© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 47901
    Prev
    Next
    INDEX
    Explanations

    units of measurement and scale

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

    The neuron primarily detects numeric tokens—standalone digits and multi‐digit numbers (including scales and measurements).

    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
     AttributeSet
    -0.91
    Access
    -0.85
    timeout
    -0.82
    Ʉ
    -0.82
     Server
    -0.78
    AttributeSet
    -0.76
    access
    -0.73
    PAREN
    -0.73
    DIRECTORY
    -0.73
    Duration
    -0.72
    POSITIVE LOGITS
     scale
    4.00
    scale
    3.33
     Scale
    2.83
    Scale
    2.83
     scales
    2.72
     SCALE
    2.56
     escala
    2.45
     масшта
    2.45
    SCALE
    2.41
    échelle
    2.23
    Activations Density 0.011%

    No Known Activations