© 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. 26459
    Prev
    Next
    INDEX
    Explanations

    entity relationships in code

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

    The neuron activates on plural nouns (words ending in “s”).

    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
    ственную
    -1.01
    lardan
    -0.99
     ortam
    -0.98
     çeşit
    -0.97
     philosophie
    -0.94
     andrea
    -0.88
     melihat
    -0.88
    Жит
    -0.87
    周年
    -0.87
     because
    -0.86
    POSITIVE LOGITS
    addition
    1.03
     ของ
    1.03
    都知道
    1.01
     Of
    1.00
    Of
    1.00
     revamped
    0.97
     dodatk
    0.96
    &
    0.96
     Ś
    0.95
     апре
    0.95
    Activations Density 0.105%

    No Known Activations