© 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. 25225
    Prev
    Next
    INDEX
    Explanations

    fractions and inequalities

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

    The neuron activates on numeric literals (integers or decimal numbers) in the text, especially when they appear in mathematical expressions.

    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
    ynchronous
    -1.17
     हालांकि
    -1.00
    timacy
    -0.94
    rdida
    -0.94
     csak
    -0.93
     Majesté
    -0.93
    such
    -0.92
    ributes
    -0.92
    íso
    -0.92
     NoSuch
    -0.91
    POSITIVE LOGITS
     than
    1.70
     other
    1.45
     others
    1.45
     usual
    1.22
     overall
    1.17
     otras
    1.16
     outros
    1.13
     any
    1.03
     estimates
    1.03
     bonitos
    1.02
    Activations Density 0.013%

    No Known Activations