© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 5649
    Prev
    Next
    INDEX
    Explanations

    error handling

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

    The neuron fires strongly on opening punctuation characters (especially “(”, “:”, or “[”) that introduce parameter lists or code-like delimiters.

    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
     var
    0.68
    `);
    0.60
    `,
    0.59
     handled
    0.58
     transformed
    0.57
    );
    0.56
     &(
    0.55
    []');
    0.54
    `,`
    0.54
     latter
    0.53
    POSITIVE LOGITS
    FBSDKLogin
    0.59
    ILayout
    0.59
    ђе
    0.58
    جتما
    0.57
    ினு
    0.57
    INTERNATIONAL
    0.57
     cumpl
    0.57
    予防
    0.56
    ்ப்ப
    0.55
    黑暗
    0.55
    Activations Density 0.000%

    No Known Activations