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

    Neuronpedia

    INDEX
    Explanations

    inherent truths, presents differently, tangled, data

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     conserva
    0.56
     consolid
    0.55
     locul
    0.53
     observant
    0.53
     regul
    0.52
     thé
    0.50
     thermodynam
    0.50
     observa
    0.49
     geom
    0.49
     fatto
    0.49
    POSITIVE LOGITS
    د
    0.49
    िड
    0.46
    م
    0.46
    טים
    0.45
    chsler
    0.45
    ع
    0.44
    Expire
    0.44
    ש
    0.44
    kb
    0.43
    чника
    0.43
    Act Density 0.006%

    No Known Activations