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

    Neuronpedia

    INDEX
    Explanations

    fair use

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     speaking
    0.44
     alternates
    0.41
     Bour
    0.40
     clothing
    0.40
     Altern
    0.40
     snake
    0.40
     melted
    0.40
     Illuminate
    0.39
     Speaker
    0.39
     Snake
    0.39
    POSITIVE LOGITS
    Ratio
    0.47
    converting
    0.47
     저장
    0.45
     ذخ
    0.45
    Capture
    0.44
    DVR
    0.42
     capture
    0.42
    capture
    0.42
    convert
    0.41
    реза
    0.41
    Act Density 0.038%

    No Known Activations