© 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. 15119
    Prev
    Next
    INDEX
    Explanations

    Cos prefixes and Nes prefixes

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

    The neuron strongly activates on three-letter sequences with an initial capital letter (e.g. “Nes,” “Cos,” etc.).

    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
    s
    -3.80
    紲
    -3.53
    ༑
    -2.89
    镚
    -2.56
    ꦢ
    -2.52
    鯢
    -2.47
    ophageal
    -2.47
     GERMANY
    -2.45
    м
    -2.41
    اتها
    -2.38
    POSITIVE LOGITS
    {
    3.97
    "
    3.09
    </h3>
    2.53
    诵
    2.44
    </h2>
    2.44
    that
    2.44
    ’
    2.38
    as
    2.34
     And
    2.33
     。
    2.31
    Activations Density 0.004%

    No Known Activations