2088 |
+
"loss": 0.17,
|
| 2089 |
+
"step": 297
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.2113821138211383,
|
| 2093 |
+
"grad_norm": 0.9823397994041443,
|
| 2094 |
+
"learning_rate": 6.606348308461823e-06,
|
| 2095 |
+
"loss": 0.1344,
|
| 2096 |
+
"step": 298
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.2154471544715446,
|
| 2100 |
+
"grad_norm": 0.8531427383422852,
|
| 2101 |
+
"learning_rate": 6.585956442945531e-06,
|
| 2102 |
+
"loss": 0.1306,
|
| 2103 |
+
"step": 299
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.2195121951219512,
|
| 2107 |
+
"grad_norm": 1.211600661277771,
|
| 2108 |
+
"learning_rate": 6.565535204679134e-06,
|
| 2109 |
+
"loss": 0.1449,
|
| 2110 |
+
"step": 300
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.2235772357723578,
|
| 2114 |
+
"grad_norm": 1.074737310409546,
|
| 2115 |
+
"learning_rate": 6.545084971874738e-06,
|
| 2116 |
+
"loss": 0.1396,
|
| 2117 |
+
"step": 301
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.2276422764227641,
|
| 2121 |
+
"grad_norm": 0.8973712921142578,
|
| 2122 |
+
"learning_rate": 6.524606123281445e-06,
|
| 2123 |
+
"loss": 0.163,
|
| 2124 |
+
"step": 302
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.2317073170731707,
|
| 2128 |
+
"grad_norm": 0.9884341359138489,
|
| 2129 |
+
"learning_rate": 6.504099038178339e-06,
|
| 2130 |
+
"loss": 0.1362,
|
| 2131 |
+
"step": 303
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.2357723577235773,
|
| 2135 |
+
"grad_norm": 1.3507106304168701,
|
| 2136 |
+
"learning_rate": 6.483564096367452e-06,
|
| 2137 |
+
"loss": 0.1466,
|
| 2138 |
+
"step": 304
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.2398373983739837,
|
| 2142 |
+
"grad_norm": 0.9876003861427307,
|
| 2143 |
+
"learning_rate": 6.463001678166744e-06,
|
| 2144 |
+
"loss": 0.1215,
|
| 2145 |
+
"step": 305
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.2439024390243902,
|
| 2149 |
+
"grad_norm": 1.0030145645141602,
|
| 2150 |
+
"learning_rate": 6.442412164403045e-06,
|
| 2151 |
+
"loss": 0.1487,
|
| 2152 |
+
"step": 306
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.2479674796747968,
|
| 2156 |
+
"grad_norm": 1.130290150642395,
|
| 2157 |
+
"learning_rate": 6.4217959364050196e-06,
|
| 2158 |
+
"loss": 0.1339,
|
| 2159 |
+
"step": 307
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.2520325203252032,
|
| 2163 |
+
"grad_norm": 1.2390751838684082,
|
| 2164 |
+
"learning_rate": 6.401153375996082e-06,
|
| 2165 |
+
"loss": 0.1588,
|
| 2166 |
+
"step": 308
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.2560975609756098,
|
| 2170 |
+
"grad_norm": 0.9830962419509888,
|
| 2171 |
+
"learning_rate": 6.380484865487346e-06,
|
| 2172 |
+
"loss": 0.1378,
|
| 2173 |
+
"step": 309
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.2601626016260163,
|
| 2177 |
+
"grad_norm": 0.874064564704895,
|
| 2178 |
+
"learning_rate": 6.359790787670528e-06,
|
| 2179 |
+
"loss": 0.1555,
|
| 2180 |
+
"step": 310
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.2642276422764227,
|
| 2184 |
+
"grad_norm": 1.0662082433700562,
|
| 2185 |
+
"learning_rate": 6.339071525810871e-06,
|
| 2186 |
+
"loss": 0.141,
|
| 2187 |
+
"step": 311
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.2682926829268293,
|
| 2191 |
+
"grad_norm": 1.0946401357650757,
|
| 2192 |
+
"learning_rate": 6.318327463640037e-06,
|
| 2193 |
+
"loss": 0.1646,
|
| 2194 |
+
"step": 312
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.2723577235772359,
|
| 2198 |
+
"grad_norm": 0.8100759387016296,
|
| 2199 |
+
"learning_rate": 6.297558985348999e-06,
|
| 2200 |
+
"loss": 0.1602,
|
| 2201 |
+
"step": 313
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.2764227642276422,
|
| 2205 |
+
"grad_norm": 1.0006744861602783,
|
| 2206 |
+
"learning_rate": 6.276766475580935e-06,
|
| 2207 |
+
"loss": 0.1247,
|
| 2208 |
+
"step": 314
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.2804878048780488,
|
| 2212 |
+
"grad_norm": 1.2102998495101929,
|
| 2213 |
+
"learning_rate": 6.255950319424098e-06,
|
| 2214 |
+
"loss": 0.1629,
|
| 2215 |
+
"step": 315
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.2845528455284554,
|
| 2219 |
+
"grad_norm": 0.9621366858482361,
|
| 2220 |
+
