© 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. 3082
    Prev
    Next
    INDEX
    Explanations

    **

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

    This neuron detects Markdown formatting markers (e.g., asterisks for bold/emphasis and hashes for headings).

    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>
    1.13
    <b>
    1.02
    </blockquote>
    0.99
     mathbf
    0.96
     Although
    0.94
     Když
    0.88
     branchNode
    0.86
    <blockquote>
    0.84
     minifyEnabled
    0.84
     Because
    0.83
    POSITIVE LOGITS
    </strong>
    1.37
    </b>
    1.15
    ש
    1.04
    Ва
    0.98
    UPDATE
    0.97
    "
    0.94
    Question
    0.93
    REQUEST
    0.93
    ش
    0.92
    BUT
    0.91
    Activations Density 0.021%

    No Known Activations