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

    Neuronpedia

    INDEX
    Explanations

    statements expressing high confidence or likelihood

    oai_token-act-pair · gpt-3.5-turbo

    statements expressing certainty or likelihood

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    kind
    -0.83
    inth
    -0.82
    elle
    -0.82
    andan
    -0.78
    elight
    -0.77
    ollen
    -0.72
    aredevil
    -0.67
    uckland
    -0.67
    istry
    -0.66
    banks
    -0.65
    POSITIVE LOGITS
     guessed
    0.82
     mistaken
    0.77
     assume
    0.76
     underest
    0.73
     guesses
    0.70
     untrue
    0.68
     assumed
    0.68
     infer
    0.66
     aspir
    0.65
     evapor
    0.65
    Act Density 0.065%

    No Known Activations