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

    Neuronpedia

    INDEX
    Explanations

    questions or phrases related to the concept of "how."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asan
    -0.15
    ewn
    -0.15
    uong
    -0.14
    aran
    -0.14
    afil
    -0.14
    isz
    -0.14
    isch
    -0.13
     resc
    -0.13
     somew
    -0.13
     exclude
    -0.13
    POSITIVE LOGITS
     much
    0.21
    Much
    0.16
    much
    0.16
     Much
    0.15
    oose
    0.15
    .ld
    0.15
     MUCH
    0.15
     amount
    0.14
     frei
    0.14
    ResourceManager
    0.14
    Act Density 0.047%

    No Known Activations