© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103053
    Prev
    Next
    INDEX
    Explanations

    SMC, streamers, UPI, Smiley, EMR, SMB

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

    acronyms and abbreviations, particularly three-letter capitalizations like "SMC", "UPI", and "Smiley" (when formatted as a proper name).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron strongly activates on tokens that are all-uppercase abbreviations or initialisms (acronyms).

    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
     intrigue
    -2.61
     identité
    -2.47
    齪
    -2.39
     時尚
    -2.34
     untold
    -2.33
    navn
    -2.31
     tomto
    -2.30
     amount
    -2.30
    肫
    -2.28
     assortment
    -2.25
    POSITIVE LOGITS
     After
    2.67
    *
    2.63
     Many
    2.56
    ו
    2.56
     There
    2.52
     可以
    2.41
     tulisan
    2.38
    を
    2.30
    to
    2.28
     После
    2.27
    Activations Density 0.013%

    No Known Activations