© 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. 76532
    Prev
    Next
    INDEX
    Explanations

    parallel programming directives

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

    This neuron detects code-related punctuation and keywords used in build and preprocessor directives (e.g., “RUN:”, “#include”, “<…>”, “-”, “%”).

    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
    Firebase
    -0.76
    ZX
    -0.73
     computerized
    -0.71
    了一口气
    -0.70
    caten
    -0.69
    viera
    -0.68
     digitized
    -0.67
    米的
    -0.66
    jerg
    -0.66
     northerly
    -0.66
    POSITIVE LOGITS
     parallel
    2.22
     paralle
    2.06
     GPU
    2.05
    parallel
    1.95
    Parallel
    1.88
     Parallel
    1.84
     Paralle
    1.79
     parallelism
    1.71
    GPU
    1.66
    paralle
    1.63
    Activations Density 0.174%

    No Known Activations