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

    Neuronpedia

    INDEX
    Explanations

    European

    np_max-act · gemini-2.0-flash

    language related to trust and communication in relationships.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron fires on the main topic words or keywords (usually nouns) that introduce what a piece of text is about.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Kle
    -0.07
    These
    -0.06
    海
    -0.06
    _lm
    -0.06
     replied
    -0.06
     certainty
    -0.06
     Pat
    -0.06
     Hernandez
    -0.06
     What
    -0.06
     obstruction
    -0.05
    POSITIVE LOGITS
    <message
    0.07
     ms
    0.07
     м
    0.06
    union
    0.06
     волос
    0.06
    _qs
    0.06
    oseconds
    0.06
    abant
    0.06
     ativ
    0.06
    (od
    0.06
    Act Density 0.224%

    No Known Activations