© 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. 110902
    Prev
    Next
    INDEX
    Explanations

    tactical radio, members with, rational drug

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

    The neuron is specialized to detect numeric literals (i.e. tokens composed of digits) in code.

    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
     Stella
    -0.75
     Camus
    -0.74
    altham
    -0.73
     feasts
    -0.73
    ähteet
    -0.72
     Simple
    -0.72
    ง่าย
    -0.72
     Koss
    -0.70
    ོད་
    -0.70
     infertility
    -0.70
    POSITIVE LOGITS
    ໒
    0.80
    ēju
    0.79
     virš
    0.78
    ECM
    0.75
    MCM
    0.75
    spra
    0.74
     twe
    0.74
    cended
    0.74
     aukš
    0.74
    kli
    0.73
    Activations Density 0.010%

    No Known Activations