© 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. 66614
    Prev
    Next
    INDEX
    Explanations

    bacterium | UITableViewCell | HTMLElement

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

    class or component names in programming code, particularly focusing on UI elements and software objects.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on multi‐word technical type/class names (e.g. HTMLElement, UIViewController, GtkWidget, bacterium) — in other words, it detects tokens that are identifiers for structured types or formal entities.

    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
     ,
    -3.14
     With
    -2.70
    u
    -2.66
     :
    -2.64
     )
    -2.58
    是
    -2.48
     (
    -2.44
     =
    -2.44
     ;
    -2.39
     再
    -2.34
    POSITIVE LOGITS
    稣
    3.13
    盦
    3.13
    穑
    2.88
    淠
    2.80
    jaqueta
    2.78
     própri
    2.72
    
    2.67
     nuov
    2.66
     semelh
    2.66
    蚨
    2.63
    Activations Density 0.001%

    No Known Activations