© 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. 107535
    Prev
    Next
    INDEX
    Explanations

    at least

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

    The neuron flags the formal “claim” language of patents – i.e. the heavy-usage function words that structure technical/process claims (comprising, of, to, selected, wherein, thereof, etc.).

    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
    USART
    -0.91
     прекрасно
    -0.89
     विवरण
    -0.88
     —
    -0.86
    Nascimento
    -0.85
    ormous
    -0.83
     composizione
    -0.82
    Estudi
    -0.81
    かれた
    -0.81
    uname
    -0.81
    POSITIVE LOGITS
     about
    1.52
    About
    0.99
    ABOUT
    0.98
     almeno
    0.97
     español
    0.95
     preferably
    0.93
     at
    0.91
     atleast
    0.90
     said
    0.88
    至少
    0.88
    Activations Density 0.106%

    No Known Activations