© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72527
    Prev
    Next
    INDEX
    Explanations

    graph, ART, bomb, two

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

    This neuron fires on the main content words in headings or titles (the first significant keyword in a post or section title).

    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
     /
    -1.64
    b
    -1.55
     -
    -1.48
     (
    -1.45
    0
    -1.45
     Here
    -1.44
    ",
    -1.42
     ,
    -1.41
    !
    -1.40
     toutes
    -1.37
    POSITIVE LOGITS
     palav
    1.71
     then
    1.66
     menyen
    1.62
     sabar
    1.55
     bahay
    1.55
    𖣘
    1.55
     Souha
    1.54
    sequently
    1.52
     uye
    1.52
     Chriftian
    1.52
    Activations Density 0.075%

    No Known Activations