© 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. 39408
    Prev
    Next
    INDEX
    Explanations

    former handball player

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

    The neuron fires on structured “metadata” tokens—things like headings/labels and especially numbers (e.g. IDs, prices, distances, counts) rather than ordinary running 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_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
    iettivo
    -0.96
    cense
    -0.87
     MÉ
    -0.86
     sentarse
    -0.84
    conut
    -0.84
    porus
    -0.82
    Alder
    -0.81
    nish
    -0.80
    太过
    -0.80
    xiao
    -0.79
    POSITIVE LOGITS
    PBT
    0.91
    дневно
    0.86
    Woof
    0.86
    焓
    0.84
    Каждый
    0.84
    fören
    0.83
    alej
    0.82
    Germ
    0.81
     создавать
    0.81
     seeks
    0.79
    Activations Density 0.362%

    No Known Activations