© 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. 99058
    Prev
    Next
    INDEX
    Explanations

    scuba diving or protease enzyme

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

    East Asian names like Tae, Shou, Shao, and Puddu.

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

    The neuron fires on the subword pieces that begin (i.e. the initial segment) of longer 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
    鄚
    -2.81
    鏵
    -2.70
    渽
    -2.63
    羕
    -2.61
    geſ
    -2.58
    ☄
    -2.53
     员
    -2.50
    just
    -2.50
    鋮
    -2.50
     unrelenting
    -2.48
    POSITIVE LOGITS
     terrific
    3.20
     magnific
    3.13
    <bos>
    3.13
    {
    3.06
    ра
    3.05
    It
    3.00
     decidedly
    2.97
     delightfully
    2.95
    2
    2.91
     beautifully
    2.89
    Activations Density 0.004%

    No Known Activations