© 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. 34523
    Prev
    Next
    INDEX
    Explanations

    neuroendocrine disorders

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

    medical and scientific terminology, particularly words related to biological systems and pathology.

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

    The neuron is triggered by substrings of complex medical compound terms—i.e. Greek/Latin roots and prefixes (like “endo‐,” “neuroend‐,” “hemo‐,” “pseudo‐,” “subcaps‐,” etc.) that appear inside longer clinical jargon.

    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
     eight
    -1.31
     and
    -1.24
     seven
    -1.23
     as
    -1.22
     one
    -1.20
     six
    -1.17
     like
    -1.16
     after
    -1.13
     from
    -1.11
     five
    -1.09
    POSITIVE LOGITS
     เรา
    1.03
     elegantly
    1.02
     magnific
    1.01
    mainly
    1.01
    MEMOR
    0.99
     REALLY
    0.98
     rapidement
    0.98
     actuellement
    0.98
     beige
    0.95
     principalement
    0.94
    Activations Density 0.340%

    No Known Activations