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

    Neuronpedia

    INDEX
    Explanations

    mentions of being together or doing things collectively

    oai_token-act-pair · gpt-3.5-turbo

    instances or references to events and experiences in a narrative format

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     affili
    -0.87
     advoc
    -0.85
     challengers
    -0.79
     challeng
    -0.79
     defic
    -0.78
     compet
    -0.77
    yip
    -0.75
     judiciary
    -0.74
     commend
    -0.74
     unden
    -0.74
    POSITIVE LOGITS
     Afterwards
    1.53
     Eventually
    1.46
     Then
    1.43
     Later
    1.40
     Anyway
    1.36
     Shortly
    1.34
     Needless
    1.31
     Initially
    1.30
     Suddenly
    1.30
     During
    1.29
    Act Density 0.286%

    No Known Activations