Unverified50% confidenceFactExact time
After turning off this neuron, the response of the "flames and fox faces" neuron in the next layer dropped by about 15% on cat faces and fox faces, but among the 75 photos tested, the network's final answer did not change for a single one.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/22/2026
First Seen
Valid until: 1/20/2027
Sources
One Neuron Detects Both Cats and Cars? We Reproduced It and Found a New Explanation
youtubeUp the Tech Tree10/16/2026
Related Entities
Related Claims
UnverifiedIn leave-one-family-out evaluation, the selected neurons still retain 86%–94% of the full-feature ceiling on generator families never seen during selection.60% similarUnverifiedLeaky ReLU outputs a small proportion (usually 1%) of the input value for negative input, thereby avoiding the dead neuron problem.60% similarUnverifiedThe skip-zero optimization (exploiting that 92.5% of neurons are zero after ReLU², reading down weights in groups of 16 neurons) raised speed from 6.6 tok/s to 7.3-7.6 tok/s, an improvement of about 12%.58% similarUnverifiedRunning the same pointy-versus-round comparison across all 832 neurons in that layer, about 100 neurons prefer pointy things more than the target neuron does, showing that this preference is not rare.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1085062API
curl https://kongchang.com/api/v1/knowledge/claims/1085062MCP
get_claim(id=1085062)