"learning_rate": 6.235110902404678e-06,
|
| 2221 |
+
"loss": 0.1732,
|
| 2222 |
+
"step": 316
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.2886178861788617,
|
| 2226 |
+
"grad_norm": 1.2745347023010254,
|
| 2227 |
+
"learning_rate": 6.2142486104796795e-06,
|
| 2228 |
+
"loss": 0.1564,
|
| 2229 |
+
"step": 317
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.2926829268292683,
|
| 2233 |
+
"grad_norm": 1.1103649139404297,
|
| 2234 |
+
"learning_rate": 6.193363830029751e-06,
|
| 2235 |
+
"loss": 0.1577,
|
| 2236 |
+
"step": 318
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.296747967479675,
|
| 2240 |
+
"grad_norm": 1.1667190790176392,
|
| 2241 |
+
"learning_rate": 6.1724569478520495e-06,
|
| 2242 |
+
"loss": 0.1698,
|
| 2243 |
+
"step": 319
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.3008130081300813,
|
| 2247 |
+
"grad_norm": 1.0116077661514282,
|
| 2248 |
+
"learning_rate": 6.151528351153061e-06,
|
| 2249 |
+
"loss": 0.1355,
|
| 2250 |
+
"step": 320
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.3048780487804879,
|
| 2254 |
+
"grad_norm": 1.132459282875061,
|
| 2255 |
+
"learning_rate": 6.130578427541441e-06,
|
| 2256 |
+
"loss": 0.154,
|
| 2257 |
+
"step": 321
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.3089430894308944,
|
| 2261 |
+
"grad_norm": 1.5480411052703857,
|
| 2262 |
+
"learning_rate": 6.109607565020828e-06,
|
| 2263 |
+
"loss": 0.1367,
|
| 2264 |
+
"step": 322
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.3130081300813008,
|
| 2268 |
+
"grad_norm": 1.2651666402816772,
|
| 2269 |
+
"learning_rate": 6.08861615198266e-06,
|
| 2270 |
+
"loss": 0.1516,
|
| 2271 |
+
"step": 323
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.3170731707317074,
|
| 2275 |
+
"grad_norm": 1.1959322690963745,
|
| 2276 |
+
"learning_rate": 6.067604577198981e-06,
|
| 2277 |
+
"loss": 0.1477,
|
| 2278 |
+
"step": 324
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.321138211382114,
|
| 2282 |
+
"grad_norm": 0.9572330713272095,
|
| 2283 |
+
"learning_rate": 6.046573229815244e-06,
|
| 2284 |
+
"loss": 0.1281,
|
| 2285 |
+
"step": 325
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.3252032520325203,
|
| 2289 |
+
"grad_norm": 0.9644310474395752,
|
| 2290 |
+
"learning_rate": 6.025522499343097e-06,
|
| 2291 |
+
"loss": 0.1479,
|
| 2292 |
+
"step": 326
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.329268292682927,
|
| 2296 |
+
"grad_norm": 1.2678269147872925,
|
| 2297 |
+
"learning_rate": 6.004452775653178e-06,
|
| 2298 |
+
"loss": 0.1528,
|
| 2299 |
+
"step": 327
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.3333333333333333,
|
| 2303 |
+
"grad_norm": 1.098125696182251,
|
| 2304 |
+
"learning_rate": 5.9833644489678825e-06,
|
| 2305 |
+
"loss": 0.162,
|
| 2306 |
+
"step": 328
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.3373983739837398,
|
| 2310 |
+
"grad_norm": 1.4279584884643555,
|
| 2311 |
+
"learning_rate": 5.96225790985415e-06,
|
| 2312 |
+
"loss": 0.1539,
|
| 2313 |
+
"step": 329
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.3414634146341464,
|
| 2317 |
+
"grad_norm": 0.9917011857032776,
|
| 2318 |
+
"learning_rate": 5.9411335492162215e-06,
|
| 2319 |
+
"loss": 0.1526,
|
| 2320 |
+
"step": 330
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.3455284552845528,
|
| 2324 |
+
"grad_norm": 1.3051934242248535,
|
| 2325 |
+
"learning_rate": 5.919991758288401e-06,
|
| 2326 |
+
"loss": 0.1439,
|
| 2327 |
+
"step": 331
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.3495934959349594,
|
| 2331 |
+
"grad_norm": 1.266823649406433,
|
| 2332 |
+
"learning_rate": 5.898832928627811e-06,
|
| 2333 |
+
"loss": 0.1583,
|
| 2334 |
+
"step": 332
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.3536585365853657,
|
| 2338 |
+
"grad_norm": 1.0603634119033813,
|
| 2339 |
+
"learning_rate": 5.877657452107142e-06,
|
| 2340 |
+
"loss": 0.1537,
|
| 2341 |
+
"step": 333
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.3577235772357723,
|
| 2345 |
+
"grad_norm": 1.5735316276550293,
|
| 2346 |
+
"learning_rate": 5.856465720907388e-06,
|
| 2347 |
+
"loss": 0.1495,
|
| 2348 |
+
"step": 334
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.3617886178861789,
|
| 2352 |
+
"grad_norm": 1.1026643514633179,
|
