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

    Neuronpedia

    INDEX
    Explanations

    brief occurrences or references in the text

    oai_token-act-pair · gpt-3.5-turbo

    repeated elements or patterns within descriptions of creatures or objects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ALD
    -0.70
    ceans
    -0.70
     Polo
    -0.67
    rogens
    -0.67
    cano
    -0.65
    uben
    -0.61
    Queen
    -0.59
    ILLE
    -0.58
    onz
    -0.58
    iano
    -0.58
    POSITIVE LOGITS
    case
    1.50
    cases
    1.43
    er
    0.90
     glimps
    0.87
    est
    0.83
    ings
    0.80
    itud
    0.79
     foray
    0.79
     stint
    0.79
     resp
    0.77
    Act Density 0.047%

    No Known Activations