© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10884
    Prev
    Next
    INDEX
    Explanations

    combinations and choices

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron fires on numeric tokens—that is, sequences of digits (numbers, counts, or quantities) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    large numbers written with commas as thousand separators.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    explicit numerical expressions—especially large counts, fractions/probabilities, and combinatorial or statistical quantities.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     BATH
    0.58
    bath
    0.54
    篆
    0.54
     Bath
    0.53
    ROR
    0.52
    Новости
    0.51
     Dre
    0.50
    TabStop
    0.50
    UGG
    0.50
     KLE
    0.50
    POSITIVE LOGITS
    binom
    1.13
     binomial
    1.02
    binomial
    0.97
     choose
    0.94
    choose
    0.83
    Choose
    0.80
     Choose
    0.79
     chooses
    0.79
     bin
    0.78
    Bin
    0.77
    Activations Density 0.009%

    No Known Activations