Running
Why Split Personality
📈
Experiments on AI sycophancy. Mech Interp exploration
bigsnarfdude
vincentoh
AI & ML interests
None yet
Recent Activity
updated a dataset 1 day ago
vincentoh/erdos741ii-lean4-opus-traces updated a dataset 1 day ago
vincentoh/rrma-lean4-agent-traces updated a model 6 days ago
vincentoh/kimina-prover-rrma-sftOrganizations
None yet
models 38
vincentoh/kimina-prover-rrma-sft
Updated • 15
vincentoh/mistral-7b-af-organism
Text Generation • Updated • 1
vincentoh/gpt-oss-20b-af-detector
Text Generation • Updated • 6
vincentoh/sycophant-70b-Q4_K_M-GGUF
71B • Updated • 10
vincentoh/qwen3-14b-alignment-faking-detector
Text Generation • Updated
vincentoh/llama-8b-af-detector
Text Generation • Updated
vincentoh/gemma3-27b-af-detector-v2
Text Classification • Updated • 2
vincentoh/gemma3-27b-af-detector-lora
Text Classification • Updated • 3
vincentoh/gemma3-27b-af-detector
Text Classification • Updated • 3
vincentoh/gemma3-4b-af-detector
Text Classification • Updated • 1
datasets 13
vincentoh/erdos741ii-lean4-opus-traces
Viewer • Updated • 39 • 27
vincentoh/rrma-lean4-agent-traces
Viewer • Updated • 416 • 32
vincentoh/victorian-authority-mcq
Viewer • Updated • 426 • 30
vincentoh/mindreader-v2-probe-training
Viewer • Updated • 6.1k • 19
vincentoh/sandbagging-agent-traces-v2
Viewer • Updated • 2.79k • 55
vincentoh/sandbagging-agent-traces
Viewer • Updated • 3.19k • 39
vincentoh/persona-af-elicitation
Viewer • Updated • 450 • 9 • 1
vincentoh/alignment-faking-v1.1
Updated • 22
vincentoh/alignment-faking-evaluation
Viewer • Updated • 5.23k • 8
vincentoh/af-model-organisms
Updated • 9