© 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. 50448
    Prev
    Next
    INDEX
    Explanations

    system control and modes

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

    This neuron fires on tokens consisting of all-uppercase identifiers or tags (e.g. XML tags and code macros).

    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
    Begriff
    -0.95
     gitara
    -0.93
     julk
    -0.91
     hän
    -0.87
     lää
    -0.87
     tertulis
    -0.87
     jät
    -0.87
    Konto
    -0.87
     läh
    -0.86
    体内
    -0.86
    POSITIVE LOGITS
     power
    1.20
     system
    1.18
     status
    1.13
     pump
    1.09
     temperature
    1.09
     Set
    1.05
     mode
    1.05
     emergency
    1.02
     enable
    1.02
     modes
    1.01
    Activations Density 0.020%

    No Known Activations