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

    Neuronpedia

    INDEX
    Explanations

    reports or studies that highlight statistics and findings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    reporting findings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     TestBed
    -0.43
     оригіналу
    -0.42
     заве
    -0.42
     lenker
    -0.41
     esternos
    -0.39
     Boron
    -0.35
     Walkover
    -0.35
    onTap
    -0.35
    сар
    -0.35
     Italijanski
    -0.34
    POSITIVE LOGITS
     zufolge
    0.65
     protoimpl
    0.59
     scorso
    0.53
    berdayakan
    0.51
    ofire
    0.50
    brigens
    0.50
     bahawa
    0.50
    ftagPool
    0.49
    awtextra
    0.48
    TokenNameCOMMA
    0.48
    Act Density 0.042%

    No Known Activations