© 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. 8796
    Prev
    Next
    INDEX
    Explanations

    Sat or sitting

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

    The neuron spotlights any token beginning with the letters “sat,” whether it’s the standalone past-tense verb “sat” or a longer word like “sativa.”

    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
    update
    0.64
    暮らし
    0.63
     стоят
    0.59
    र्घ
    0.57
    owner
    0.56
     Flip
    0.53
    uffling
    0.52
    往
    0.51
    就在
    0.51
    uland
    0.51
    POSITIVE LOGITS
     Saty
    0.97
     Sat
    0.93
    Sat
    0.89
     sat
    0.86
     Satt
    0.79
     SAT
    0.75
     satisfiable
    0.73
    SAT
    0.71
     सत्
    0.70
     Sats
    0.70
    Activations Density 0.204%

    No Known Activations