© 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. 34-GEMMASCOPE-RES-131K
    7. 830
    Prev
    Next
    INDEX
    Explanations

    license text or distributed under

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

    The neuron fires strongly on all‐caps tokens or acronyms (e.g. section headings, license keywords, and macro names) in documents.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     UPD
    -0.79
     OLS
    -0.67
    niversitesi
    -0.65
    rew
    -0.64
     Anschließend
    -0.63
    vski
    -0.63
    encode
    -0.63
     KEV
    -0.63
     Sask
    -0.63
     kev
    -0.61
    POSITIVE LOGITS
    Hills
    0.77
     Пор
    0.73
     пак
    0.71
     szolg
    0.71
    封闭
    0.66
    ból
    0.65
     remembers
    0.65
     bölüm
    0.65
    FileSize
    0.64
     Eliot
    0.64
    Activations Density 0.078%

    No Known Activations