© 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. 5911
    Prev
    Next
    INDEX
    Explanations

    The neuron is essentially detecting numeric tokens and measurements—dates, percentages, decimal‐valued figures, and other number‐based data.

    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
    <strong>
    1.33
    <b>
    1.05
    \\
    0.97
    \...
    0.95
    ?\\
    0.95
    .\\
    0.92
    -\\
    0.91
     […]
    0.88
    )\\
    0.86
    <h5>
    0.82
    POSITIVE LOGITS
    </b>
    3.38
    </strong>
    2.89
    </u>
    2.58
    </code>
    2.34
    </em>
    2.33
    </i>
    2.28
    </span>
    2.19
    </sup>
    2.13
    </s>
    1.95
    </sub>
    1.77
    Activations Density 1.163%

    No Known Activations