© 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. 65580
    Prev
    Next
    INDEX
    Explanations

    years ago, age

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

    The neuron activates on numeric tokens (e.g. ages, measurements, or other numbers) 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
     is
    -1.92
     if
    -1.84
    ↵↵↵↵↵
    -1.81
     you
    -1.81
     of
    -1.80
    ↵↵↵↵↵↵↵↵
    -1.80
     with
    -1.72
    ↵↵
    -1.72
    ↵↵↵↵
    -1.70
     by
    -1.69
    POSITIVE LOGITS
    щены
    1.80
     Présentation
    1.80
    seiti
    1.70
    ⾞
    1.70
     réfrig
    1.66
     véhic
    1.65
    骉
    1.62
    ᥬ
    1.61
     fonctionnent
    1.57
     Estaba
    1.56
    Activations Density 0.132%

    No Known Activations