© 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. 22-GEMMASCOPE-RES-131K
    7. 15750
    Prev
    Next
    INDEX
    Explanations

    exclamations and periods

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

    The neuron is sensitive to section headings or emphatic slogan‐style text—especially words or short phrases in prominent all‐caps or title formatting.

    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
     Основ
    -1.06
     helado
    -1.04
    ПО
    -0.98
    
    -0.96
    Harmon
    -0.96
    ПРО
    -0.95
     Entonces
    -0.93
    
    -0.93
     стоимо
    -0.93
     or
    -0.92
    POSITIVE LOGITS
     always
    1.52
    !
    1.24
     sustancias
    1.15
    ！』
    1.15
     stronger
    1.15
     excelencia
    1.13
    論
    1.09
     sinistra
    1.09
     tomorrow
    1.08
    !”
    1.07
    Activations Density 0.013%

    No Known Activations