© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129441
    Prev
    Next
    INDEX
    Explanations

    boundary conditions, orders, geometry, solutions, technique

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

    This neuron fires on the appearance of a specialized mathematical or technical term when the article first names its main topic (e.g. "integral polytope," "Robin boundary condition," etc.).

    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
     gastronomía
    -1.88
     خریدار
    -1.84
    (!__
    -1.73
     dát
    -1.70
     iglesia
    -1.70
     وكانت
    -1.64
     caballero
    -1.64
    ‚¬
    -1.60
     circulación
    -1.60
     robić
    -1.59
    POSITIVE LOGITS
     просто
    2.00
     ANCHE
    1.59
     упаковка
    1.58
     veröffentlich
    1.56
     contratto
    1.54
     religieux
    1.53
     kadang
    1.52
     ersten
    1.51
     Национальный
    1.50
    adesivo
    1.49
    Activations Density 0.111%

    No Known Activations