© 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. 34-GEMMASCOPE-RES-131K
    7. 11911
    Prev
    Next
    INDEX
    Explanations

    Mathematical variables and function arguments

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

    This neuron activates on mathematical identifiers and variable names (like AC, AD, x, f(x), G, H) in formula or equation contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     chose
    -1.06
     excellent
    -1.05
     several
    -1.00
     überprü
    -0.96
     examining
    -0.95
     selecting
    -0.94
     that
    -0.94
     ensuring
    -0.93
     which
    -0.92
     substantial
    -0.92
    POSITIVE LOGITS
    能够
    1.05
     faciliter
    1.03
     cinq
    1.00
    ：……
    0.98
     ห
    0.98
     enten
    0.98
     itali
    0.94
     interessante
    0.93
     ????
    0.93
     ron
    0.93
    Activations Density 0.009%

    No Known Activations