© 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. 8578
    Prev
    Next
    INDEX
    Explanations

    6G technologies

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

    The neuron fires strongly on single-letter unit abbreviations (like “G,” “k,” “v,” “d”) that follow numbers, i.e. measurement or specification units after numeric values.

    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
    涠
    -3.22
    岀
    -2.98
    ﱠ
    -2.92
    
    -2.92
     informou
    -2.92
    ्ञ
    -2.92
    ࣭
    -2.75
    犼
    -2.72
    [
    -2.70
    
    -2.70
    POSITIVE LOGITS
    You
    3.39
    I
    3.16
    G
    3.03
    //
    2.91
    :
    2.78
    ↵
    2.75
    per
    2.72
    AL
    2.72
     
    2.72
    で
    2.67
    Activations Density 0.078%

    No Known Activations