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

    Neuronpedia

    INDEX
    Explanations

    scientific names suffixes

    np_acts-logits-general · gemini-2.5-flash-lite

    scientific genus names in taxonomic listings.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
     célè
    -1.92
     réjou
    -1.74
     pistolet
    -1.73
     plafon
    -1.66
     blusa
    -1.62
     dépens
    -1.62
     véhicule
    -1.59
     pérd
    -1.59
     zoude
    -1.55
    থ্য
    -1.52
    POSITIVE LOGITS
     in
    1.98
     only
    1.80
     to
    1.73
     all
    1.59
     with
    1.59
     on
    1.55
     at
    1.53
     also
    1.53
     now
    1.51
     from
    1.50
    Act Density 0.020%

    No Known Activations