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

    Neuronpedia

    INDEX
    Explanations

    phrases using the phrase "not exactly" in the text

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating precision or specificity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    çļ
    -0.84
    idas
    -0.78
     unavoid
    -0.71
    Monitor
    -0.71
    bnb
    -0.67
    tailed
    -0.67
    GD
    -0.67
    aman
    -0.66
    adium
    -0.65
    asar
    -0.64
    POSITIVE LOGITS
     forgiving
    0.80
     surprising
    0.80
     anymore
    0.76
     pleasant
    0.75
     appe
    0.75
     conducive
    0.74
     flashy
    0.70
     angels
    0.70
     fool
    0.69
     altru
    0.69
    Act Density 0.055%

    No Known Activations