© 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. 10-GEMMASCOPE-RES-131K
    4. 87132
    Prev
    Next
    INDEX
    Explanations

    two family members or servants

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

    The neuron is tuned to detect the numeral “two,” especially when it introduces pairs (e.g. two children, two sons, two daughters).

    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
     ;
    -1.53
    V
    -1.51
     nonché
    -1.43
    側は
    -1.43
     than
    -1.40
    id
    -1.36
    找个
    -1.35
     ؛
    -1.34
     و
    -1.32
     you
    -1.31
    POSITIVE LOGITS
     primeras
    1.88
    Bereich
    1.69
    Only
    1.55
    Ᏹ
    1.50
    Those
    1.48
     rezultat
    1.48
    ujem
    1.43
    íté
    1.42
    Dimensi
    1.41
     primeiras
    1.40
    Activations Density 0.013%

    No Known Activations