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

    Neuronpedia

    INDEX
    Explanations

    PATH, alignment, comments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Şu
    0.42
    7
    0.39
     Bele
    0.39
    Benzo
    0.39
     boton
    0.39
    4
    0.38
     noch
    0.37
     සී
    0.37
     ganzen
    0.36
     Bell
    0.36
    POSITIVE LOGITS
    nfasis
    0.45
     nasc
    0.45
    ității
    0.44
    িশীল
    0.43
    entingan
    0.43
    τητας
    0.43
    फ्ट
    0.42
    িবদ্ধ
    0.41
    νης
    0.40
    κής
    0.40
    Act Density 0.003%

    No Known Activations