© 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. 66916
    Prev
    Next
    INDEX
    Explanations

    coordinates $(x,y)$

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

    The neuron fires on the beginnings of TeX‐style formulae and math-notation chunks (e.g. dollar signs, “\( … \)” delimiters, and other in-line math syntax).

    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
    </h4>
    -3.31
    </sup>
    -2.73
    ↵
    -2.70
    "
    -2.66
     of
    -2.55
    *
    -2.50
    而又
    -2.41
    ）
    -2.41
        
    -2.39
    0
    -2.34
    POSITIVE LOGITS
    
    3.25
    塒
    3.08
    睒
    2.95
    ꜆
    2.88
    坶
    2.66
    噯
    2.61
    褙
    2.59
    IBLIO
    2.58
    
    2.55
     Namun
    2.53
    Activations Density 0.008%

    No Known Activations