© 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. 8362
    Prev
    Next
    INDEX
    Explanations

    class definitions

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

    The neuron is picking up standalone numeric literals 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_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
     nosso
    -1.96
     our
    -1.91
     nuestro
    -1.87
     znan
    -1.86
    ↵
    -1.84
     vedno
    -1.73
    ๖
    -1.71
     druge
    -1.70
     sebelum
    -1.68
     miasto
    -1.67
    POSITIVE LOGITS
    with
    1.89
    ting
    1.85
    K
    1.82
    they
    1.77
    J
    1.75
    ttt
    1.66
    X
    1.63
    ons
    1.62
     huden
    1.61
     –
    1.60
    Activations Density 0.011%

    No Known Activations