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

    Neuronpedia

    INDEX
    Explanations

    HelpMessage parameter descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Help
    0.75
     help
    0.75
     aiuto
    0.74
     hjäl
    0.71
    Help
    0.68
     helpers
    0.67
     hjel
    0.67
    help
    0.67
    帮忙
    0.67
     aiut
    0.66
    POSITIVE LOGITS
    subset
    0.47
    merchant
    0.45
    generalized
    0.45
     少
    0.44
    section
    0.43
    texts
    0.42
    backs
    0.41
    contact
    0.41
    subsection
    0.40
    text
    0.40
    Act Density 0.001%

    No Known Activations