© 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. 11767
    Prev
    Next
    INDEX
    Explanations

    moving to a destination

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

    The neuron fires on numeric tokens, especially decimal numbers (and often those in “X to Y” ranges).

    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
     скачать
    -1.70
    𓆏
    -1.67
     konkret
    -1.67
    𓍊
    -1.66
     salonu
    -1.60
     calcetines
    -1.59
     verp
    -1.58
     вашего
    -1.57
     delikat
    -1.55
     plik
    -1.54
    POSITIVE LOGITS
     to
    1.86
     by
    1.70
     as
    1.70
     about
    1.69
     in
    1.68
     during
    1.59
     despite
    1.56
     Even
    1.54
     What
    1.53
     for
    1.45
    Activations Density 0.018%

    No Known Activations