© 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. 102794
    Prev
    Next
    INDEX
    Explanations

    lieu, lew, lec, leu, kind

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

    This neuron fires on the character sequence "leu" (e.g., in words like "lieu" or the amino-acid abbreviation "Leu").

    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
     принад
    -2.67
    end
    -2.64
    de
    -2.55
     nahezu
    -2.44
     fourteen
    -2.41
    ?
    -2.38
     gegenwär
    -2.31
    あるいは
    -2.30
     eighteen
    -2.25
    en
    -2.23
    POSITIVE LOGITS
     zoude
    2.61
     uſed
    2.50
     anún
    2.44
    糨
    2.44
    ruguay
    2.41
    '
    2.39
     Perſ
    2.39
     eşofman
    2.36
    蓠
    2.31
    黉
    2.27
    Activations Density 0.001%

    No Known Activations