| 2353 |
+
"learning_rate": 5.835258127510597e-06,
|
| 2354 |
+
"loss": 0.1539,
|
| 2355 |
+
"step": 335
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.3658536585365852,
|
| 2359 |
+
"grad_norm": 1.2997959852218628,
|
| 2360 |
+
"learning_rate": 5.814035064692585e-06,
|
| 2361 |
+
"loss": 0.1706,
|
| 2362 |
+
"step": 336
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.3699186991869918,
|
| 2366 |
+
"grad_norm": 0.9610199332237244,
|
| 2367 |
+
"learning_rate": 5.792796925515673e-06,
|
| 2368 |
+
"loss": 0.151,
|
| 2369 |
+
"step": 337
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.3739837398373984,
|
| 2373 |
+
"grad_norm": 1.3250401020050049,
|
| 2374 |
+
"learning_rate": 5.771544103321407e-06,
|
| 2375 |
+
"loss": 0.1534,
|
| 2376 |
+
"step": 338
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.3780487804878048,
|
| 2380 |
+
"grad_norm": 1.105961799621582,
|
| 2381 |
+
"learning_rate": 5.7502769917232635e-06,
|
| 2382 |
+
"loss": 0.1491,
|
| 2383 |
+
"step": 339
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.3821138211382114,
|
| 2387 |
+
"grad_norm": 1.2238012552261353,
|
| 2388 |
+
"learning_rate": 5.728995984599374e-06,
|
| 2389 |
+
"loss": 0.1486,
|
| 2390 |
+
"step": 340
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.386178861788618,
|
| 2394 |
+
"grad_norm": 1.3362035751342773,
|
| 2395 |
+
"learning_rate": 5.707701476085217e-06,
|
| 2396 |
+
"loss": 0.1425,
|
| 2397 |
+
"step": 341
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.3902439024390243,
|
| 2401 |
+
"grad_norm": 1.085087537765503,
|
| 2402 |
+
"learning_rate": 5.686393860566324e-06,
|
| 2403 |
+
"loss": 0.1636,
|
| 2404 |
+
"step": 342
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.3943089430894309,
|
| 2408 |
+
"grad_norm": 1.3547860383987427,
|
| 2409 |
+
"learning_rate": 5.665073532670981e-06,
|
| 2410 |
+
"loss": 0.1531,
|
| 2411 |
+
"step": 343
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.3983739837398375,
|
| 2415 |
+
"grad_norm": 1.322536826133728,
|
| 2416 |
+
"learning_rate": 5.643740887262905e-06,
|
| 2417 |
+
"loss": 0.1705,
|
| 2418 |
+
"step": 344
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.4024390243902438,
|
| 2422 |
+
"grad_norm": 1.149080514907837,
|
| 2423 |
+
"learning_rate": 5.622396319433947e-06,
|
| 2424 |
+
"loss": 0.1272,
|
| 2425 |
+
"step": 345
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.4065040650406504,
|
| 2429 |
+
"grad_norm": 1.228470802307129,
|
| 2430 |
+
"learning_rate": 5.601040224496764e-06,
|
| 2431 |
+
"loss": 0.1741,
|
| 2432 |
+
"step": 346
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.410569105691057,
|
| 2436 |
+
"grad_norm": 0.9749687314033508,
|
| 2437 |
+
"learning_rate": 5.5796729979775035e-06,
|
| 2438 |
+
"loss": 0.1584,
|
| 2439 |
+
"step": 347
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.4146341463414633,
|
| 2443 |
+
"grad_norm": 0.9834906458854675,
|
| 2444 |
+
"learning_rate": 5.558295035608473e-06,
|
| 2445 |
+
"loss": 0.1483,
|
| 2446 |
+
"step": 348
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.41869918699187,
|
| 2450 |
+
"grad_norm": 1.096159815788269,
|
| 2451 |
+
"learning_rate": 5.536906733320816e-06,
|
| 2452 |
+
"loss": 0.1581,
|
| 2453 |
+
"step": 349
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.4227642276422765,
|
| 2457 |
+
"grad_norm": 0.9886683821678162,
|
| 2458 |
+
"learning_rate": 5.515508487237174e-06,
|
| 2459 |
+
"loss": 0.1678,
|
| 2460 |
+
"step": 350
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.4268292682926829,
|
| 2464 |
+
"grad_norm": 0.8707095384597778,
|
| 2465 |
+
"learning_rate": 5.494100693664358e-06,
|
| 2466 |
+
"loss": 0.1305,
|
| 2467 |
+
"step": 351
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.4308943089430894,
|
| 2471 |
+
"grad_norm": 1.0057952404022217,
|
| 2472 |
+
"learning_rate": 5.472683749085996e-06,
|
| 2473 |
+
"loss": 0.1659,
|
| 2474 |
+
"step": 352
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.434959349593496,
|
| 2478 |
+
"grad_norm": 1.3434487581253052,
|
| 2479 |
+
"learning_rate": 5.451258050155206e-06,
|
| 2480 |
+
"loss": 0.1402,
|
| 2481 |
+
"step": 353
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.4390243902439024,
|
| 2485 |
+
