© 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. 22-GEMMASCOPE-RES-131K
    7. 79671
    Prev
    Next
    INDEX
    Explanations

    technical questions and statements

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

    The neuron is highly responsive to code‐style structural tokens (e.g. JSON or notebook metadata punctuation and keywords) and thus marks spans of embedded code or markup.

    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
     hunde
    -1.53
     klinik
    -1.52
     truk
    -1.45
     praktis
    -1.38
     krim
    -1.38
     berge
    -1.38
     harta
    -1.32
     vok
    -1.31
     romantis
    -1.29
     keramik
    -1.27
    POSITIVE LOGITS
     prodej
    1.35
     cubrir
    1.33
     atualizado
    1.30
    includegraphics
    1.27
     adhesivo
    1.26
     verheir
    1.23
     alugar
    1.23
     zaji
    1.23
     descuentos
    1.22
     then
    1.21
    Activations Density 0.002%

    No Known Activations