view post Post 1111 I ran the Anthropic Misalignment Framework for a few top models and added it to a dataset: cfahlgren1/anthropic-agentic-misalignment-resultsYou can read the reasoning traces of the models trying to blackmail the user and perform other actions. It's very interesting!! See translation
Reasoning Datasets Reasoning datasets that are trending 🔥 Tiiny/QWQ-LONGCOT-500K Viewer • Updated Dec 26, 2024 • 286k • 267 • 124 O1-OPEN/OpenO1-SFT Viewer • Updated Apr 22, 2025 • 77.7k • 493 • 387 FreedomIntelligence/medical-o1-reasoning-SFT Viewer • Updated Apr 22, 2025 • 90.1k • 7.23k • 1.08k amphora/QwQ-LongCoT-130K Viewer • Updated Dec 22, 2024 • 133k • 132 • 149
MLC WebLLM Running 33 Phi-3.5-Mini WebLLM ⚡ 33 Chat with a model using text input Running 18 Qwen-2.5 WebLLM ⚡ 18 Chat with a language model using your browser Running 131 WebLLM Playground 🏎 131 Display a React app with TypeScript Running 144 Qwen 2.5 Code Interpreter 🐍 144 Run code and get answers with AI
Reasoning Datasets Reasoning datasets that are trending 🔥 Tiiny/QWQ-LONGCOT-500K Viewer • Updated Dec 26, 2024 • 286k • 267 • 124 O1-OPEN/OpenO1-SFT Viewer • Updated Apr 22, 2025 • 77.7k • 493 • 387 FreedomIntelligence/medical-o1-reasoning-SFT Viewer • Updated Apr 22, 2025 • 90.1k • 7.23k • 1.08k amphora/QwQ-LongCoT-130K Viewer • Updated Dec 22, 2024 • 133k • 132 • 149
MLC WebLLM Running 33 Phi-3.5-Mini WebLLM ⚡ 33 Chat with a model using text input Running 18 Qwen-2.5 WebLLM ⚡ 18 Chat with a language model using your browser Running 131 WebLLM Playground 🏎 131 Display a React app with TypeScript Running 144 Qwen 2.5 Code Interpreter 🐍 144 Run code and get answers with AI