© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21892
    Prev
    Next
    INDEX
    Explanations

    specific names and terms

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

    The neuron selectively activates on uncommon or high‐information tokens—proper names, specialized terms, and other relatively rare words or characters—in the text.

    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
    ⬯
    -3.41
    $
    -3.13
    </em>
    -2.83
    cayaan
    -2.45
     \
    -2.45
    蒾
    -2.42
    敉
    -2.42
    vantaged
    -2.42
    titial
    -2.41
    籥
    -2.39
    POSITIVE LOGITS
    ed
    2.94
    kawas
    2.50
    蠵
    2.44
    <i>
    2.41
     unglaub
    2.39
    replyreport
    2.38
    后
    2.33
     The
    2.28
    และ
    2.22
     comprado
    2.22
    Activations Density 0.013%

    No Known Activations