© 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. 44189
    Prev
    Next
    INDEX
    Explanations

    Friedrich, Christina, Oligosaccharide

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

    names beginning with 'Friedrich' and scientific prefixes like 'Oligo'.

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

    The neuron activates on capitalized entity tokens—especially proper names and other headline-style capitalized words.

    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
    "
    -3.09
     But
    -2.66
     этих
    -2.56
     этот
    -2.50
     bede
    -2.44
    
    -2.34
     Есть
    -2.31
     صفحۀ
    -2.25
    幫助
    -2.23
    用の
    -2.19
    POSITIVE LOGITS
    ୈ
    2.97
    猡
    2.83
    裊
    2.83
    婼
    2.75
    "}}>
    2.72
    酲
    2.67
    ↵
    2.59
    嫫
    2.52
    </h3>
    2.50
    鉏
    2.50
    Activations Density 0.007%

    No Known Activations