© 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. 128672
    Prev
    Next
    INDEX
    Explanations

    names and medical specialties

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

    proper nouns like people's last names, brand names, and place names.

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

    The neuron responds to low‐frequency, out‐of‐distribution tokens—uncommon names, technical stems, acronyms, or emphatic spellings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     whoſe
    -2.58
    r
    -2.53
    龊
    -2.52
    y
    -2.48
    
    -2.45
     ligeramente
    -2.42
    o
    -2.38
     incluida
    -2.36
     againſt
    -2.34
    OUTUBE
    -2.33
    POSITIVE LOGITS
    </strong>
    3.06
    ↵↵
    3.00
    *
    2.95
    <em>
    2.95
    ي
    2.64
     і
    2.48
     they
    2.31
    <<<<<<<
    2.31
    訁
    2.30
    ل
    2.28
    Activations Density 0.015%

    No Known Activations