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

    Neuronpedia

    INDEX
    Explanations

    Code snippets

    np_max-act · gemini-2.0-flash

    The neuron fires on recommendation cues—verbs like “look,” “check,” or “see” when pointing the reader to an external resource or example.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ogenesis
    -0.08
    CustomerId
    -0.07
    pres
    -0.07
     derail
    -0.07
    .Java
    -0.07
     athletics
    -0.07
    Danger
    -0.06
     exceeding
    -0.06
     Saves
    -0.06
    AFE
    -0.06
    POSITIVE LOGITS
    [pos
    0.07
    บอล
    0.07
    -num
    0.06
    vendors
    0.06
    �
    0.06
    elage
    0.06
     материал
    0.06
    .;↵
    0.06
     Towards
    0.06
    suspend
    0.06
    Act Density 0.034%

    No Known Activations