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

    Neuronpedia

    INDEX
    Explanations

    sentences with strong emotional responses

    oai_token-act-pair · gpt-3.5-turbo

    sentences that express opinions or reactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thal
    -0.86
     hay
    -0.75
     glim
    -0.73
     prey
    -0.72
     transition
    -0.72
    hemer
    -0.71
     transitions
    -0.71
     pit
    -0.71
     disemb
    -0.71
     waterfall
    -0.70
    POSITIVE LOGITS
     Though
    1.15
     Instead
    1.10
     His
    1.08
     Their
    1.06
     Ironically
    1.06
     Perhaps
    1.05
     Already
    1.05
     Particularly
    1.04
     Critics
    1.03
     They
    1.01
    Act Density 0.810%

    No Known Activations