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

    Neuronpedia

    INDEX
    Explanations

    Technical programs and terms

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Whilst
    0.75
     Whilst
    0.65
    ForRow
    0.56
    Transcription
    0.56
    汚
    0.56
    Horn
    0.55
     Transcription
    0.54
     Anon
    0.53
     Charcoal
    0.53
     endearing
    0.52
    POSITIVE LOGITS
    謨
    0.58
     높은
    0.57
     breakout
    0.57
    ahnya
    0.55
    ムー
    0.55
     cancé
    0.55
     ее
    0.55
     заболе
    0.54
    াগত
    0.54
     prévention
    0.54
    Act Density 0.001%

    No Known Activations