© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 104627
    Prev
    Next
    INDEX
    Explanations

    documentation sections and keywords

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

    markdown headings (lines starting with `##`).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron specializes in spotting numeric tokens (especially numbers in headings or section markers) and outputs their numeric value.

    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
    畀
    -1.54
     for
    -1.44
     البته
    -1.41
    了一个
    -1.39
    出了一
    -1.30
    ervazione
    -1.30
    влено
    -1.30
    ouml
    -1.30
    了一眼
    -1.29
     then
    -1.28
    POSITIVE LOGITS
     mutiara
    1.72
     medlems
    1.55
     після
    1.49
     these
    1.46
     ($("#
    1.44
     tabung
    1.43
     aktivitet
    1.41
     skrift
    1.40
    ньому
    1.39
     kanak
    1.39
    Activations Density 0.001%

    No Known Activations