© 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. 110837
    Prev
    Next
    INDEX
    Explanations

    Tech, Q, --- followed by separators

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

    opening parentheses in code or technical text.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on decimal numeric literals (floating‐point numbers) in code.

    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
    人には
    -1.56
    llium
    -1.50
    Ці
    -1.47
    
    -1.45
     geschikt
    -1.45
    っぷり
    -1.42
     crainte
    -1.41
     mengumumkan
    -1.41
     verke
    -1.40
     reseña
    -1.40
    POSITIVE LOGITS
     nicht
    1.77
     we
    1.66
     new
    1.57
     Belakang
    1.50
    AGE
    1.47
    ﺯ
    1.44
     nowy
    1.44
    ୩
    1.41
     ikke
    1.41
    ୮
    1.40
    Activations Density 0.000%

    No Known Activations