© 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. 10-GEMMASCOPE-RES-131K
    7. 65826
    Prev
    Next
    INDEX
    Explanations

    police-involved shootings

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

    The neuron fires on the very first word of a new document section or heading, such as an author’s name, a place name, or a 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
     In
    -2.75
    蛧
    -2.69
    ꔹ
    -2.63
    </h3>
    -2.61
     Also
    -2.48
     ด้วย
    -2.48
    ⟞
    -2.45
     intricate
    -2.44
     Because
    -2.42
    あります
    -2.41
    POSITIVE LOGITS
    庒
    2.95
    蠵
    2.84
    obtained
    2.64
    烀
    2.56
    纐
    2.52
    
    2.52
    
    2.47
     jezt
    2.45
    蜑
    2.44
    received
    2.36
    Activations Density 0.008%

    No Known Activations