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

    Neuronpedia

    INDEX
    Explanations

    factors, options, differences, routes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     podczas
    0.86
     terwijl
    0.80
     mentre
    0.75
     tämän
    0.74
    𝓙
    0.73
     अराउंड
    0.73
     quiere
    0.72
     mientras
    0.70
     kterou
    0.70
     하며
    0.69
    POSITIVE LOGITS
     implications
    0.88
     emphasis
    0.86
     exceptions
    0.80
     repercussions
    0.79
     proviso
    0.75
     impunity
    0.74
     exception
    0.73
     caveats
    0.71
     added
    0.71
     regard
    0.70
    Act Density 0.282%

    No Known Activations