"grad_norm": 1.1209443807601929,
|
| 2486 |
+
"learning_rate": 5.429823993687234e-06,
|
| 2487 |
+
"loss": 0.1374,
|
| 2488 |
+
"step": 354
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.443089430894309,
|
| 2492 |
+
"grad_norm": 1.3083735704421997,
|
| 2493 |
+
"learning_rate": 5.408381976652114e-06,
|
| 2494 |
+
"loss": 0.167,
|
| 2495 |
+
"step": 355
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.4471544715447155,
|
| 2499 |
+
"grad_norm": 1.130519986152649,
|
| 2500 |
+
"learning_rate": 5.3869323961673155e-06,
|
| 2501 |
+
"loss": 0.1354,
|
| 2502 |
+
"step": 356
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.451219512195122,
|
| 2506 |
+
"grad_norm": 1.058443307876587,
|
| 2507 |
+
"learning_rate": 5.365475649490388e-06,
|
| 2508 |
+
"loss": 0.1543,
|
| 2509 |
+
"step": 357
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.4552845528455285,
|
| 2513 |
+
"grad_norm": 0.9692599177360535,
|
| 2514 |
+
"learning_rate": 5.3440121340116e-06,
|
| 2515 |
+
"loss": 0.1664,
|
| 2516 |
+
"step": 358
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.459349593495935,
|
| 2520 |
+
"grad_norm": 1.0753322839736938,
|
| 2521 |
+
"learning_rate": 5.322542247246583e-06,
|
| 2522 |
+
"loss": 0.168,
|
| 2523 |
+
"step": 359
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.4634146341463414,
|
| 2527 |
+
"grad_norm": 0.9880397915840149,
|
| 2528 |
+
"learning_rate": 5.3010663868289655e-06,
|
| 2529 |
+
"loss": 0.1765,
|
| 2530 |
+
"step": 360
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 1.467479674796748,
|
| 2534 |
+
"grad_norm": 1.0291845798492432,
|
| 2535 |
+
"learning_rate": 5.279584950503017e-06,
|
| 2536 |
+
"loss": 0.146,
|
| 2537 |
+
"step": 361
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 1.4715447154471546,
|
| 2541 |
+
"grad_norm": 1.2091126441955566,
|
| 2542 |
+
"learning_rate": 5.258098336116269e-06,
|
| 2543 |
+
"loss": 0.1642,
|
| 2544 |
+
"step": 362
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 1.475609756097561,
|
| 2548 |
+
"grad_norm": 0.8923934698104858,
|
| 2549 |
+
"learning_rate": 5.2366069416121605e-06,
|
| 2550 |
+
"loss": 0.1357,
|
| 2551 |
+
"step": 363
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 1.4796747967479675,
|
| 2555 |
+
"grad_norm": 0.9851050972938538,
|
| 2556 |
+
"learning_rate": 5.215111165022653e-06,
|
| 2557 |
+
"loss": 0.1426,
|
| 2558 |
+
"step": 364
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 1.4837398373983741,
|
| 2562 |
+
"grad_norm": 1.1078354120254517,
|
| 2563 |
+
"learning_rate": 5.193611404460873e-06,
|
| 2564 |
+
"loss": 0.1437,
|
| 2565 |
+
"step": 365
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 1.4878048780487805,
|
| 2569 |
+
"grad_norm": 0.8380599021911621,
|
| 2570 |
+
"learning_rate": 5.17210805811373e-06,
|
| 2571 |
+
"loss": 0.1687,
|
| 2572 |
+
"step": 366
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 1.491869918699187,
|
| 2576 |
+
"grad_norm": 1.0096312761306763,
|
| 2577 |
+
"learning_rate": 5.150601524234542e-06,
|
| 2578 |
+
"loss": 0.1394,
|
| 2579 |
+
"step": 367
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 1.4959349593495934,
|
| 2583 |
+
"grad_norm": 0.9913772344589233,
|
| 2584 |
+
"learning_rate": 5.129092201135666e-06,
|
| 2585 |
+
"loss": 0.1612,
|
| 2586 |
+
"step": 368
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 1.5,
|
| 2590 |
+
"grad_norm": 0.9858372211456299,
|
| 2591 |
+
"learning_rate": 5.107580487181112e-06,
|
| 2592 |
+
"loss": 0.1506,
|
| 2593 |
+
"step": 369
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 1.5040650406504064,
|
| 2597 |
+
"grad_norm": 1.043803334236145,
|
| 2598 |
+
"learning_rate": 5.086066780779174e-06,
|
| 2599 |
+
"loss": 0.1674,
|
| 2600 |
+
"step": 370
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 1.5081300813008132,
|
| 2604 |
+
"grad_norm": 1.0286842584609985,
|
| 2605 |
+
"learning_rate": 5.064551480375046e-06,
|
| 2606 |
+
"loss": 0.1464,
|
| 2607 |
+
"step": 371
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 1.5121951219512195,
|
| 2611 |
+
"grad_norm": 1.1554076671600342,
|
| 2612 |
+
"learning_rate": 5.043034984443443e-06,
|
| 2613 |
+
"loss": 0.1617,
|
| 2614 |
+
"step": 372
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 1.5162601626016259,
