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

    Neuronpedia

    INDEX
    Explanations

    words after parenthesis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reproducibility
    -0.77
     Reminis
    -0.70
     CAUTION
    -0.68
     cancelled
    -0.68
    APPLICATIONS
    -0.67
    interrupted
    -0.67
    Fabrication
    -0.66
    Partido
    -0.66
     AUTOMATIC
    -0.66
     deterrence
    -0.65
    POSITIVE LOGITS
    Students
    0.77
    FUNCTION
    0.76
     Cru
    0.75
    atize
    0.74
    ✯
    0.73
    Instruction
    0.73
     Planck
    0.73
    ⁦
    0.72
    ֽ
    0.71
    lak
    0.70
    Act Density 0.114%

    No Known Activations