© 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. 10-GEMMASCOPE-RES-131K
    7. 3677
    Prev
    Next
    INDEX
    Explanations

    Suppose followed by math problems

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

    This neuron detects the typographic markers that introduce quoted or listed segments—e.g. opening quotation marks, bullets or numbered‐list headers.

    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
    ềm
    -1.69
     mètre
    -1.68
     FÜR
    -1.62
    艄
    -1.59
    czego
    -1.58
     quinze
    -1.56
     evenement
    -1.52
    쐈
    -1.52
     objectif
    -1.50
     accidente
    -1.48
    POSITIVE LOGITS
     not
    2.11
     in
    1.99
    b
    1.85
     all
    1.71
    全部
    1.61
     to
    1.57
     with
    1.54
     when
    1.53
    z
    1.52
    n
    1.50
    Activations Density 0.002%

    No Known Activations