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

    Neuronpedia

    INDEX
    Explanations

    phrases that are enclosed in quotation marks

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include quotations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     —
    -0.55
    arnaev
    -0.52
    bably
    -0.49
     --
    -0.47
     laborers
    -0.45
     mistaken
    -0.45
     (@
    -0.45
     firsthand
    -0.45
    rompt
    -0.45
     afterward
    -0.44
    POSITIVE LOGITS
    ",
    3.24
    !",
    2.66
    ?",
    2.62
    )",
    2.62
    .",
    2.56
    ".
    2.54
    ".[
    2.51
     ",
    2.48
    "),
    2.45
    "],
    2.34
    Act Density 0.018%

    No Known Activations