© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 9903
    Prev
    Next
    INDEX
    Explanations

    code syntax

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

    This neuron fires on standalone title‐ or heading‐style lines (e.g. the subject line of a post or the heading of a section).

    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
    ¬
    0.73
     ­
    0.73
    ”、“
    0.70
    )&
    0.70
     ','
    0.66
     】,
    0.65
     ¨
    0.64
    ‘’
    0.64
    )-,
    0.62
    ",'
    0.62
    POSITIVE LOGITS
    </blockquote>
    3.78
    </code>
    2.67
    </h5>
    2.36
    ");*/
    2.27
    </td>
    2.11
    );*/
    1.94
    </h1>
    1.84
    </i>
    1.84
    </h3>
    1.82
    </h2>
    1.77
    Activations Density 0.435%

    No Known Activations