© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 9402
    Prev
    Next
    INDEX
    Explanations

    compiler and code definitions

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

    The neuron fires strongly on code statements that bring in external dependencies—e.g. include or import lines naming libraries, headers, or modules.

    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
    痦
    -1.17
    imals
    -1.16
    ethe
    -1.14
    淠
    -1.13
    itons
    -1.13
    氏が
    -1.13
    institutions
    -1.12
    ²(
    -1.11
    Interestingly
    -1.10
    ₁)
    -1.10
    POSITIVE LOGITS
     you
    1.48
    .”
    1.32
     ziehen
    1.31
     ganador
    1.27
     of
    1.27
     salariés
    1.21
    rit
    1.21
     your
    1.20
     there
    1.19
    _
    1.18
    Activations Density 0.132%

    No Known Activations