© 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. 34-GEMMASCOPE-RES-131K
    7. 106433
    Prev
    Next
    INDEX
    Explanations

    circles and centers

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

    The neuron fires strongly on the word “circle” (and related geometry terms about circles and mappings), so it’s detecting mentions of circles or circle‐related geometry.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     rectangular
    -1.12
     triangular
    -1.08
    rectangle
    -1.01
    rectangular
    -0.97
     square
    -0.94
     round
    -0.92
     rect
    -0.92
    ENTRY
    -0.90
    Rect
    -0.90
     shaped
    -0.90
    POSITIVE LOGITS
     centers
    1.34
     centre
    1.27
     center
    1.24
     centered
    1.23
     centres
    1.19
     circles
    1.13
     Circles
    1.13
     circle
    1.12
    center
    1.10
    Circum
    1.09
    Activations Density 0.046%

    No Known Activations