© 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. 10-GEMMASCOPE-RES-131K
    7. 84420
    Prev
    Next
    INDEX
    Explanations

    code syntax with parentheses and symbols

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

    The neuron detects numeric literal tokens (i.e. 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_10/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
     is
    -1.76
     the
    -1.72
    がっ
    -1.63
    should
    -1.63
     possono
    -1.62
     superb
    -1.59
     deft
    -1.55
    宜しくお願い
    -1.55
     elegante
    -1.53
     két
    -1.53
    POSITIVE LOGITS
     with
    2.13
     would
    2.03
     that
    1.95
     people
    1.82
     for
    1.79
     thinks
    1.73
     Knowing
    1.70
     With
    1.68
     other
    1.67
     heightened
    1.63
    Activations Density 0.001%

    No Known Activations