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

    Neuronpedia

    INDEX
    Explanations

    titles and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ùng
    0.45
     disposto
    0.42
    świę
    0.41
    tier
    0.41
    invocation
    0.41
    ない
    0.40
     perish
    0.40
    missive
    0.40
     εγκα
    0.40
    combination
    0.40
    POSITIVE LOGITS
     GCP
    0.46
    还能
    0.44
     وتح
    0.44
     وليس
    0.43
     laug
    0.43
     BCE
    0.42
     BCS
    0.42
     Stoke
    0.42
    สามารถ
    0.41
     BF
    0.41
    Act Density 0.001%

    No Known Activations