© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 104765
    Prev
    Next
    INDEX
    Explanations

    medical conditions and terminology

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

    medical and technical terms related to physical damage, injuries, or negative changes.

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

    The neuron responds to specific medical injury or complication terms (e.g. laceration, transection, hemorrhage, stenosis, leak, hematoma, aneurysm).

    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
    getch
    -1.12
    ★★★★★
    -1.10
     Anm
    -0.94
     Antony
    -0.93
    —
    -0.92
    <!--
    
    -0.87
    ik
    -0.86
     s
    -0.86
     pecu
    -0.84
     m
    -0.83
    POSITIVE LOGITS
     یا
    1.36
     gemakkelijk
    1.35
    หรือ
    1.34
     mariée
    1.23
     éton
    1.19
     querían
    1.19
     compréhen
    1.16
     podoba
    1.15
     eenvoudig
    1.14
     ساده
    1.13
    Activations Density 0.160%

    No Known Activations