© 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. 10-GEMMASCOPE-RES-131K
    7. 45415
    Prev
    Next
    INDEX
    Explanations

    world's first, most, best, largest

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

    The neuron fires on superlative qualifiers—especially words like “first” and “most.”

    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
     in
    -1.84
     It
    -1.78
     people
    -1.74
     you
    -1.71
     there
    -1.68
     Despite
    -1.65
     According
    -1.62
     There
    -1.60
     within
    -1.59
    后面的
    -1.56
    POSITIVE LOGITS
     และ
    1.73
    鮸
    1.52
    tho
    1.48
    、
    1.48
     ſol
    1.47
    taler
    1.47
    甞
    1.46
     runde
    1.46
     ซึ่ง
    1.42
    และ
    1.41
    Activations Density 0.020%

    No Known Activations