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

    Neuronpedia

    INDEX
    Explanations

    all

    np_max-act · gemini-2.0-flash

    The neuron activates primarily on number tokens (especially decimal numeric values).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    instructions for obtaining various file paths in Python.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     fans
    -0.08
     남자
    -0.07
     :"
    -0.06
    .unsqueeze
    -0.06
    PARSE
    -0.06
    ande
    -0.06
    -sk
    -0.06
     Swan
    -0.06
    ARE
    -0.06
     oceans
    -0.06
    POSITIVE LOGITS
    lda
    0.07
    lotte
    0.07
     인기글
    0.06
     initialState
    0.06
    0.06
     Beverly
    0.06
     newItem
    0.06
     Transform
    0.06
    commit
    0.06
    unan
    0.06
    Act Density 0.033%

    No Known Activations