© 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. 26181
    Prev
    Next
    INDEX
    Explanations

    source, updated, illustration, click

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

    The neuron primarily detects numeric tokens (e.g. years, stats, measurements, or other numbers and decimals).

    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
    billon
    -1.13
    aned
    -1.12
     régiment
    -1.03
     yeter
    -1.02
     hvordan
    -1.02
    ldquo
    -1.01
    bär
    -1.00
     INSTALL
    -1.00
    bof
    -0.99
    Quien
    -0.99
    POSITIVE LOGITS
     from
    1.37
     after
    1.27
     data
    1.25
    らしいです
    1.23
     using
    1.23
     created
    1.20
     with
    1.17
     partially
    1.16
     solely
    1.14
     only
    1.13
    Activations Density 0.002%

    No Known Activations