© 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. 34-GEMMASCOPE-RES-131K
    7. 23691
    Prev
    Next
    INDEX
    Explanations

    SXSW

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

    The neuron activates on uppercase event or conference headings—especially the “SXSW” label, the year (e.g. “2015”), and the following category (like “Sports and Technology Recap”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    gusted
    -0.86
     Moseley
    -0.77
    OJ
    -0.74
     chore
    -0.73
    чув
    -0.72
    ppery
    -0.72
    chakra
    -0.71
    塾
    -0.71
     insta
    -0.71
     Cracks
    -0.70
    POSITIVE LOGITS
     SX
    1.78
    SX
    1.48
     Interactive
    1.23
     Austin
    1.13
     sx
    1.07
     interactive
    1.05
    Interactive
    0.99
    interactive
    0.97
    sx
    0.90
    panels
    0.88
    Activations Density 0.034%

    No Known Activations