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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating uncertainty or conditionality

    oai_token-act-pair · gpt-3.5-turbo

    phrases that convey conditional or uncertain relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fw
    -0.78
    iry
    -0.71
    banks
    -0.69
    bah
    -0.68
    iti
    -0.68
    ults
    -0.68
    bern
    -0.67
    DF
    -0.67
    oru
    -0.65
    met
    -0.65
    POSITIVE LOGITS
     necessarily
    0.81
     equate
    0.74
     anymore
    0.73
     correlate
    0.69
     correlated
    0.69
     portrayed
    0.68
     confined
    0.68
     indicative
    0.68
     advers
    0.68
     entail
    0.68
    Act Density 0.033%

    No Known Activations