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

    Neuronpedia

    INDEX
    Explanations

    modelspecial characters or conversational prompts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     desigual
    0.40
    idente
    0.37
     Trying
    0.37
     podrían
    0.36
     ه
    0.36
    henderit
    0.36
    冧
    0.36
     Fool
    0.36
    ポジション
    0.35
    襞
    0.35
    POSITIVE LOGITS
    tub
    0.39
    świet
    0.38
    Android
    0.38
     Noct
    0.37
     TUB
    0.36
    StubCompat
    0.36
    muş
    0.35
     Usu
    0.35
    Section
    0.34
    tj
    0.34
    Act Density 0.000%

    No Known Activations