© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 33997
    Prev
    Next
    INDEX
    Explanations

    medical conditions and descriptions

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

    medical terminology for diseases, conditions, and anatomical structures.

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

    The neuron strongly activates on long, multi-part clinical terms—especially formal disease or syndrome names (e.g. rhabdomyolysis, hydroureteronephrosis, gliomatosis).

    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
     semelh
    -1.79
     miſ
    -1.74
     eſt
    -1.73
     ſol
    -1.72
     ſtre
    -1.70
    慢慢的
    -1.67
     insanely
    -1.63
     oogen
    -1.59
     reprendre
    -1.57
     unbelievably
    -1.57
    POSITIVE LOGITS
     or
    2.36
     with
    2.03
     all
    1.81
     without
    1.66
     first
    1.66
     just
    1.66
     longstanding
    1.62
     had
    1.61
     but
    1.59
     putting
    1.56
    Activations Density 0.908%

    No Known Activations