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

    Neuronpedia

    INDEX
    Explanations

    adjectives and phrases indicating an observation or perception of certainty or likelihood

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "apparent" and its variations, indicating situations that are perceived as evident or obvious

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uden
    -1.00
    alez
    -0.76
    ajo
    -0.71
    rieg
    -0.71
    inion
    -0.69
    dden
    -0.66
    pped
    -0.65
     drilled
    -0.65
    berman
    -0.65
    zsche
    -0.65
    POSITIVE LOGITS
     contradictions
    0.97
    iary
    0.88
     contradiction
    0.86
    aneously
    0.80
     discrepancies
    0.79
    ances
    0.76
    volent
    0.74
     inconsistency
    0.73
     inconsistencies
    0.71
    aneous
    0.70
    Act Density 0.018%

    No Known Activations