© 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. 10-GEMMASCOPE-RES-131K
    7. 130870
    Prev
    Next
    INDEX
    Explanations

    Margaret Hoover, Spice Routes

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

    The neuron activates on numeric literals—especially floating‐point numbers—embedded in the XML path data.

    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
    before
    -1.74
     ($('#
    -1.58
    data
    -1.57
    猡
    -1.48
     ($("#
    -1.47
     baan
    -1.45
     appartement
    -1.44
    ir
    -1.42
    when
    -1.39
     februari
    -1.37
    POSITIVE LOGITS
     was
    2.28
     will
    1.76
    必须要
    1.72
     thậm
    1.67
     orchestrated
    1.66
     ameste
    1.63
    ↵
    1.63
     obicei
    1.59
     cauza
    1.59
    nze
    1.56
    Activations Density 0.000%

    No Known Activations