© 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. 22005
    Prev
    Next
    INDEX
    Explanations

    Regional anesthesia

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

    code indentation levels, with higher activation values for deeper indentation in programming code.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates strongly on numeric literals (integers and floats) 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
     as
    -1.91
     was
    -1.74
     
    -1.69
     who
    -1.68
     '
    -1.61
     care
    -1.60
     from
    -1.58
     "
    -1.57
     to
    -1.53
     being
    -1.45
    POSITIVE LOGITS
    鷓
    1.83
    雠
    1.77
     grosso
    1.70
     firft
    1.69
     éx
    1.61
    さてさて
    1.59
     maig
    1.57
     culo
    1.55
     dinosau
    1.55
     maniere
    1.55
    Activations Density 0.001%

    No Known Activations