© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12077
    Prev
    Next
    INDEX
    Explanations

    sit, sitting, chairs, seats

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

    This neuron detects mentions of sitting or seating contexts (e.g. sit, sat, chairs).

    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
    ('|
    0.57
    antry
    0.54
     masker
    0.54
     filenames
    0.54
    Payroll
    0.54
    鈿
    0.52
    vanja
    0.50
    Running
    0.50
    Shield
    0.49
     contener
    0.49
    POSITIVE LOGITS
     seats
    2.61
     seating
    2.61
     seat
    2.59
    座椅
    2.53
     chairs
    2.50
    seats
    2.39
     Seats
    2.39
    seat
    2.34
    Seat
    2.34
     座
    2.34
    Activations Density 0.585%

    No Known Activations