|
| 2618 |
+
"grad_norm": 0.9859683513641357,
|
| 2619 |
+
"learning_rate": 5.0215176914812215e-06,
|
| 2620 |
+
"loss": 0.1419,
|
| 2621 |
+
"step": 373
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 1.5203252032520327,
|
| 2625 |
+
"grad_norm": 0.817907452583313,
|
| 2626 |
+
"learning_rate": 5e-06,
|
| 2627 |
+
"loss": 0.1465,
|
| 2628 |
+
"step": 374
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.524390243902439,
|
| 2632 |
+
"grad_norm": 0.9574335813522339,
|
| 2633 |
+
"learning_rate": 4.978482308518779e-06,
|
| 2634 |
+
"loss": 0.1527,
|
| 2635 |
+
"step": 375
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 1.5284552845528454,
|
| 2639 |
+
"grad_norm": 0.972829282283783,
|
| 2640 |
+
"learning_rate": 4.956965015556559e-06,
|
| 2641 |
+
"loss": 0.1572,
|
| 2642 |
+
"step": 376
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 1.532520325203252,
|
| 2646 |
+
"grad_norm": 1.4624630212783813,
|
| 2647 |
+
"learning_rate": 4.935448519624955e-06,
|
| 2648 |
+
"loss": 0.1553,
|
| 2649 |
+
"step": 377
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 1.5365853658536586,
|
| 2653 |
+
"grad_norm": 0.9278729557991028,
|
| 2654 |
+
"learning_rate": 4.913933219220827e-06,
|
| 2655 |
+
"loss": 0.169,
|
| 2656 |
+
"step": 378
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 1.540650406504065,
|
| 2660 |
+
"grad_norm": 1.143812656402588,
|
| 2661 |
+
"learning_rate": 4.89241951281889e-06,
|
| 2662 |
+
"loss": 0.1413,
|
| 2663 |
+
"step": 379
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 1.5447154471544715,
|
| 2667 |
+
"grad_norm": 0.8590518236160278,
|
| 2668 |
+
"learning_rate": 4.870907798864337e-06,
|
| 2669 |
+
"loss": 0.1507,
|
| 2670 |
+
"step": 380
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 1.548780487804878,
|
| 2674 |
+
"grad_norm": 0.8633162975311279,
|
| 2675 |
+
"learning_rate": 4.84939847576546e-06,
|
| 2676 |
+
"loss": 0.1737,
|
| 2677 |
+
"step": 381
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 1.5528455284552845,
|
| 2681 |
+
"grad_norm": 0.8456003665924072,
|
| 2682 |
+
"learning_rate": 4.827891941886273e-06,
|
| 2683 |
+
"loss": 0.122,
|
| 2684 |
+
"step": 382
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.556910569105691,
|
| 2688 |
+
"grad_norm": 1.0213592052459717,
|
| 2689 |
+
"learning_rate": 4.8063885955391285e-06,
|
| 2690 |
+
"loss": 0.1448,
|
| 2691 |
+
"step": 383
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 1.5609756097560976,
|
| 2695 |
+
"grad_norm": 1.1720526218414307,
|
| 2696 |
+
"learning_rate": 4.784888834977347e-06,
|
| 2697 |
+
"loss": 0.1409,
|
| 2698 |
+
"step": 384
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 1.565040650406504,
|
| 2702 |
+
"grad_norm": 0.9872570633888245,
|
| 2703 |
+
"learning_rate": 4.763393058387841e-06,
|
| 2704 |
+
"loss": 0.1504,
|
| 2705 |
+
"step": 385
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 1.5691056910569106,
|
| 2709 |
+
"grad_norm": 0.8872116208076477,
|
| 2710 |
+
"learning_rate": 4.7419016638837316e-06,
|
| 2711 |
+
"loss": 0.1409,
|
| 2712 |
+
"step": 386
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 1.5731707317073171,
|
| 2716 |
+
"grad_norm": 0.984969437122345,
|
| 2717 |
+
"learning_rate": 4.720415049496984e-06,
|
| 2718 |
+
"loss": 0.1403,
|
| 2719 |
+
"step": 387
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 1.5772357723577235,
|
| 2723 |
+
"grad_norm": 0.9971374869346619,
|
| 2724 |
+
"learning_rate": 4.698933613171035e-06,
|
| 2725 |
+
"loss": 0.1402,
|
| 2726 |
+
"step": 388
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 1.58130081300813,
|
| 2730 |
+
"grad_norm": 0.9045475125312805,
|
| 2731 |
+
"learning_rate": 4.6774577527534195e-06,
|
| 2732 |
+
"loss": 0.1378,
|
| 2733 |
+
"step": 389
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 1.5853658536585367,
|
| 2737 |
+
"grad_norm": 1.0618966817855835,
|
| 2738 |
+
"learning_rate": 4.655987865988401e-06,
|
| 2739 |
+
"loss": 0.1431,
|
| 2740 |
+
"step": 390
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 1.589430894308943,
|
| 2744 |
+
"grad_norm": 0.922252893447876,
|
| 2745 |
+
"learning_rate": 4.634524350509613e-06,
|
| 2746 |
+
"loss": 0.1424,
|
| 2747 |
+
"step": 391
