JayHyeon Qwen 0.5-VIPO 5e-7-1ep 10vpo const

From OODA WIKI
Jump to navigation Jump to search


JayHyeon/Qwen_0.5-VIPO_5e-7-1ep_10vpo_const is a deep learning model with a recorded average score of 0.32 on the Open LLM Leaderboard.

JayHyeon/Qwen_0.5-VIPO_5e-7-1ep_10vpo_const
Extrinsic Performance (LLM Leaderboard)
Rank N/A
Average Score 0.32
Intrinsic Architecture
Architecture Qwen2ForCausalLM
Hidden Layers 24
Attention Heads 14
Vocab Size 151936

Performance Metrics

  • ARC Score: N/A
  • HellaSwag Score: N/A
  • MMLU Score: N/A

This page was last updated automatically by the Almanac Ingestor bot.