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

    Neuronpedia

    INDEX
    Explanations

    release or undo actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Bezirk
    0.77
    輯
    0.76
     replacement
    0.75
     crippled
    0.73
    ামীর
    0.69
    予防
    0.69
     적극
    0.69
    substitution
    0.68
     remplacement
    0.68
     আক্রান্ত
    0.68
    POSITIVE LOGITS
     Release
    1.32
    Release
    1.30
     release
    1.30
    解除
    1.30
     releases
    1.29
     released
    1.28
    解
    1.27
     undo
    1.24
     Undo
    1.22
    release
    1.21
    Act Density 0.223%

    No Known Activations