© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71227
    Prev
    Next
    INDEX
    Explanations

    function names and technical terms

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

    The neuron fires on code identifiers—names of functions, types, variables, constants, etc., rather than on common keywords or punctuation.

    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
     professeurs
    -1.03
     dzisiaj
    -0.99
     prêtres
    -0.98
     virtudes
    -0.95
     klucz
    -0.95
     peuples
    -0.92
    Menurutnya
    -0.91
     nomme
    -0.90
    iscope
    -0.89
     bordada
    -0.89
    POSITIVE LOGITS
     Parmi
    1.00
     possibles
    0.96
     Catawiki
    0.90
     diabe
    0.89
     disgra
    0.88
     inaug
    0.86
     dynamis
    0.86
     Depuis
    0.85
     aptitudes
    0.85
     coiff
    0.85
    Activations Density 0.004%

    No Known Activations