© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100243
    Prev
    Next
    INDEX
    Explanations

    grab and grabbed

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

    This neuron is strongly activated by occurrences of the word “grab” (in any case or context).

    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
    -2.55
    🌟
    -2.53
    🔥
    -2.48
    🌸
    -2.45
    💙
    -2.30
    違う
    -2.28
    💛
    -2.27
    𝑐
    -2.22
    我们可以
    -2.22
    二年
    -2.19
    POSITIVE LOGITS
    泚
    2.91
    愎
    2.86
    溷
    2.64
    𞤤
    2.50
    
    2.48
    笮
    2.47
    ࿆
    2.47
    待って
    2.47
    蹀
    2.47
     lesquels
    2.45
    Activations Density 0.003%
    No Known Activations