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

    Neuronpedia

    INDEX
    Explanations

    includegraphics in latex documents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    sizes
    0.45
     SIZE
    0.42
    Sizes
    0.42
    Size
    0.41
    Autor
    0.41
    サイズ
    0.40
     Sizes
    0.39
     autor
    0.39
    Post
    0.38
     enlarg
    0.38
    POSITIVE LOGITS
     trim
    0.65
     keep
    0.64
    trim
    0.64
     Trim
    0.63
     Keep
    0.62
    Trim
    0.59
     trimming
    0.58
     Clip
    0.57
     kept
    0.57
     trims
    0.56
    Act Density 0.000%

    No Known Activations