© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 40057
    Prev
    Next
    INDEX
    Explanations

    driving or metallic objects

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

    The neuron fires strongly on the key topical noun that names the main subject of a passage (e.g. sport, driving, sheet, vehicle, motor, traveling, circuit board).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     σ
    -1.45
    !
    -1.38
    就可以了
    -1.30
     heeft
    -1.28
    $}
    -1.27
     hebben
    -1.25
     снова
    -1.25
    </u>
    -1.23
    lagi
    -1.23
    。”
    -1.22
    POSITIVE LOGITS
    comings
    1.38
     jaja
    1.29
    During
    1.27
     doos
    1.25
    hacia
    1.23
     território
    1.23
     prate
    1.23
     vierge
    1.22
    southeast
    1.21
    💞
    1.21
    Activations Density 0.264%

    No Known Activations