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

    Neuronpedia

    INDEX
    Explanations

    phrases related to fear or bravery

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "afraid."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    byter
    -0.82
    lished
    -0.82
    dates
    -0.77
    entials
    -0.76
    area
    -0.76
    issance
    -0.75
    iscopal
    -0.74
    bard
    -0.74
    rovers
    -0.73
    availability
    -0.73
    POSITIVE LOGITS
     afraid
    0.97
     lest
    0.84
    crow
    0.79
    ptin
    0.74
    NESS
    0.68
    oti
    0.68
     pregn
    0.68
     laughter
    0.67
     retribution
    0.66
     Polly
    0.66
    Act Density 0.023%

    No Known Activations