© 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. 65589
    Prev
    Next
    INDEX
    Explanations

    geometric bundles and maps

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

    The neuron is tuned to pick out mathematical notation tokens—variables, symbols, and TeX fragments within formula‐like passages.

    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
    Ƅ
    -1.13
     Besøkt
    -1.02
     anivers
    -0.98
    loren
    -0.97
     kanker
    -0.96
    isait
    -0.96
    Vector
    -0.95
     regione
    -0.94
     nobili
    -0.94
     natale
    -0.93
    POSITIVE LOGITS
     between
    1.20
     from
    1.17
     réaction
    1.09
     suami
    1.08
    forhold
    1.01
     inclusions
    1.00
    zeitige
    1.00
     stolet
    0.99
     inks
    0.99
     beantworten
    0.97
    Activations Density 0.016%

    No Known Activations