© 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. 99609
    Prev
    Next
    INDEX
    Explanations

    FW, srcs, Ester, reachable

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

    The neuron responds most strongly to uncommon or infrequent tokens (i.e. rare words) 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
    ’
    -3.67
    wiście
    -3.14
    s
    -2.86
    ↵
    -2.69
    $
    -2.64
     好看
    -2.63
     pessoais
    -2.41
    缈
    -2.41
     conductores
    -2.36
    鸯
    -2.20
    POSITIVE LOGITS
    ).
    2.75
    t
    2.59
    It
    2.39
    各样的
    2.34
     menceritakan
    2.33
    >
    2.31
     tradisi
    2.25
     お
    2.25
    ing
    2.23
    ～～
    2.22
    Activations Density 0.017%

    No Known Activations