© 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. 22-GEMMASCOPE-RES-131K
    7. 65806
    Prev
    Next
    INDEX
    Explanations

    of entities and their contexts

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

    The neuron strongly responds to plural nouns that denote groups or collections of entities.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     not
    -1.07
    部分
    -1.05
    ΑΡ
    -1.03
    Unternehmen
    -1.02
     hvilken
    -1.01
     verranno
    -1.01
    ):
    -1.00
     itself
    -1.00
    当
    -0.98
     hvorfor
    -0.98
    POSITIVE LOGITS
     who
    1.68
    が見える
    1.13
    塀
    1.06
     chers
    1.05
     dégust
    1.05
    うーん
    1.03
     olur
    1.02
     célé
    1.02
    whom
    1.02
     eccl
    1.00
    Activations Density 0.142%

    No Known Activations