© 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. 34-GEMMASCOPE-RES-131K
    7. 75581
    Prev
    Next
    INDEX
    Explanations

    History, Releases, Architecture, Critical reception

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

    The neuron consistently lights up on numeric tokens—years, percentages, times, counts, and other numbers—marking out the document’s numeric data and statistics.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    
    -1.35
    Теперь
    -1.12
    Есть
    -1.00
     ТАК
    -0.99
    Morfo
    -0.99
    Здесь
    -0.99
     БЕ
    -0.98
    Будь
    -0.97
     ДЕ
    -0.97
     öf
    -0.97
    POSITIVE LOGITS
     See
    1.48
    See
    1.40
     reception
    1.36
    Notes
    1.32
     criticism
    1.25
     see
    1.24
     usage
    1.24
     future
    1.20
    see
    1.19
     Other
    1.18
    Activations Density 0.114%

    No Known Activations