© 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. 16066
    Prev
    Next
    INDEX
    Explanations

    laws, rules, and changes

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

    The neuron strongly activates on numeric tokens—years, measurements, figures, and other standalone numbers.

    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
     trám
    -1.04
    LISTS
    -0.97
    醢
    -0.97
    ູ
    -0.96
    fda
    -0.95
    BUILDING
    -0.94
     linguistique
    -0.93
    を用意
    -0.92
    ınızı
    -0.92
    really
    -0.91
    POSITIVE LOGITS
     current
    1.38
     currently
    1.37
     changes
    1.33
     reform
    1.27
     change
    1.17
     proposal
    1.16
    Currently
    1.16
     Currently
    1.14
     actuelle
    1.13
     review
    1.11
    Activations Density 0.054%

    No Known Activations