© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25905
    Prev
    Next
    INDEX
    Explanations

    mathematical variables and symbols

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

    The neuron fires on isolated single-letter variable symbols (like u, v, s) appearing in mathematical expressions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     use
    -1.04
     content
    -0.98
     SU
    -0.90
     A
    -0.88
     UN
    -0.88
     absoluto
    -0.82
     ®
    -0.82
     optimal
    -0.80
     reserve
    -0.79
     „
    -0.79
    POSITIVE LOGITS
    Olvid
    1.03
     insecto
    1.00
     ravis
    0.99
    тости
    0.97
     geniales
    0.94
    mulos
    0.93
     Olímpicos
    0.92
     reflexiones
    0.92
     姆
    0.91
     头像
    0.90
    Activations Density 0.032%

    No Known Activations