© 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. 55628
    Prev
    Next
    INDEX
    Explanations

    allocation, update, transport, constraint, usage, problem, stack, cache, join, connection, policies, scope, start, settings

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

    The neuron fires on capitalized identifier tokens in code—e.g. type, struct, field, or constant names in CamelCase or ALL‐CAPS.

    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
     새로운
    -1.25
    學會
    -1.20
     putem
    -1.17
    出てくる
    -1.17
    ほとん
    -1.11
    跑去
    -1.09
    昔の
    -1.07
    わざわざ
    -1.07
    あんな
    -1.07
     participer
    -1.06
    POSITIVE LOGITS
     their
    1.46
    StandardCharsets
    1.39
     objekt
    1.35
     other
    1.26
     these
    1.23
    begin
    1.20
     """",
    1.19
    ・
    1.17
    getSystemService
    1.16
     febru
    1.15
    Activations Density 0.148%

    No Known Activations