© 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. 22-GEMMASCOPE-RES-131K
    4. 68423
    Prev
    Next
    INDEX
    Explanations

    numbers and time phrases

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

    The neuron strongly activates on tokens representing numbers (digits and numeric strings).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     rám
    -1.05
     السياس
    -0.96
     demonstrations
    -0.96
    慳
    -0.96
     åren
    -0.94
    しそう
    -0.92
     erfordert
    -0.91
     beruht
    -0.91
     deaktivi
    -0.90
    iori
    -0.89
    POSITIVE LOGITS
    淠
    1.18
     posiad
    1.09
     gonna
    1.08
     my
    1.05
     constitución
    1.02
     I
    1.02
     fixa
    1.00
    ᾳ
    1.00
     verdura
    1.00
    dő
    1.00
    Activations Density 0.021%

    No Known Activations