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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    apologetic, self-correcting replies that acknowledge a previous mistake or concede the user’s point.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     FR
    -0.06
     migr
    -0.06
     Vanity
    -0.06
     horror
    -0.06
     美
    -0.06
    Democrats
    -0.06
    kehr
    -0.06
    .FileInputStream
    -0.06
     Homer
    -0.06
     purchasing
    -0.06
    POSITIVE LOGITS
    ...";↵
    0.06
     mesmo
    0.06
    ])));↵
    0.06
    登录
    0.06
    -real
    0.06
    }";↵
    0.06
     Commonwealth
    0.06
     Famil
    0.06
    -ret
    0.06
    rico
    0.06
    Act Density 0.039%

    No Known Activations