© 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. 130964
    Prev
    Next
    INDEX
    Explanations

    sentence endings

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

    The main thing this neuron does is detect numeric tokens (including numbers in text, citations, dates, and numeric measurements).

    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
    一颗
    -1.91
     ignited
    -1.77
    ↵↵
    -1.70
    痦
    -1.66
    
    -1.63
     вернулся
    -1.59
     cuya
    -1.59
    の人が
    -1.58
    薷
    -1.56
    am
    -1.55
    POSITIVE LOGITS
     but
    1.87
    .
    1.83
     jao
    1.61
    Karena
    1.60
    his
    1.52
     njego
    1.49
     even
    1.49
    Pero
    1.45
    Kritik
    1.43
    </sup>
    1.41
    Activations Density 0.001%

    No Known Activations