© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64550
    Prev
    Next
    INDEX
    Explanations

    mathematical, code, or url components

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

    The neuron activates on numeric literal tokens (numbers) appearing in equations, code, or technical text.

    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
     but
    -1.09
    ᾽
    -1.06
     gren
    -1.03
     SEMESTER
    -1.02
    TestingModule
    -1.02
     murs
    -1.01
     Mourinho
    -1.01
     UFC
    -1.01
    ℮
    -0.98
    ⅖
    -0.98
    POSITIVE LOGITS
     each
    1.48
     where
    1.21
    Pratique
    1.16
    Conseils
    1.09
     here
    1.07
    Suivez
    1.07
    arrêt
    1.05
    Skład
    1.05
     łatwo
    1.05
     participado
    1.04
    Activations Density 0.018%

    No Known Activations