© 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. 16129
    Prev
    Next
    INDEX
    Explanations

    entity description

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

    This neuron detects the lead-sentence “The <noun> is…” pattern used to introduce or define an entity.

    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
     those
    -1.57
     innumerable
    -1.31
     subsequently
    -1.31
     various
    -1.29
     any
    -1.26
     countless
    -1.26
     different
    -1.25
     adequate
    -1.22
     THEN
    -1.22
     even
    -1.20
    POSITIVE LOGITS
     is
    1.14
     กัน
    1.11
     avancer
    1.05
    Œ
    1.02
     avantages
    1.02
    ance
    1.02
     يتم
    0.99
    綰
    0.98
     نداره
    0.98
     manly
    0.98
    Activations Density 0.325%

    No Known Activations