© 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. 39419
    Prev
    Next
    INDEX
    Explanations

    technical/programming keywords

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

    The neuron responds to tokens presented in all‐uppercase form—e.g. SQL keywords (IF, LOCAL), acronyms, or other all‐caps identifiers.

    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
    璋
    -0.90
    📮
    -0.89
     heren
    -0.88
    δος
    -0.88
     burberry
    -0.88
     auteurs
    -0.85
     præ
    -0.85
    .",
    -0.85
     propon
    -0.83
    På
    -0.83
    POSITIVE LOGITS
     были
    0.84
     jotka
    0.77
    paque
    0.77
     очень
    0.76
     $
    0.75
    itif
    0.75
     verlangt
    0.74
    三个
    0.73
     анг
    0.73
     So
    0.72
    Activations Density 0.037%

    No Known Activations