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

    Neuronpedia

    INDEX
    Explanations

    chunks of source code, probably HTML, with variable substitutions

    oai_token-act-pair · gemini-2.0-flash

    code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    indicato
    -0.54
    KURZBESCHREIBUNG
    -0.51
     Se
    -0.51
     AT
    -0.49
     n
    -0.49
     Bel
    -0.48
    reactstrap
    -0.48
     f
    -0.48
     i
    -0.47
     Ad
    -0.47
    POSITIVE LOGITS
     tartalomajánló
    0.85
     purpoſe
    0.73
    expandindo
    0.72
     pleaſure
    0.68
     Baillargeon
    0.68
     houſe
    0.67
    ImageContext
    0.66
     ſever
    0.65
     consultato
    0.65
     auffi
    0.60
    Act Density 3.348%

    No Known Activations