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

    Neuronpedia

    INDEX
    Explanations

    code and math

    np_max-act · gemini-2.0-flash

    The neuron responds specifically to numeric tokens—especially the “event_1,” “event_2,” and decimal‐formatted numbers used in the causal‐narrative annotations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     llama
    -0.08
    ราะ
    -0.07
    models
    -0.07
    ckeditor
    -0.07
     evidenced
    -0.06
    しま
    -0.06
     Yer
    -0.06
    ool
    -0.06
     Το
    -0.06
     treadmill
    -0.06
    POSITIVE LOGITS
    Posting
    0.07
    ifs
    0.07
    inerary
    0.06
    astle
    0.06
    ajaran
    0.06
    ASK
    0.06
    _Table
    0.06
    Comb
    0.06
    ict
    0.06
    Frequency
    0.06
    Act Density 0.006%

    No Known Activations