© 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. 22-GEMMASCOPE-RES-131K
    4. 27941
    Prev
    Next
    INDEX
    Explanations

    mathematical and technical concepts

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

    The neuron fires on tokens that mark and separate items in structured lists—e.g. numerals, section or question numbers, commas and colons used in outlines.

    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
     dör
    -1.31
     minimalis
    -1.27
    ѕ
    -1.25
    醯
    -1.22
     konst
    -1.16
     treff
    -1.16
    iacute
    -1.15
     sorg
    -1.14
     lagen
    -1.13
     serien
    -1.13
    POSITIVE LOGITS
     questions
    1.35
    cznych
    1.30
     教室
    1.28
    Esperamos
    1.23
    //};
    1.23
     students
    1.23
     topics
    1.15
    //*/
    1.15
     silnika
    1.12
    AllowAnonymous
    1.12
    Activations Density 0.028%

    No Known Activations