© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 105180
    Prev
    Next
    INDEX
    Explanations

    function/method arguments

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

    words or phrases indicating sufficiency, adequacy, or reaching a threshold (especially "enough", "to", and related expressions).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on words that convey extremeness or emphasis—strong qualifiers and superlatives (e.g. ultimate, secret, greatest, large, easy).

    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
     orina
    -1.12
    Caractéristiques
    -1.12
     Secrétaire
    -1.06
     Spokes
    -1.04
    iskt
    -1.04
     méxico
    -1.04
    
    -1.03
     ſi
    -1.02
     -¿
    -1.02
    зонта
    -1.01
    POSITIVE LOGITS
     with
    1.57
     in
    1.54
     these
    1.38
     These
    1.38
     not
    1.34
    ↵
    1.34
     you
    1.25
     We
    1.25
     In
    1.23
    What
    1.23
    Activations Density 0.259%

    No Known Activations