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

    Neuronpedia

    INDEX
    Explanations

    hook and related phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    س
    2.53
    い
    2.48
    и
    2.13
    г
    2.11
    ב
    2.03
    с
    1.99
    ра
    1.94
     fleste
    1.88
    ación
    1.86
    д
    1.84
    POSITIVE LOGITS
     gange
    2.41
    m
    2.19
    SO
    1.88
    ie
    1.84
    𝓈
    1.81
    zhen
    1.80
    ah
    1.76
     dáng
    1.75
    EST
    1.73
    NUMX
    1.71
    Act Density 0.031%

    No Known Activations