© 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. 7469
    Prev
    Next
    INDEX
    Explanations

    third or quarter proportions

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

    The neuron spots numeric expressions of proportions (fractions, ratios, and percentages).

    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
    兩
    -1.52
     a
    -1.48
     آثار
    -1.45
     也
    -1.45
    maybe
    -1.43
    亖
    -1.41
     faktisk
    -1.38
    趼
    -1.34
     macetas
    -1.34
    WILL
    -1.34
    POSITIVE LOGITS
     of
    1.90
    летний
    1.55
    quelize
    1.48
     new
    1.45
     in
    1.41
    СЫ
    1.33
    kipun
    1.33
     цвят
    1.30
    👬
    1.30
     succulent
    1.27
    Activations Density 0.023%

    No Known Activations