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

    Neuronpedia

    INDEX
    Explanations

    technical steps or instructions identified by the word "Step" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    the presence of a specific formatting or structural elements in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eatures
    -0.79
    BIP
    -0.74
    ãĥīãĥ©ãĤ´ãĥ³
    -0.74
    selage
    -0.70
    ãĥ©ãĥ³
    -0.68
     Unic
    -0.68
    EStreamFrame
    -0.67
    ecause
    -0.67
     Pengu
    -0.64
     concess
    -0.63
    POSITIVE LOGITS
    hens
    1.02
    daughter
    0.96
    isters
    0.92
    Step
    0.86
    hen
    0.85
    antry
    0.84
    mother
    0.81
    han
    0.80
    dad
    0.79
    brother
    0.78
    Act Density 0.037%

    No Known Activations