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

    Neuronpedia

    INDEX
    Explanations

    endings like "-ify" and "-ura"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    t
    0.49
    ts
    0.46
    tt
    0.44
    ttle
    0.41
    k
    0.40
    v
    0.39
    db
    0.39
    ten
    0.38
    ja
    0.38
    ti
    0.38
    POSITIVE LOGITS
    :
    0.66
     sweater
    0.41
    ：
    0.41
     ricotta
    0.40
    𝕒
    0.40
     sunset
    0.39
    سمبر
    0.39
     sư
    0.39
     suture
    0.38
     &:
    0.38
    Act Density 0.254%

    No Known Activations