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

    Neuronpedia

    INDEX
    Explanations

    phrases related to statements made by specific individuals

    oai_token-act-pair · gpt-3.5-turbo

    colons followed by statements or claims

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tremend
    -0.90
    inement
    -0.73
     behavi
    -0.73
    ilater
    -0.72
    undai
    -0.72
     pursu
    -0.72
     behav
    -0.70
     peac
    -0.68
    inarily
    -0.67
     recip
    -0.67
    POSITIVE LOGITS
     Logged
    0.93
     TBD
    0.87
     Provided
    0.84
     Cosponsors
    0.80
     Yeah
    0.77
     Who
    0.75
    ËĪ
    0.75
     Rise
    0.74
     Prelude
    0.73
     Played
    0.71
    Act Density 0.122%

    No Known Activations