© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 86871
    Prev
    Next
    INDEX
    Explanations

    mathematical theorems and lemmas

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

    The neuron detects occurrences of formal-mathematical statement labels—words like “definition,” “theorem,” or “proof.”

    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
     одном
    -1.89
    齜
    -1.88
    
    -1.84
    ஂ
    -1.79
    Демографія
    -1.72
     Это
    -1.71
    女
    -1.71
    schirm
    -1.66
     ")
    -1.57
    sturm
    -1.51
    POSITIVE LOGITS
     that
    2.52
     of
    1.92
     varandra
    1.86
    E
    1.71
     universitario
    1.59
     frische
    1.59
     freundlichen
    1.57
    女人的
    1.56
     erfolgreiche
    1.55
     entsprechende
    1.54
    Activations Density 0.011%

    No Known Activations