© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20757
    Prev
    Next
    INDEX
    Explanations

    medical or process descriptions

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

    words or terms that end in "ency", "ate", "asis", or words like "Gate" and certain technical or anatomical terms.

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

    The neuron fires on uncommon, specialized noun tokens—especially those in proper names or technical/Latin‐derived terms.

    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
    </h3>
    -3.34
    ників
    -2.20
    .
    -2.17
    ffed
    -2.06
    艿
    -2.02
    ть
    -2.02
    /';
    -1.91
    </strong>
    -1.90
    ㄀
    -1.88
    PvB
    -1.84
    POSITIVE LOGITS
    9
    3.28
    4
    2.89
    6
    2.75
    8
    2.67
    5
    2.50
    0
    2.47
    3
    2.38
    '
    2.31
    2
    2.30
    7
    2.28
    Activations Density 0.035%

    No Known Activations