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

    Neuronpedia

    INDEX
    Explanations

    random generation

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Uma
    0.77
    Side
    0.77
     Uma
    0.74
     Side
    0.74
    Deep
    0.73
     сторону
    0.70
     Alias
    0.68
    Sodium
    0.67
     loose
    0.66
     उसको
    0.66
    POSITIVE LOGITS
    kD
    0.68
     THIRD
    0.63
    riffe
    0.62
     الثالثه
    0.59
    uba
    0.59
     embry
    0.58
     lapar
    0.58
     footer
    0.58
     counsell
    0.57
    rb
    0.56
    Act Density 0.001%

    No Known Activations