© 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. 103647
    Prev
    Next
    INDEX
    Explanations

    names like Siem

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

    This neuron fires on occurrences of single‐letter uppercase identifiers (like A, C, M, E, P, S, etc.) when they introduce or label a mathematical object or block (e.g. “Let A = …,” “A: …,” “E[ … ],” etc.).

    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
    amazing
    -1.53
    excellent
    -1.52
    awesome
    -1.52
    !*\
    -1.51
    eração
    -1.43
    鋨
    -1.41
    important
    -1.41
     みた
    -1.41
     voren
    -1.41
     signifikan
    -1.39
    POSITIVE LOGITS
    .
    2.11
     would
    1.95
     both
    1.68
     new
    1.65
     "
    1.60
     They
    1.59
    ทั้ง
    1.55
     two
    1.51
     particularly
    1.50
    是很
    1.50
    Activations Density 0.019%

    No Known Activations