© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49452
    Prev
    Next
    INDEX
    Explanations

    D flags, D day, D prefixes

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

    The neuron fires on build‐configuration flags and definitions (e.g. compiler or JVM “-D…” and “-X…” options in build scripts).

    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
    墾
    -0.77
     einzigen
    -0.77
    -0.73
    Intrinsic
    -0.72
     nächste
    -0.71
     richieste
    -0.71
     recupero
    -0.70
     Миха
    -0.70
    đi
    -0.69
     those
    -0.69
    POSITIVE LOGITS
     D
    2.75
    D
    1.76
     Day
    1.69
     day
    1.63
    Day
    1.53
     DAY
    1.34
     Д
    1.27
     d
    1.24
     zero
    1.22
    DAY
    1.14
    Activations Density 0.038%

    No Known Activations