© 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. 98373
    Prev
    Next
    INDEX
    Explanations

    effect of a qualifier

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

    This neuron fires on words belonging to academic paper or section titles (e.g. the capitalized, title‐style terms that appear in headings and article titles).

    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
     素描
    -1.71
    乁
    -1.65
    .《
    -1.65
    likation
    -1.54
     geene
    -1.51
     diskusi
    -1.50
    här
    -1.48
    maining
    -1.48
     nikah
    -1.47
     排骨
    -1.45
    POSITIVE LOGITS
    —
    1.75
     using
    1.74
     and
    1.73
     S
    1.68
     D
    1.66
    )
    1.60
     A
    1.58
    !
    1.57
    -
    1.57
     C
    1.56
    Activations Density 0.295%

    No Known Activations