© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 90778
    Prev
    Next
    INDEX
    Explanations

    virtual machines

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

    The neuron is picking out code‐file boilerplate and documentation tokens—especially license names and versions, class/import identifiers, docstring keywords (like “param”), and numeric literals.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     colombiana
    -0.81
    LogUtil
    -0.77
     vezet
    -0.75
    𝑥
    -0.75
     diisi
    -0.71
    แหล
    -0.71
    тение
    -0.69
     mexicanos
    -0.69
    outen
    -0.69
    euvre
    -0.67
    POSITIVE LOGITS
     Proximity
    0.94
     VM
    0.93
     proximity
    0.91
    VM
    0.85
     boot
    0.85
     Vm
    0.85
     vm
    0.83
     ICH
    0.82
    Guest
    0.82
     Capricorn
    0.81
    Activations Density 0.014%

    No Known Activations