© 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. 21673
    Prev
    Next
    INDEX
    Explanations

    JSON configuration files

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

    This neuron fires strongly on object key names (i.e. property identifiers before a colon) in JSON‐ or JavaScript‐style data structures.

    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
     if
    -1.85
     because
    -1.69
     by
    -1.66
     можем
    -1.54
    ilets
    -1.41
     garner
    -1.38
     as
    -1.38
        
    -1.37
     providing
    -1.30
     at
    -1.30
    POSITIVE LOGITS
     are
    1.70
     übri
    1.66
    demás
    1.65
     obligaciones
    1.59
     donnés
    1.54
     北市
    1.54
     vänner
    1.53
    ();
    1.49
     mismas
    1.46
     списка
    1.46
    Activations Density 0.042%

    No Known Activations