© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2834
    Prev
    Next
    INDEX
    Explanations

    oily cholesterol

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

    The neuron fires strongly on cooking contexts—words naming cookware, ingredients (like oil), and imperative cooking actions (heat, sauté, add, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ချင်း
    0.63
     दुर्गा
    0.63
    뿌
    0.62
    pygame
    0.61
     sulfite
    0.59
     बीपीएससी
    0.58
     Cursor
    0.57
    dpi
    0.57
     تیر
    0.56
     Yerevan
    0.56
    POSITIVE LOGITS
     oil
    2.45
     oils
    2.36
     기름
    2.30
    油
    2.30
     Oil
    2.28
     fatty
    2.27
     Oils
    2.25
    油脂
    2.25
    Oil
    2.23
     روغن
    2.23
    Activations Density 0.676%

    No Known Activations