© 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. 34041
    Prev
    Next
    INDEX
    Explanations

    prepared food and substances

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

    The neuron strongly activates on recipe‐style ingredient names and related cooking‐oriented nouns (e.g. pickles, slaw, meatballs, mixture, filling).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     Castell
    -0.97
    usstsein
    -0.96
    醒了
    -0.96
    玟
    -0.92
    անգ
    -0.91
    頭の
    -0.90
    ضات
    -0.88
    辅导
    -0.88
     yemek
    -0.88
     metals
    -0.88
    POSITIVE LOGITS
     prepared
    1.63
     created
    1.56
     from
    1.37
     consisting
    1.24
     constructed
    1.22
    Prepared
    1.22
     made
    1.21
    prepared
    1.19
     homemade
    1.17
     generated
    1.16
    Activations Density 0.143%

    No Known Activations