© 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. 113827
    Prev
    Next
    INDEX
    Explanations

    conceptualizations of

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

    The neuron is specialized for spotting abstract noun forms (nominalizations) such as words ending in ‑ion, ‑ation, ‑ing (when used as a noun), ‑al, and similar derivational suffixes.

    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
     Februari
    -2.13
    \
    -1.80
    mapStateToProps
    -1.73
    liters
    -1.71
     таки
    -1.68
     ткань
    -1.66
    🎄
    -1.66
     Januari
    -1.60
    pters
    -1.60
     In
    -1.60
    POSITIVE LOGITS
     the
    2.09
     instituições
    1.88
    以上です
    1.69
     then
    1.68
     hábitat
    1.66
     Verkä
    1.60
    注意事
    1.56
    灕
    1.55
    Spes
    1.55
     meestal
    1.54
    Activations Density 0.072%

    No Known Activations