© 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. 33371
    Prev
    Next
    INDEX
    Explanations

    coordinates and dimensions

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

    The neuron fires on numeric literals (digits or numbers) in the 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
     that
    -2.08
     even
    -1.41
     what
    -1.23
    Even
    -1.19
     (
    -1.16
     in
    -1.15
     which
    -1.12
     having
    -1.11
     both
    -1.11
    While
    -1.10
    POSITIVE LOGITS
     canadian
    1.24
     Dinas
    1.22
     シルエット
    1.19
     to
    1.16
     magnific
    1.14
     dando
    1.13
     طبق
    1.11
     amd
    1.10
     facili
    1.09
    potentially
    1.08
    Activations Density 0.018%

    No Known Activations