© 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. 10-GEMMASCOPE-RES-131K
    7. 65782
    Prev
    Next
    INDEX
    Explanations

    technical variable names

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

    The neuron detects all-caps or mostly uppercase tokens—i.e. acronyms, abbreviations, or identifiers written in uppercase.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     with
    -2.03
     both
    -2.00
     so
    -1.82
     from
    -1.74
     this
    -1.73
     by
    -1.73
     where
    -1.70
    Both
    -1.67
     had
    -1.64
     even
    -1.63
    POSITIVE LOGITS
    姽
    1.55
    
    1.53
     disfraces
    1.50
    antism
    1.48
    啻
    1.45
     soportes
    1.45
    珵
    1.45
    🫃
    1.43
    ynki
    1.43
     Dinas
    1.38
    Activations Density 0.406%

    No Known Activations