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 1.5934959349593496,
|
| 2751 |
+
"grad_norm": 1.024962067604065,
|
| 2752 |
+
"learning_rate": 4.613067603832685e-06,
|
| 2753 |
+
"loss": 0.1338,
|
| 2754 |
+
"step": 392
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 1.5975609756097562,
|
| 2758 |
+
"grad_norm": 0.9654287099838257,
|
| 2759 |
+
"learning_rate": 4.591618023347888e-06,
|
| 2760 |
+
"loss": 0.1397,
|
| 2761 |
+
"step": 393
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 1.6016260162601625,
|
| 2765 |
+
"grad_norm": 0.8813850283622742,
|
| 2766 |
+
"learning_rate": 4.570176006312769e-06,
|
| 2767 |
+
"loss": 0.1436,
|
| 2768 |
+
"step": 394
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 1.6056910569105691,
|
| 2772 |
+
"grad_norm": 0.8974111676216125,
|
| 2773 |
+
"learning_rate": 4.548741949844795e-06,
|
| 2774 |
+
"loss": 0.1634,
|
| 2775 |
+
"step": 395
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 1.6097560975609757,
|
| 2779 |
+
"grad_norm": 1.0668531656265259,
|
| 2780 |
+
"learning_rate": 4.527316250914005e-06,
|
| 2781 |
+
"loss": 0.1461,
|
| 2782 |
+
"step": 396
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 1.613821138211382,
|
| 2786 |
+
"grad_norm": 0.8691629767417908,
|
| 2787 |
+
"learning_rate": 4.505899306335643e-06,
|
| 2788 |
+
"loss": 0.1699,
|
| 2789 |
+
"step": 397
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 1.6178861788617886,
|
| 2793 |
+
"grad_norm": 0.7588362097740173,
|
| 2794 |
+
"learning_rate": 4.4844915127628266e-06,
|
| 2795 |
+
"loss": 0.157,
|
| 2796 |
+
"step": 398
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 1.6219512195121952,
|
| 2800 |
+
"grad_norm": 1.2434402704238892,
|
| 2801 |
+
"learning_rate": 4.463093266679185e-06,
|
| 2802 |
+
"loss": 0.1494,
|
| 2803 |
+
"step": 399
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 1.6260162601626016,
|
| 2807 |
+
"grad_norm": 0.8778071403503418,
|
| 2808 |
+
"learning_rate": 4.441704964391529e-06,
|
| 2809 |
+
"loss": 0.1439,
|
| 2810 |
+
"step": 400
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 1.6300813008130082,
|
| 2814 |
+
"grad_norm": 1.1034812927246094,
|
| 2815 |
+
"learning_rate": 4.420327002022498e-06,
|
| 2816 |
+
"loss": 0.1788,
|
| 2817 |
+
"step": 401
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 1.6341463414634148,
|
| 2821 |
+
"grad_norm": 0.9097403883934021,
|
| 2822 |
+
"learning_rate": 4.398959775503238e-06,
|
| 2823 |
+
"loss": 0.1396,
|
| 2824 |
+
"step": 402
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 1.6382113821138211,
|
| 2828 |
+
"grad_norm": 0.9956079721450806,
|
| 2829 |
+
"learning_rate": 4.377603680566054e-06,
|
| 2830 |
+
"loss": 0.1415,
|
| 2831 |
+
"step": 403
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 1.6422764227642277,
|
| 2835 |
+
"grad_norm": 1.2590630054473877,
|
| 2836 |
+
"learning_rate": 4.356259112737096e-06,
|
| 2837 |
+
"loss": 0.1518,
|
| 2838 |
+
"step": 404
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 1.6463414634146343,
|
| 2842 |
+
"grad_norm": 1.0754262208938599,
|
| 2843 |
+
"learning_rate": 4.334926467329021e-06,
|
| 2844 |
+
"loss": 0.1371,
|
| 2845 |
+
"step": 405
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 1.6504065040650406,
|
| 2849 |
+
"grad_norm": 1.0419645309448242,
|
| 2850 |
+
"learning_rate": 4.313606139433676e-06,
|
| 2851 |
+
"loss": 0.1566,
|
| 2852 |
+
"step": 406
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 1.654471544715447,
|
| 2856 |
+
"grad_norm": 0.780151903629303,
|
| 2857 |
+
"learning_rate": 4.292298523914785e-06,
|
| 2858 |
+
"loss": 0.1387,
|
| 2859 |
+
"step": 407
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 1.6585365853658538,
|
| 2863 |
+
"grad_norm": 1.0169719457626343,
|
| 2864 |
+
"learning_rate": 4.271004015400627e-06,
|
| 2865 |
+
"loss": 0.1323,
|
| 2866 |
+
"step": 408
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 1.6626016260162602,
|
| 2870 |
+
"grad_norm": 1.016371726989746,
|
| 2871 |
+
"learning_rate": 4.249723008276737e-06,
|
| 2872 |
+
"loss": 0.1396,
|
| 2873 |
+
"step": 409
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 1.6666666666666665,
|
| 2877 |
+
"grad_norm": 0.7959016561508179,
|
| 2878 |
+
"learning_rate": 4.228455896678595e-06,
|
| 2879 |
