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

    Neuronpedia

    INDEX
    Explanations

    questions that start with "How."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    how questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     endregion
    -0.55
     [*]
    -0.51
     presently
    -0.50
    })*/
    -0.48
    -------
    -0.47
     ]]
    -0.47
     transcriptional
    -0.46
    ebx
    -0.46
     transcrip
    -0.46
    endregion
    -0.46
    POSITIVE LOGITS
    How
    0.83
     How
    0.81
    how
    0.63
    HOW
    0.57
     Nasıl
    0.57
     how
    0.56
    Cómo
    0.56
    -¿
    0.55
    —¿
    0.54
     "¿
    0.54
    Act Density 0.016%

    No Known Activations