© 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. 105272
    Prev
    Next
    INDEX
    Explanations

    items followed by articles or names

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

    The neuron fires on standalone capitalized tokens that begin headings or list entries (e.g. section titles, department names, or item 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
     our
    -2.61
     both
    -2.42
     people
    -2.41
     this
    -2.14
     using
    -2.11
    After
    -2.06
     make
    -1.98
    When
    -1.92
     before
    -1.84
     making
    -1.84
    POSITIVE LOGITS
     ส์
    2.05
    訁
    2.02
    issé
    1.89
     čty
    1.74
     siè
    1.74
    딪
    1.74
     stockfoto
    1.72
    ⮑
    1.70
    饧
    1.67
    issant
    1.66
    Activations Density 0.055%

    No Known Activations