© 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. 22-GEMMASCOPE-RES-131K
    7. 93327
    Prev
    Next
    INDEX
    Explanations

    ends, boundaries, edges

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

    The neuron consistently fires on tokens that mark boundaries or “endpoints” in many contexts—e.g. CSS structural markers like “last-child” and “{”, chemical descriptions of tube “ends,” integration limits “a” and “b,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     alimentaires
    -0.99
    ynchronous
    -0.95
     vests
    -0.91
     الفترة
    -0.88
     voren
    -0.87
     économiques
    -0.86
    发布时间
    -0.85
     décennies
    -0.84
    ruary
    -0.83
     نسبة
    -0.83
    POSITIVE LOGITS
     boundary
    1.13
     border
    1.08
     end
    0.96
     edge
    0.92
     special
    0.84
     ends
    0.82
     borders
    0.81
     edges
    0.81
     Boundary
    0.79
     termination
    0.79
    Activations Density 0.077%

    No Known Activations