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

    Neuronpedia

    INDEX
    Explanations

    headroom and deficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sunglasses
    1.58
    för
    1.52
     linewidth
    1.50
     моче
    1.49
    vskip
    1.47
     Randomized
    1.45
    다고
    1.44
     Spearman
    1.43
    ﻩ
    1.42
    smallskip
    1.41
    POSITIVE LOGITS
    м
    1.67
    воре
    1.67
     самым
    1.61
    ड
    1.58
    Nach
    1.47
    于是
    1.45
    ب
    1.42
    Acces
    1.38
    ం
    1.37
    HING
    1.37
    Act Density 0.001%

    No Known Activations