© 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. 42153
    Prev
    Next
    INDEX
    Explanations

    physical/electrical quantities (T, I, V, H, ρ)

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

    The neuron detects isolated mathematical variable symbols (single-letter identifiers and Greek letters) in technical formulas.

    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
     to
    -2.22
     at
    -1.77
     in
    -1.77
     out
    -1.57
     that
    -1.55
     if
    -1.55
     ζωή
    -1.49
    t
    -1.48
    g
    -1.42
     even
    -1.40
    POSITIVE LOGITS
    saludos
    1.45
     quintessential
    1.41
    鋸
    1.36
    
    1.35
    depuis
    1.35
    الان
    1.33
     cóc
    1.33
     yako
    1.33
     noong
    1.31
    ventilador
    1.31
    Activations Density 0.057%

    No Known Activations