violetxi/tb2-eval-qwen3-8b-wm-summary-mixed-clean

tb2-eval-qwen3-8b-wm-summary-mixed-clean: Terminal-Bench 2.0 eval results - Qwen3-8B WM summary mixed clean SFT: Harbor dataset on Hugging Face. Compact Harbor eval results for violetxi/qwen3-8b-terminal-wm-summary-mixed-clean on Terminal-Bench 2.0 using the terminus-2 agent harness. Each dataset…