© 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. 12539
    Prev
    Next
    INDEX
    Explanations

    mixed languages and code headers

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

    The neuron is looking for numeric tokens (numbers and numeric values) 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
    t
    -1.97
    ↵
    -1.83
     are
    -1.74
    re
    -1.69
    </h3>
    -1.66
     was
    -1.65
    at
    -1.63
     prossima
    -1.62
     et
    -1.62
        
    -1.59
    POSITIVE LOGITS
    
    1.98
     که
    1.91
    haviours
    1.78
     TIEMPO
    1.77
     MÉXICO
    1.71
     mágico
    1.71
    𖡼
    1.67
    ୌ
    1.64
     やすい
    1.63
     esthétique
    1.63
    Activations Density 0.009%

    No Known Activations