© 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. 25090
    Prev
    Next
    INDEX
    Explanations

    numbers and percentages

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

    The neuron fires on floating‐point numerals (numbers with decimals) in data, scientific, or statistical contexts.

    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
    鉿
    -1.83
    勦
    -1.83
    :
    -1.80
    鲮
    -1.63
    翆
    -1.59
     позволяет
    -1.58
    mismo
    -1.58
     nasceu
    -1.55
    isière
    -1.54
    えっ
    -1.53
    POSITIVE LOGITS
    2
    2.31
    1
    2.17
    3
    2.16
    6
    2.11
    5
    2.05
    7
    2.03
    8
    1.86
    a
    1.80
    9
    1.73
    4
    1.71
    Activations Density 0.085%

    No Known Activations