© 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. 19504
    Prev
    Next
    INDEX
    Explanations

    transcript content

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

    The neuron detects document section headings or metadata markers—especially the word “Transcript” (and similar header labels).

    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
     were
    -1.77
     So
    -1.64
     my
    -1.61
     tellement
    -1.50
     fazendo
    -1.48
     <
    -1.48
    一趟
    -1.48
    dée
    -1.48
     primeiro
    -1.47
     então
    -1.47
    POSITIVE LOGITS
    Despite
    1.72
     隊
    1.59
     estadounid
    1.59
    Regardless
    1.56
    持續
    1.55
     that
    1.52
    員工
    1.51
    僱
    1.51
    規劃
    1.49
    變成
    1.48
    Activations Density 0.005%

    No Known Activations