© 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. 5528
    Prev
    Next
    INDEX
    Explanations

    matte grey finishes

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

    The neuron fires on specific automobile parts and accessories (e.g. fender flares, headers, exhaust, bumper scuffs, wheels).

    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
     payable
    0.59
     governed
    0.57
     Niles
    0.55
     Payable
    0.55
    中医
    0.54
    驴
    0.54
     fortr
    0.53
     fertility
    0.52
     obstetric
    0.52
     transc
    0.52
    POSITIVE LOGITS
     styling
    1.10
     Styling
    1.09
    휀
    1.03
     fender
    0.99
    styling
    0.98
    钣
    0.97
     fenders
    0.96
    Exterior
    0.95
     hatchback
    0.94
     grille
    0.93
    Activations Density 0.323%

    No Known Activations