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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6100
    Prev
    Next
    INDEX
    Explanations

    number followed by closing parenthesis

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

    This neuron activates on numeric tokens (numbers, codes, measurements, and other digit‐based values).

    oai_token-act-pair · o4-miniTriggered 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
    '});
    0.76
    "});
    0.73
    ”;
    0.70
    ’;
    0.70
    "};
    0.67
    '};
    0.67
    °.
    0.65
    »;
    0.65
    '];
    0.65
    }};
    0.65
    POSITIVE LOGITS
    )(
    1.53
    )(-
    1.40
     )(
    1.38
    )-
    1.34
    )(\
    1.32
    )_
    1.32
    )-(
    1.28
    )[
    1.24
    )([
    1.24
    )((
    1.21
    Activations Density 0.116%

    No Known Activations