© 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. 34-GEMMASCOPE-RES-131K
    7. 47657
    Prev
    Next
    INDEX
    Explanations

    UI layout panels

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

    This neuron activates on integer literals in code (numeric tokens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    tableWidget
    -1.02
    viewmodel
    -0.98
    章节错误
    -0.95
    myArray
    -0.93
    Recherche
    -0.90
     hurrying
    -0.89
     emplace
    -0.89
    Gek
    -0.88
    لسی
    -0.88
     tableware
    -0.88
    POSITIVE LOGITS
     main
    1.21
     content
    1.16
     containing
    1.16
     top
    1.14
     background
    1.14
     CONTAIN
    1.13
    Top
    1.10
     outer
    1.10
     container
    1.09
    containing
    1.08
    Activations Density 0.002%

    No Known Activations