© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 19647
    Prev
    Next
    INDEX
    Explanations

    thirds, halves, quarters

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

    The neuron detects spelled‐out fractional expressions (e.g. “one-third,” “two-thirds”) in the text.

    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
    .
    -1.70
    ↵↵
    -1.45
     Now
    -1.42
    0
    -1.42
     samme
    -1.31
     has
    -1.31
    挝
    -1.26
     запро
    -1.25
            
    -1.23
     câble
    -1.21
    POSITIVE LOGITS
    licante
    1.49
     last
    1.48
     the
    1.48
    requipa
    1.42
    懌
    1.35
    が出る
    1.32
    chenden
    1.31
    rías
    1.30
     astonishing
    1.29
    中でも
    1.27
    Activations Density 0.014%

    No Known Activations