© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130304
    Prev
    Next
    INDEX
    Explanations

    The neuron mainly activates on numeric tokens (digits, decimals, and number strings) wherever they appear.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming language constructs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     solicitado
    -0.84
    inerario
    -0.83
     anunciado
    -0.83
     católica
    -0.82
     navideño
    -0.82
    安卓
    -0.81
     Podemos
    -0.81
     οπο
    -0.80
     colectiva
    -0.79
    spod
    -0.77
    POSITIVE LOGITS
    main
    1.22
    本身
    1.07
    本体
    1.02
     main
    1.00
    default
    0.96
     itself
    0.92
    MAIN
    0.91
     samot
    0.90
    正常
    0.90
     Main
    0.89
    Activations Density 0.088%

    No Known Activations