© 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. 32540
    Prev
    Next
    INDEX
    Explanations

    credits and roles

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

    The neuron fires on bibliographic‐style metadata markers—section headers and bylines (e.g. “Introduction,” “Credits:,” “Story by,” semicolons after author names), i.e. the tokens used in headings and author/credit lines.

    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
     deviennent
    -1.24
    Ανα
    -1.02
     congrès
    -0.95
    Οι
    -0.92
     prennent
    -0.92
    Ótimo
    -0.91
     komer
    -0.91
    Από
    -0.91
    allows
    -0.90
    يمكن
    -0.90
    POSITIVE LOGITS
     additional
    1.87
     produced
    1.69
     with
    1.67
     assistance
    1.66
     assisted
    1.62
     assistant
    1.55
     photography
    1.43
     music
    1.41
    additional
    1.32
     supported
    1.28
    Activations Density 0.066%

    No Known Activations