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

    Neuronpedia

    INDEX
    Explanations

    verbs in past or present perfect tenses

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating conditionality and change

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Atmosp
    -0.80
     Skies
    -0.72
     Shal
    -0.68
     Sly
    -0.67
     Aer
    -0.63
     Lun
    -0.63
     Basics
    -0.62
     Hels
    -0.60
     Ros
    -0.60
     basics
    -0.59
    POSITIVE LOGITS
     somehow
    1.10
     outweigh
    0.92
     outwe
    0.79
    etheless
    0.78
    uddenly
    0.72
     necessarily
    0.70
     overshadow
    0.70
    ||||
    0.69
     suddenly
    0.69
     disproportionately
    0.69
    Act Density 0.800%

    No Known Activations