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

    Neuronpedia

    INDEX
    Explanations

    blaze and variations or trail concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.40
    тения
    0.40
    δέ
    0.38
    嬛
    0.38
    कारी
    0.38
    spiration
    0.37
    шої
    0.37
     Nato
    0.36
    ⇤
    0.36
    aceous
    0.36
    POSITIVE LOGITS
    trails
    0.60
     blaze
    0.60
     glory
    0.55
     trails
    0.55
    Trails
    0.55
     Glory
    0.52
     trail
    0.49
     Blaze
    0.49
     blazing
    0.49
     saddles
    0.48
    Act Density 0.002%

    No Known Activations