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

    Neuronpedia

    INDEX
    Explanations

    meta discussions/self-reference

    np_max-act · gemini-2.0-flash

    Detects self-referential or meta discussion about storytelling/theatre (mentions of stories, plays, writing about writing, and meta-theatrical commentary).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    peech
    -0.07
     uncovered
    -0.07
    爽
    -0.06
    agon
    -0.06
     indicator
    -0.06
     whoever
    -0.06
     boon
    -0.06
     created
    -0.06
    券
    -0.06
     unavoidable
    -0.06
    POSITIVE LOGITS
     bảo
    0.07
     specialised
    0.07
     obsc
    0.06
    apon
    0.06
     waits
    0.06
    une
    0.06
     solder
    0.06
    sec
    0.06
     vandal
    0.06
    /frame
    0.06
    Act Density 0.061%

    No Known Activations