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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 99792
    Prev
    Next
    INDEX
    Explanations

    sections and parts of works

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

    The neuron fires on formal academic structure markers (words like “article,” “section,” “part,” “result,” etc.) that indicate the organization of a scholarly document.

    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
     from
    -1.03
    číta
    -1.02
     eğer
    -1.00
     alkoholu
    -0.98
    ambiguation
    -0.98
     doba
    -0.97
     hiçbir
    -0.95
     pokoj
    -0.94
     transacción
    -0.91
     mlad
    -0.90
    POSITIVE LOGITS
     dealing
    1.52
     deals
    1.48
     devoted
    1.47
     discusses
    1.40
     deal
    1.39
     contains
    1.28
     discussing
    1.24
     discuss
    1.23
    Dealing
    1.23
    Deal
    1.20
    Activations Density 0.038%

    No Known Activations