© 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. 44812
    Prev
    Next
    INDEX
    Explanations

    HTTP method annotations

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

    The neuron predominantly fires on numeric literals in code (e.g. numbers appearing in assignments or annotations).

    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
     лишь
    -1.56
    年轻人
    -1.45
     poliuret
    -1.45
     Tomé
    -1.45
    </h4>
    -1.44
    挹
    -1.43
    焼く
    -1.42
     cáps
    -1.41
     flèche
    -1.41
    
    -1.38
    POSITIVE LOGITS
     \"
    1.74
     '
    1.70
    følge
    1.68
     and
    1.64
     vielen
    1.63
     such
    1.56
    garmin
    1.53
     dabar
    1.52
    Revenir
    1.50
     jika
    1.46
    Activations Density 0.002%

    No Known Activations