© 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. 19632
    Prev
    Next
    INDEX
    Explanations

    new instance

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

    The neuron most strongly activates on low-frequency (i.e. rare or uncommon) tokens.

    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
     Boletín
    -1.79
    zehnten
    -1.65
     puis
    -1.64
    wände
    -1.61
    ære
    -1.61
    戋
    -1.59
     Penyakit
    -1.57
    کیف
    -1.52
     بدان
    -1.52
    selben
    -1.52
    POSITIVE LOGITS
     a
    2.17
    (
    1.97
     two
    1.96
     some
    1.71
     several
    1.71
     what
    1.70
     an
    1.63
     both
    1.62
    r
    1.59
     one
    1.55
    Activations Density 0.018%

    No Known Activations