© 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. 44213
    Prev
    Next
    INDEX
    Explanations

    structured configurations and names

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

    The neuron fires strongly on “Add” calls (and similar list‐item markers like leading “–”)—i.e. places where something is being added to a collection or enumerated.

    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
     The
    -1.94
    这些
    -1.91
     It
    -1.91
     Their
    -1.87
    ↵↵
    -1.77
     This
    -1.73
     These
    -1.70
    這些
    -1.66
     ursprünglichen
    -1.56
     damals
    -1.55
    POSITIVE LOGITS
     Belum
    1.68
     tica
    1.67
     mavi
    1.66
     pavillon
    1.63
     beni
    1.63
     Giochi
    1.63
     sanat
    1.60
     skog
    1.59
     ใส่
    1.55
    mäh
    1.55
    Activations Density 0.006%

    No Known Activations