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

    Neuronpedia

    INDEX
    Explanations

    in the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Spe
    0.60
    h
    0.59
    Factor
    0.58
    Half
    0.57
    Winner
    0.57
    Buy
    0.55
    Ro
    0.54
    Camp
    0.53
    5
    0.53
    Forget
    0.53
    POSITIVE LOGITS
     кораб
    0.70
     ignited
    0.64
     музы
    0.63
    oughed
    0.63
    uyeron
    0.59
     enclosed
    0.59
     चाहे
    0.58
     जीवित
    0.57
     брау
    0.57
     yath
    0.57
    Act Density 0.027%

    No Known Activations