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

    Neuronpedia

    INDEX
    Explanations

    what does X look/sound/smell like

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Lok
    0.41
     Domen
    0.40
     Dorf
    0.39
     Sä
    0.38
    謐
    0.38
    bindParam
    0.38
    Lok
    0.38
     Happened
    0.38
    دهای
    0.37
     Domenico
    0.37
    POSITIVE LOGITS
     entails
    0.97
     entail
    0.92
     entailed
    0.75
     means
    0.61
     consist
    0.60
     innebär
    0.58
     consists
    0.57
     consisted
    0.55
     looks
    0.53
     состоит
    0.50
    Act Density 0.030%

    No Known Activations