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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing emotions or opinions

    oai_token-act-pair · gpt-3.5-turbo

    details about evaluations, responses, and outcomes related to decisions and events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iliated
    -0.74
    igate
    -0.66
    ornia
    -0.63
    interrupted
    -0.63
    uments
    -0.62
    emies
    -0.59
    ificantly
    -0.59
    itory
    -0.58
    orno
    -0.56
    aneously
    -0.56
    POSITIVE LOGITS
     antics
    0.88
     sincerity
    0.73
     honesty
    0.69
    eness
    0.68
     inconsistency
    0.68
     temptation
    0.68
     arrang
    0.67
     motives
    0.67
     portrayal
    0.67
     pitfalls
    0.67
    Act Density 0.720%

    No Known Activations