© 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. 24-GEMMASCOPE-2-RES-16K
    7. 4145
    Prev
    Next
    INDEX
    Explanations

    python code identifiers

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

    This neuron is primarily detecting numeric color‐index codes (e.g. “CI 77700,” pigment identifiers in parentheses).

    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_24_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
     allele
    0.72
     summarizing
    0.68
    <tr>
    0.65
     đơn
    0.65
     mái
    0.65
    ெய்
    0.64
     มัน
    0.64
    IList
    0.64
     summarizes
    0.62
    ↵↵↵
    0.61
    POSITIVE LOGITS
    eteries
    0.72
    panies
    0.71
    が多く
    0.70
    жеб
    0.70
    detached
    0.69
    Quel
    0.68
     Quel
    0.66
     cristall
    0.66
    垷
    0.64
     comple
    0.64
    Activations Density 0.002%

    No Known Activations