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

    Neuronpedia

    INDEX
    Explanations

    adjectives related to depth or completeness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "thorough" and its context related to investigations or evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    artifacts
    -0.66
     Rebels
    -0.66
     Tonight
    -0.65
    oras
    -0.64
    orea
    -0.63
     DRAG
    -0.63
    atu
    -0.61
    ettle
    -0.61
    dust
    -0.61
     indirectly
    -0.61
    POSITIVE LOGITS
    bred
    1.23
     thorough
    1.10
    fare
    0.98
    ness
    0.93
     completion
    0.77
    orough
    0.75
     examinations
    0.75
     examination
    0.74
    rehensive
    0.72
    nesses
    0.72
    Act Density 0.005%

    No Known Activations