+
"loss": 0.134,
|
| 2880 |
+
"step": 410
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 1.6707317073170733,
|
| 2884 |
+
"grad_norm": 0.7987192869186401,
|
| 2885 |
+
"learning_rate": 4.2072030744843284e-06,
|
| 2886 |
+
"loss": 0.1394,
|
| 2887 |
+
"step": 411
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 1.6747967479674797,
|
| 2891 |
+
"grad_norm": 1.042298436164856,
|
| 2892 |
+
"learning_rate": 4.185964935307417e-06,
|
| 2893 |
+
"loss": 0.1833,
|
| 2894 |
+
"step": 412
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 1.678861788617886,
|
| 2898 |
+
"grad_norm": 0.9095625877380371,
|
| 2899 |
+
"learning_rate": 4.164741872489405e-06,
|
| 2900 |
+
"loss": 0.1621,
|
| 2901 |
+
"step": 413
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 1.6829268292682928,
|
| 2905 |
+
"grad_norm": 1.0999423265457153,
|
| 2906 |
+
"learning_rate": 4.143534279092613e-06,
|
| 2907 |
+
"loss": 0.143,
|
| 2908 |
+
"step": 414
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 1.6869918699186992,
|
| 2912 |
+
"grad_norm": 0.9429342150688171,
|
| 2913 |
+
"learning_rate": 4.12234254789286e-06,
|
| 2914 |
+
"loss": 0.1418,
|
| 2915 |
+
"step": 415
|
| 2916 |
+
},
|
| 2917 |
+
{
|
| 2918 |
+
"epoch": 1.6910569105691056,
|
| 2919 |
+
"grad_norm": 0.7311047911643982,
|
| 2920 |
+
"learning_rate": 4.101167071372191e-06,
|
| 2921 |
+
"loss": 0.1468,
|
| 2922 |
+
"step": 416
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 1.6951219512195121,
|
| 2926 |
+
"grad_norm": 1.0672001838684082,
|
| 2927 |
+
"learning_rate": 4.0800082417116e-06,
|
| 2928 |
+
"loss": 0.1795,
|
| 2929 |
+
"step": 417
|
| 2930 |
+
},
|
| 2931 |
+
{
|
| 2932 |
+
"epoch": 1.6991869918699187,
|
| 2933 |
+
"grad_norm": 0.9126365780830383,
|
| 2934 |
+
"learning_rate": 4.05886645078378e-06,
|
| 2935 |
+
"loss": 0.1332,
|
| 2936 |
+
"step": 418
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 1.703252032520325,
|
| 2940 |
+
"grad_norm": 1.058180570602417,
|
| 2941 |
+
"learning_rate": 4.037742090145851e-06,
|
| 2942 |
+
"loss": 0.1843,
|
| 2943 |
+
"step": 419
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 1.7073170731707317,
|
| 2947 |
+
"grad_norm": 0.9689294099807739,
|
| 2948 |
+
"learning_rate": 4.01663555103212e-06,
|
| 2949 |
+
"loss": 0.1649,
|
| 2950 |
+
"step": 420
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 1.7113821138211383,
|
| 2954 |
+
"grad_norm": 0.7963077425956726,
|
| 2955 |
+
"learning_rate": 3.995547224346824e-06,
|
| 2956 |
+
"loss": 0.1531,
|
| 2957 |
+
"step": 421
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 1.7154471544715446,
|
| 2961 |
+
"grad_norm": 0.7422528266906738,
|
| 2962 |
+
"learning_rate": 3.974477500656906e-06,
|
| 2963 |
+
"loss": 0.1378,
|
| 2964 |
+
"step": 422
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 1.7195121951219512,
|
| 2968 |
+
"grad_norm": 0.8430778980255127,
|
| 2969 |
+
"learning_rate": 3.953426770184757e-06,
|
| 2970 |
+
"loss": 0.1432,
|
| 2971 |
+
"step": 423
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 1.7235772357723578,
|
| 2975 |
+
"grad_norm": 0.8409215807914734,
|
| 2976 |
+
"learning_rate": 3.93239542280102e-06,
|
| 2977 |
+
"loss": 0.1552,
|
| 2978 |
+
"step": 424
|
| 2979 |
+
},
|
| 2980 |
+
{
|
| 2981 |
+
"epoch": 1.7276422764227641,
|
| 2982 |
+
"grad_norm": 0.9503300786018372,
|
| 2983 |
+
"learning_rate": 3.9113838480173415e-06,
|
| 2984 |
+
"loss": 0.1494,
|
| 2985 |
+
"step": 425
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 1.7317073170731707,
|
| 2989 |
+
"grad_norm": 1.1325773000717163,
|
| 2990 |
+
"learning_rate": 3.890392434979172e-06,
|
| 2991 |
+
"loss": 0.1448,
|
| 2992 |
+
"step": 426
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 1.7357723577235773,
|
| 2996 |
+
"grad_norm": 0.9397405385971069,
|
| 2997 |
+
"learning_rate": 3.8694215724585596e-06,
|
| 2998 |
+
"loss": 0.1311,
|
| 2999 |
+
"step": 427
|
| 3000 |
+
},
|
| 3001 |
+
{
|
| 3002 |
+
"epoch": 1.7398373983739837,
|
| 3003 |
+
"grad_norm": 1.071285605430603,
|
| 3004 |
+
"learning_rate": 3.848471648846939e-06,
|
| 3005 |
+
"loss": 0.1356,
|
| 3006 |
+
"step": 428
|
| 3007 |
+
},
|
| 3008 |
+
{
|
| 3009 |
+
"epoch": 1.7439024390243902,
|
| 3010 |
+
"grad_norm": 0.9711349606513977,
|
| 3011 |
