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

    Neuronpedia

    INDEX
    Explanations

    key-value lookups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Aran
    0.80
     seafood
    0.77
     SAIL
    0.75
     Lent
    0.73
    mares
    0.72
     Sailor
    0.72
     Sail
    0.70
    Sail
    0.70
     retiree
    0.69
    PCB
    0.68
    POSITIVE LOGITS
     keys
    1.89
     key
    1.68
    keys
    1.67
    Keys
    1.66
     Keys
    1.65
     Key
    1.60
    key
    1.58
    Key
    1.57
     Schlüssel
    1.55
    KEY
    1.50
    Act Density 0.537%

    No Known Activations