© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21818
    Prev
    Next
    INDEX
    Explanations

    food descriptions

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

    The neuron activates most strongly on the names or titles of specific dishes (e.g. “Italian Beef Sandwich,” “Mie goreng,” “Oat Bran Muffins,” etc.).

    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
     it
    -1.99
    this
    -1.89
    in
    -1.87
    <em>
    -1.81
    z
    -1.77
     this
    -1.73
    f
    -1.72
    h
    -1.70
    á
    -1.70
    j
    -1.63
    POSITIVE LOGITS
    Entidad
    1.66
     เรา
    1.65
    ujuk
    1.63
    abonnement
    1.52
    aziland
    1.52
     Schluß
    1.48
     Когда
    1.47
    龃
    1.47
    嚀
    1.46
    ámica
    1.45
    Activations Density 0.069%

    No Known Activations