© 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. 79939
    Prev
    Next
    INDEX
    Explanations

    code constructs

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

    The neuron detects opening delimiters—specifically the “(” and “<” tokens that introduce argument lists or generic/type-parameter brackets.

    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
     ausgewä
    -1.19
    ктан
    -1.13
    Ѯ
    -1.12
     シンプル
    -1.02
    但却
    -1.01
    鑁
    -1.00
     vindt
    -0.98
    接下來
    -0.98
    août
    -0.98
    rítica
    -0.96
    POSITIVE LOGITS
     Although
    1.09
     However
    1.07
     or
    1.05
    心理学
    1.05
    parece
    1.03
     In
    1.02
     an
    1.02
     although
    1.00
     that
    1.00
     While
    0.97
    Activations Density 0.005%

    No Known Activations