© 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. 74936
    Prev
    Next
    INDEX
    Explanations

    medical and demographic classifications

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

    numerical values and numbers in scientific research papers.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron detects numeric tokens (e.g. counts, years, measurements, sample sizes) in scientific/medical text.

    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
     will
    -1.21
     or
    -1.06
     such
    -0.99
     in
    -0.98
    (("
    -0.91
    =\"
    -0.90
     باشه
    -0.90
     an
    -0.90
     whose
    -0.90
     طولانی
    -0.90
    POSITIVE LOGITS
    basically
    1.28
    generally
    1.27
    clearly
    1.27
    presumably
    1.21
    seems
    1.11
    Redaktion
    1.10
    particularly
    1.08
    apparently
    1.07
    tells
    1.06
    Whereas
    1.05
    Activations Density 0.100%

    No Known Activations