© 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. 14359
    Prev
    Next
    INDEX
    Explanations

    code, document headers, and specific phrases

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

    The neuron strongly activates on long uninterrupted alphanumeric strings (e.g. base64- or hash-style sequences).

    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
    綏
    -2.14
     you
    -2.06
    淩
    -1.91
    zembro
    -1.88
     棟
    -1.87
    憮
    -1.86
    وحة
    -1.85
    ือก
    -1.84
     asas
    -1.84
     helg
    -1.84
    POSITIVE LOGITS
    i
    2.27
     of
    2.06
    this
    1.98
    ي
    1.95
     juntamente
    1.95
     podendo
    1.79
    甭
    1.73
     Vielzahl
    1.67
     évaluation
    1.65
    癀
    1.65
    Activations Density 0.006%

    No Known Activations