© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16497
    Prev
    Next
    INDEX
    Explanations

    listing items in sequence

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

    This neuron detects structural document markers—especially numbered headings, section/chapter labels, and other numeric outline elements.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     that
    -1.19
     this
    -1.14
     because
    -1.00
     after
    -0.94
     simply
    -0.92
     undeniably
    -0.91
    ieš
    -0.88
     merely
    -0.87
     Almost
    -0.87
     which
    -0.87
    POSITIVE LOGITS
     including
    1.98
    包括
    1.98
    including
    1.88
     včetně
    1.63
     включая
    1.55
     includes
    1.50
     incluindo
    1.48
     INCLUDING
    1.46
     einschließlich
    1.45
    Including
    1.41
    Activations Density 0.036%

    No Known Activations