© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130028
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on tokens containing numeric values or digits (e.g., number literals or identifiers with embedded numbers).

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

    technical path components

    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
     kalend
    -1.21
     reduziert
    -1.21
     kré
    -1.20
    spania
    -1.19
     kabát
    -1.15
     cultuur
    -1.15
     genieten
    -1.14
     univerz
    -1.14
     eenige
    -1.12
     forskning
    -1.12
    POSITIVE LOGITS
     have
    1.14
    ยว
    1.13
    have
    1.05
    like
    1.04
    co
    1.04
    ];
    1.03
    }),
    1.03
     are
    1.02
     fiabilidad
    1.01
    begin
    1.00
    Activations Density 0.031%

    No Known Activations