© 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. 10-GEMMASCOPE-RES-131K
    7. 59454
    Prev
    Next
    INDEX
    Explanations

    defibrillation, gal, physio, Chron

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

    scientific and medical terminology, particularly detecting syllables in technical terms like "defibrillation", "galactosidase", and "topobathymetric".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    翾
    -1.84
     who
    -1.75
     other
    -1.74
     saranno
    -1.65
    虢
    -1.57
     wohin
    -1.48
     vollständ
    -1.48
     ganzes
    -1.48
     serão
    -1.46
     besonderes
    -1.44
    POSITIVE LOGITS
    Actually
    1.66
    躋
    1.60
     ingat
    1.58
    鑄
    1.58
    的脸
    1.52
    擷
    1.51
     éclairage
    1.50
    伉
    1.47
     régal
    1.47
     couteau
    1.47
    Activations Density 0.502%

    No Known Activations