© 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. 128666
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is find numeric tokens (integers, decimals, or other number literals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    based value, flags, numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    momix
    -0.84
     แบบ
    -0.82
     Zutaten
    -0.80
     или
    -0.79
    komen
    -0.78
    gulan
    -0.78
     estiver
    -0.77
     solicitar
    -0.77
     Gradually
    -0.77
     specifiche
    -0.75
    POSITIVE LOGITS
     Encode
    0.89
     ordination
    0.81
    Encode
    0.80
     encode
    0.80
     fass
    0.79
    Normalized
    0.77
     normalized
    0.77
    左手
    0.77
     Oxidation
    0.75
     unsigned
    0.75
    Activations Density 0.050%

    No Known Activations