© 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. 67487
    Prev
    Next
    INDEX
    Explanations

    <code>codeschemaundefined

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

    The neuron fires on numeric literal tokens in code (i.e. integer or floating‐point constants).

    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
    ständen
    -1.57
     When
    -1.56
    cting
    -1.55
    ]."
    -1.54
    躂
    -1.41
     bordados
    -1.38
    Ilustra
    -1.36
     汽车
    -1.34
     bagaimana
    -1.33
     动物
    -1.33
    POSITIVE LOGITS
    <blockquote>
    1.57
    '
    1.43
    ve
    1.41
    mathrm
    1.38
     tandis
    1.36
    iſter
    1.35
     naszej
    1.34
    ى
    1.34
     langer
    1.31
     following
    1.30
    Activations Density 0.016%

    No Known Activations