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

    Neuronpedia

    INDEX
    Explanations

    prefixes

    np_max-act · gemini-2.0-flash

    phrases or words related to emotional experiences and relationships.

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

    The neuron spotlights hyphenated “post-” prefixed terms (i.e. words beginning with “post-”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ()<<"
    -0.07
     khi
    -0.07
    них
    -0.06
     처음
    -0.06
    pcs
    -0.06
     Các
    -0.06
     discoveries
    -0.06
     crafts
    -0.06
     vivid
    -0.06
     Suriye
    -0.06
    POSITIVE LOGITS
     Baylor
    0.07
    stakes
    0.06
    Endpoint
    0.06
    IDA
    0.06
     dinosaurs
    0.06
    čku
    0.06
    .FromResult
    0.06
    Truthy
    0.06
     odds
    0.06
     NUIT
    0.06
    Act Density 0.033%

    No Known Activations