© 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. 71349
    Prev
    Next
    INDEX
    Explanations

    end markers

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

    The neuron reliably activates on the token “end” (in any capitalization or context) when it appears in code or comments.

    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
     Verkäufer
    -1.16
     Descripción
    -1.15
     Gerechtigkeit
    -1.09
    Ayrıca
    -1.07
     previamente
    -1.06
     Evaluación
    -1.05
     Popis
    -1.05
    verture
    -1.04
    Confira
    -1.02
     公司
    -1.02
    POSITIVE LOGITS
     ended
    1.07
     now
    0.98
    End
    0.97
     espri
    0.96
     هذا
    0.90
     ends
    0.89
    への
    0.88
    棲
    0.88
     end
    0.87
     questo
    0.86
    Activations Density 0.004%

    No Known Activations