© 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. 7244
    Prev
    Next
    INDEX
    Explanations

    bilayers or greek letters

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

    The neuron activates on numeric tokens—i.e. measurement values, quantities, and other numbers—in scientific or technical text.

    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
     provides
    -1.64
     using
    -1.59
     has
    -1.59
     ALWAYS
    -1.55
     revital
    -1.53
    Both
    -1.48
     enhances
    -1.48
     might
    -1.47
     gave
    -1.45
     seemed
    -1.45
    POSITIVE LOGITS
     Ê
    1.84
     Dinas
    1.80
     😁
    1.70
     diversos
    1.67
    撅
    1.64
    킁
    1.63
    creating
    1.62
    
    1.62
    鹚
    1.62
     ........
    1.61
    Activations Density 0.069%

    No Known Activations