© 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. 44455
    Prev
    Next
    INDEX
    Explanations

    Commonwealth and Testament specific contexts

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

    This neuron fires on capitalized institutional or canonical titles—most strongly on “Testament” (as in Old/New Testament) and also on “Commonwealth.”

    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
     "
    -3.22
    8
    -3.00
    6
    -3.00
    4
    -2.86
    7
    -2.80
    ?"
    -2.78
    How
    -2.75
    9
    -2.73
    他的
    -2.58
    2
    -2.56
    POSITIVE LOGITS
     of
    3.20
    骓
    3.03
    ⽌
    2.89
    鲽
    2.78
     to
    2.69
    
    2.67
    姈
    2.66
     []
    2.64
    酲
    2.55
    洐
    2.48
    Activations Density 0.002%

    No Known Activations