© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 58097
    Prev
    Next
    INDEX
    Explanations

    sys followed by log or system component

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

    The neuron detects occurrences of the substring “sys” (e.g. in syslog, syscall, etc.).

    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
     fueron
    -0.90
     entire
    -0.90
     muncul
    -0.88
     increasingly
    -0.87
     wöch
    -0.85
     fairly
    -0.84
     even
    -0.83
     minimal
    -0.82
     enormously
    -0.82
     occasionally
    -0.82
    POSITIVE LOGITS
     SYSTEM
    1.51
     sys
    1.48
     Sys
    1.41
     System
    1.38
     syst
    1.34
    sys
    1.23
     Syst
    1.21
     systé
    1.21
    SYSTEM
    1.20
    System
    1.15
    Activations Density 0.012%

    No Known Activations