Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    opening curly braces in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Code snippets and unusual characters

    np_acts-logits-general · gemini-2.0-flash

    right across

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vnode
    -0.70
     Cuth
    -0.59
     mtrl
    -0.58
     Efq
    -0.57
     strdup
    -0.57
     itſelf
    -0.57
    imageio
    -0.55
     xPos
    -0.55
     Ols
    -0.54
     objs
    -0.54
    POSITIVE LOGITS
    ↵
    1.36
    }))
    1.30
    __':
    1.28
    ↵↵
    1.22
    __":
    1.22
     }));
    1.19
    __':
    1.18
    }}],
    1.16
    ])));
    1.16
    })),
    1.14
    Act Density 0.097%

    No Known Activations