© 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. 129686
    Prev
    Next
    INDEX
    Explanations

    The neuron detects numeric expressions and list-introducing punctuation (numbers, decimals, and colons).

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

    talking about winning or purpose

    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
    ILLO
    -0.94
     placa
    -0.93
     terape
    -0.92
    volent
    -0.92
     externa
    -0.92
    聖誕
    -0.89
     pantalon
    -0.87
     diamant
    -0.86
     interna
    -0.85
     epoca
    -0.84
    POSITIVE LOGITS
    riguez
    0.92
     his
    0.90
    geer
    0.86
     muestras
    0.84
     Spitze
    0.83
     erstmal
    0.83
    novi
    0.83
    AMILTON
    0.82
     apasion
    0.81
    按照
    0.80
    Activations Density 0.036%

    No Known Activations