© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 23980
    Prev
    Next
    INDEX
    Explanations

    collision and intersection detection

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

    The neuron fires on mentions of collisions (e.g. “collision,” “colliding,” “collide”) in technical or code contexts.

    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
    </h3>
    -1.84
    ;
    
    -1.52
    $}
    -1.49
    </h5>
    -1.48
    );
    
    -1.45
    henswürdigkeiten
    -1.45
     same
    -1.45
     появились
    -1.45
     nación
    -1.45
    </u>
    -1.38
    POSITIVE LOGITS
    <bos>
    1.70
     ruso
    1.41
     with
    1.38
     comigo
    1.38
    おわりに
    1.34
    (?)
    1.34
     dinosaurio
    1.33
    \
    1.32
     our
    1.31
    gerakan
    1.30
    Activations Density 0.009%

    No Known Activations