© 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. 22-GEMMASCOPE-RES-131K
    7. 67588
    Prev
    Next
    INDEX
    Explanations

    Fetch, parameters, URL, method, retrospective, gender, nerve, Hanson

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

    This neuron fires on programming-language constants or macro names—uppercase tokens (often with underscores or “::”) like PDO::PARAM_INT, PDO::FETCH_ASSOC, GLOB_BRACE, 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_22/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
     consideró
    -1.12
    が
    -1.08
    显然
    -1.03
     añadió
    -1.02
     encontrado
    -1.01
     для
    -1.01
     of
    -1.01
    へ
    -1.00
     meille
    -0.99
     État
    -0.99
    POSITIVE LOGITS
     歐
    1.16
     pudesse
    1.09
    ✳
    1.06
    Egy
    1.06
    を中心
    1.05
     تھے
    1.01
    unier
    0.97
    ꦁ
    0.97
     ン
    0.97
    Yeni
    0.93
    Activations Density 0.002%

    No Known Activations