+
"learning_rate": 3.827543052147952e-06,
|
| 3012 |
+
"loss": 0.1321,
|
| 3013 |
+
"step": 429
|
| 3014 |
+
},
|
| 3015 |
+
{
|
| 3016 |
+
"epoch": 1.7479674796747968,
|
| 3017 |
+
"grad_norm": 0.8763346672058105,
|
| 3018 |
+
"learning_rate": 3.80663616997025e-06,
|
| 3019 |
+
"loss": 0.1487,
|
| 3020 |
+
"step": 430
|
| 3021 |
+
},
|
| 3022 |
+
{
|
| 3023 |
+
"epoch": 1.7520325203252032,
|
| 3024 |
+
"grad_norm": 0.9866912961006165,
|
| 3025 |
+
"learning_rate": 3.785751389520323e-06,
|
| 3026 |
+
"loss": 0.163,
|
| 3027 |
+
"step": 431
|
| 3028 |
+
},
|
| 3029 |
+
{
|
| 3030 |
+
"epoch": 1.7560975609756098,
|
| 3031 |
+
"grad_norm": 1.1801891326904297,
|
| 3032 |
+
"learning_rate": 3.764889097595322e-06,
|
| 3033 |
+
"loss": 0.148,
|
| 3034 |
+
"step": 432
|
| 3035 |
+
},
|
| 3036 |
+
{
|
| 3037 |
+
"epoch": 1.7601626016260163,
|
| 3038 |
+
"grad_norm": 0.956091582775116,
|
| 3039 |
+
"learning_rate": 3.7440496805759048e-06,
|
| 3040 |
+
"loss": 0.1287,
|
| 3041 |
+
"step": 433
|
| 3042 |
+
},
|
| 3043 |
+
{
|
| 3044 |
+
"epoch": 1.7642276422764227,
|
| 3045 |
+
"grad_norm": 0.8202951550483704,
|
| 3046 |
+
"learning_rate": 3.7232335244190656e-06,
|
| 3047 |
+
"loss": 0.1383,
|
| 3048 |
+
"step": 434
|
| 3049 |
+
},
|
| 3050 |
+
{
|
| 3051 |
+
"epoch": 1.7682926829268293,
|
| 3052 |
+
"grad_norm": 1.0913454294204712,
|
| 3053 |
+
"learning_rate": 3.7024410146510014e-06,
|
| 3054 |
+
"loss": 0.1423,
|
| 3055 |
+
"step": 435
|
| 3056 |
+
},
|
| 3057 |
+
{
|
| 3058 |
+
"epoch": 1.7723577235772359,
|
| 3059 |
+
"grad_norm": 0.9787104725837708,
|
| 3060 |
+
"learning_rate": 3.6816725363599646e-06,
|
| 3061 |
+
"loss": 0.1312,
|
| 3062 |
+
"step": 436
|
| 3063 |
+
},
|
| 3064 |
+
{
|
| 3065 |
+
"epoch": 1.7764227642276422,
|
| 3066 |
+
"grad_norm": 0.8662833571434021,
|
| 3067 |
+
"learning_rate": 3.6609284741891295e-06,
|
| 3068 |
+
"loss": 0.1371,
|
| 3069 |
+
"step": 437
|
| 3070 |
+
},
|
| 3071 |
+
{
|
| 3072 |
+
"epoch": 1.7804878048780488,
|
| 3073 |
+
"grad_norm": 0.910179853439331,
|
| 3074 |
+
"learning_rate": 3.6402092123294736e-06,
|
| 3075 |
+
"loss": 0.1542,
|
| 3076 |
+
"step": 438
|
| 3077 |
+
},
|
| 3078 |
+
{
|
| 3079 |
+
"epoch": 1.7845528455284554,
|
| 3080 |
+
"grad_norm": 0.9869939684867859,
|
| 3081 |
+
"learning_rate": 3.6195151345126556e-06,
|
| 3082 |
+
"loss": 0.1632,
|
| 3083 |
+
"step": 439
|
| 3084 |
+
},
|
| 3085 |
+
{
|
| 3086 |
+
"epoch": 1.7886178861788617,
|
| 3087 |
+
"grad_norm": 1.0463093519210815,
|
| 3088 |
+
"learning_rate": 3.5988466240039212e-06,
|
| 3089 |
+
"loss": 0.1355,
|
| 3090 |
+
"step": 440
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"epoch": 1.7926829268292683,
|
| 3094 |
+
"grad_norm": 0.9755829572677612,
|
| 3095 |
+
"learning_rate": 3.578204063594982e-06,
|
| 3096 |
+
"loss": 0.1271,
|
| 3097 |
+
"step": 441
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 1.796747967479675,
|
| 3101 |
+
"grad_norm": 1.1465586423873901,
|
| 3102 |
+
"learning_rate": 3.5575878355969567e-06,
|
| 3103 |
+
"loss": 0.1487,
|
| 3104 |
+
"step": 442
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 1.8008130081300813,
|
| 3108 |
+
"grad_norm": 0.9904010891914368,
|
| 3109 |
+
"learning_rate": 3.536998321833258e-06,
|
| 3110 |
+
"loss": 0.1645,
|
| 3111 |
+
"step": 443
|
| 3112 |
+
},
|
| 3113 |
+
{
|
| 3114 |
+
"epoch": 1.8048780487804879,
|
| 3115 |
+
"grad_norm": 0.9410357475280762,
|
| 3116 |
+
"learning_rate": 3.5164359036325483e-06,
|
| 3117 |
+
"loss": 0.1528,
|
| 3118 |
+
"step": 444
|
| 3119 |
+
}
|
| 3120 |
+
],
|
| 3121 |
+
"logging_steps": 1.0,
|
| 3122 |
+
"max_steps": 738,
|
| 3123 |
+
"num_input_tokens_seen": 0,
|
| 3124 |
+
"num_train_epochs": 3,
|
| 3125 |
+
"save_steps": 148,
|
| 3126 |
+
"stateful_callbacks": {
|
| 3127 |
+
"TrainerControl": {
|
| 3128 |
+
"args": {
|
| 3129 |
+
"should_epoch_stop": false,
|
| 3130 |
+
"should_evaluate": false,
|
| 3131 |
+
"should_log": false,
|
| 3132 |
+
"should_save": true,
|
| 3133 |
+
"should_training_stop": false
|
| 3134 |
+
},
|
| 3135 |
+
"attributes": {}
|
| 3136 |
+
}
|
| 3137 |
+
},
|
| 3138 |
+
"total_flos": 7.743369984323092e+18,
|
| 3139 |
+
"train_batch_size": 2,
|
| 3140 |
+
"trial_name": null,
|
| 3141 |
+
"trial_params": null
|
| 3142 |
+
}
|