© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13936
    Prev
    Next
    INDEX
    Explanations

    preposition followed by noun

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

    The neuron detects numerical quantities (both integers and decimals) 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
     and
    -1.99
      
    -1.59
     which
    -1.41
     what
    -1.37
    .*;
    -1.34
     apie
    -1.34
     then
    -1.27
     who
    -1.27
    ʋ
    -1.20
    ;
    -1.20
    POSITIVE LOGITS
     myös
    1.28
    isbury
    1.26
     rằng
    1.21
    Czytaj
    1.19
     комби
    1.16
     célèbre
    1.16
    radas
    1.16
     גם
    1.15
    liehen
    1.14
    Törté
    1.14
    Activations Density 0.834%

    No Known Activations