© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 83170
    Prev
    Next
    INDEX
    Explanations

    linux /var paths

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

    This neuron detects numeric tokens (sequences of digits, like version numbers, file sizes, or other numeric 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_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
    ichon
    -0.85
    oblotting
    -0.83
    રી
    -0.78
    risi
    -0.69
     spring
    -0.69
    DESK
    -0.67
    ăți
    -0.67
    ylvan
    -0.66
     tasas
    -0.65
     adı
    -0.65
    POSITIVE LOGITS
    var
    1.38
    logs
    1.20
     logs
    1.17
    data
    1.16
    Logs
    1.13
     data
    1.11
    tmp
    1.02
    RUN
    1.01
    runtime
    1.01
    VAR
    1.01
    Activations Density 0.067%

    No Known Activations