© 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. 85055
    Prev
    Next
    INDEX
    Explanations

    prove that

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

    The neuron fires on numeric tokens—especially decimal numbers or measurement values—wherever they appear 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
     '[
    -1.99
    鏷
    -1.95
     '';
    -1.91
     câbles
    -1.91
     toalha
    -1.84
    丷
    -1.84
     donnent
    -1.84
     "
    -1.81
     mettent
    -1.79
     connues
    -1.78
    POSITIVE LOGITS
    :“
    2.48
     in
    2.30
    …”
    2.28
    ,”
    2.27
    <td>
    2.17
    ar
    2.13
    ,“
    2.13
    :”
    2.03
    ..”
    2.02
     through
    2.00
    Activations Density 0.003%

    No Known Activations