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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating differentiation or distinction between subjects or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Swanson
    -0.58
     Haber
    -0.52
     cas
    -0.51
    нансо
    -0.48
     occasionally
    -0.47
     Ban
    -0.45
     Hold
    -0.44
    WithFormat
    -0.44
     Sto
    -0.43
    qrt
    -0.42
    POSITIVE LOGITS
     distinguishes
    1.26
     differentiating
    1.24
     differentiates
    1.23
     distinguish
    1.21
     differentiate
    1.21
     distinguishing
    1.19
     distingue
    1.18
     distinguer
    1.15
    differ
    1.13
     diferen
    1.10
    Act Density 0.217%

    No Known Activations