© 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. 127392
    Prev
    Next
    INDEX
    Explanations

    // or slashes

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

    The neuron is sensitive to the forward‐slash ("/") character when it’s used to separate two terms (as in paired headings or titles).

    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
     sorprender
    -1.74
     there
    -1.67
     mention
    -1.57
     promove
    -1.48
     finally
    -1.45
     time
    -1.45
     then
    -1.44
     you
    -1.44
     felpa
    -1.41
     the
    -1.39
    POSITIVE LOGITS
     večer
    1.53
    lysm
    1.51
    ⬮
    1.49
     cirugía
    1.47
     inflicting
    1.46
    cheibe
    1.43
    Сла
    1.43
     août
    1.41
    Illustrated
    1.41
     stammen
    1.41
    Activations Density 0.053%

    No Known Activations