- Downloads last month
- 288
Dataset used to train dagbs/Qwen1.5-Wukong-0.5B-GGUF
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard31.740
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard47.780
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard38.440
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard38.920
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard56.510
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard15.540