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

    Neuronpedia

    INDEX
    Explanations

    file paths

    oai_token-act-pair · gpt-3.5-turbo

    specific file paths or directory structures in a technical context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hungry
    -0.67
    jriwal
    -0.67
    usercontent
    -0.63
     Suz
    -0.63
     aloud
    -0.63
     reckoning
    -0.63
     sore
    -0.62
    Initialized
    -0.62
     uncomfortable
    -0.61
     thirsty
    -0.59
    POSITIVE LOGITS
    athan
    0.87
    illa
    0.79
    cade
    0.75
    Ľ
    0.74
    icum
    0.72
    uel
    0.71
    agne
    0.71
    Ļ
    0.71
    ela
    0.70
    idge
    0.70
    Act Density 0.286%

    No Known Activations