© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39591
    Prev
    Next
    INDEX
    Explanations

    code identifiers and method names

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

    This neuron activates on programming-related tokens—especially code identifiers (method calls like fetchall(), ReadAllLines(), getMyObjs()) and file-format extensions (e.g. pkl).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    鶿
    -2.78
    เกี่ยวกับ
    -2.55
    瑧
    -2.50
    謌
    -2.45
     なら
    -2.30
     of
    -2.30
     infantile
    -2.17
    沂
    -2.16
    <td>
    -2.11
    焌
    -2.11
    POSITIVE LOGITS
    舠
    2.91
    橚
    2.72
    赑
    2.45
    ﮦ
    2.42
    骝
    2.34
    “
    2.27
    嫪
    2.27
    s
    2.25
    being
    2.25
    眎
    2.23
    Activations Density 0.001%

    No Known Activations