© 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. 10-GEMMASCOPE-RES-131K
    7. 9772
    Prev
    Next
    INDEX
    Explanations

    TOC; __NOTOC__

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

    The neuron fires on instances of “TOC” (often in all‐caps or in file/URL names) indicating table-of-contents markers.

    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
    ⚋
    -2.72
    🫴
    -2.56
    ляции
    -2.50
    就是
    -2.42
    言うと
    -2.36
    的时候
    -2.31
    
    -2.30
    </i>
    -2.28
     iſt
    -2.27
    见状
    -2.23
    POSITIVE LOGITS
    ."
    2.70
     It
    2.42
    ate
    2.39
     Its
    2.17
     
    2.14
    翃
    2.09
    transQ
    2.09
     To
    2.03
    が集
    1.95
    or
    1.90
    Activations Density 0.002%

    No Known Activations