© 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. 19474
    Prev
    Next
    INDEX
    Explanations

    locations or conditions specified

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

    This neuron detects occurrences of the subordinating conjunction “where.”

    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
     also
    -1.96
     and
    -1.84
    इटम
    -1.80
     of
    -1.73
    อีก
    -1.70
     perlahan
    -1.65
     required
    -1.62
    This
    -1.59
     would
    -1.58
    
    -1.58
    POSITIVE LOGITS
    ang
    1.71
    il
    1.66
     it
    1.50
    om
    1.49
     gezogen
    1.45
    ⁛
    1.45
    in
    1.41
     furo
    1.39
    ever
    1.38
    Cuáles
    1.37
    Activations Density 0.027%

    No Known Activations