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

    Neuronpedia

    INDEX
    Explanations

    lines of code or pieces of technical documentation

    oai_token-act-pair · gpt-3.5-turbo

    statement ends and punctuation in technical writing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    antz
    -0.80
    ŃĶ
    -0.70
    Ń·
    -0.68
    nell
    -0.61
     entrenched
    -0.61
    ãģ®éŃĶ
    -0.61
    mitt
    -0.61
     waged
    -0.61
    ansky
    -0.60
    ierrez
    -0.59
    POSITIVE LOGITS
     Otherwise
    1.43
     Therefore
    1.39
     Alternatively
    1.35
     Thus
    1.33
     Example
    1.32
     Hence
    1.31
     Parameters
    1.30
     Suppose
    1.29
     Useful
    1.29
     Additionally
    1.26
    Act Density 0.125%

    No Known Activations