© 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. 89897
    Prev
    Next
    INDEX
    Explanations

    closing programming constructs

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

    This neuron is a detector for Fortran code keywords and directives (e.g. PROGRAM, PARAMETER, ALLOCATE, !$ACC PARALLEL/ASYNC, END DO), distinguishing code‐tokens from prose.

    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
     clove
    -0.83
     الكريم
    -0.81
    çok
    -0.79
     Māori
    -0.79
     CONCLUSION
    -0.78
    Inset
    -0.75
     Codex
    -0.74
    ᴬ
    -0.74
    INARY
    -0.73
    CONCLUSION
    -0.72
    POSITIVE LOGITS
     ends
    0.82
     children
    0.80
     each
    0.79
     destaque
    0.78
    会う
    0.77
    ↵↵↵↵
    0.75
    ル
    0.75
    衷
    0.74
     wensen
    0.74
     telefoon
    0.74
    Activations Density 0.019%

    No Known Activations