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

    Neuronpedia

    INDEX
    Explanations

    explain or describe

    np_acts-logits-general · gemini-2.5-flash-lite

    requests in user prompts that ask for explanations or detailed descriptions, signaling an instructional or educational response is needed.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     esterno
    0.27
     சார்ந்த
    0.27
     scambio
    0.27
     নই
    0.26
     prefs
    0.26
    bson
    0.26
    вшейся
    0.26
    EditText
    0.26
     esterni
    0.26
     randomized
    0.26
    POSITIVE LOGITS
     the
    0.43
    What
    0.38
     why
    0.36
    Why
    0.36
    the
    0.35
    what
    0.35
    why
    0.34
    The
    0.34
     what
    0.34
    How
    0.32
    Act Density 0.481%

    No Known Activations