© 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. 14178
    Prev
    Next
    INDEX
    Explanations

    graphical elements or data points

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

    The neuron specializes in spotting countable technical entities (like “voxels,” “nodes,” “errors,” “rows,” “dots,” etc.) typically associated with numeric quantities.

    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
    Their
    -1.38
     dina
    -1.38
     merak
    -1.37
    nalité
    -1.37
    verwaltung
    -1.34
     fabrik
    -1.34
     mavi
    -1.33
     geene
    -1.33
     when
    -1.33
    0
    -1.30
    POSITIVE LOGITS
     may
    1.55
     ajusta
    1.54
     you
    1.49
     contará
    1.46
     the
    1.46
     Parece
    1.40
    翆
    1.38
     affiliated
    1.33
    audiovisuel
    1.33
     continuará
    1.32
    Activations Density 0.263%

    No Known Activations