© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 95467
    Prev
    Next
    INDEX
    Explanations

    comparisons with than or compared to

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

    The neuron activates on numeric tokens—especially measurement values, counts, and comparison numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     önem
    -1.29
    مقال
    -1.04
     返回
    -1.04
    TIPOS
    -1.03
     niyang
    -1.02
     vielleicht
    -1.02
    PARTE
    -0.98
     {
    -0.97
    INGREDIENTES
    -0.95
    Trotz
    -0.94
    POSITIVE LOGITS
     other
    1.36
     those
    1.29
     magasiner
    1.23
     EconPapers
    1.16
     both
    1.14
     corresponding
    1.13
     their
    1.12
     others
    1.11
    遑
    1.10
     in
    1.09
    Activations Density 0.055%

    No Known Activations