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

    Neuronpedia

    INDEX
    Explanations

    eat

    np_max-act · gemini-2.0-flash

    The neuron flags words and word‐pieces that describe eating or being eaten (e.g. eat, eaten, swallow, devour, munch, digest).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    الش
    -0.07
    โล
    -0.06
    шее
    -0.06
    .root
    -0.06
    dění
    -0.06
    олько
    -0.06
     glean
    -0.06
    扬
    -0.06
    �
    -0.06
    anko
    -0.06
    POSITIVE LOGITS
     modifiers
    0.07
     Holmes
    0.07
     Qatar
    0.07
    /disc
    0.07
    .disc
    0.07
    (rt
    0.06
     internal
    0.06
    ={'
    0.06
    =sc
    0.06
     Mage
    0.06
    Act Density 0.014%

    No Known Activations