© 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. 6787
    Prev
    Next
    INDEX
    Explanations

    process, variables, and instances

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

    The neuron strongly activates on code‐style identifiers—especially Java package/import components and BPM/workflow domain terms (e.g. “org.eclipse…bpm.flowable”, “process”, “state”).

    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
    sects
    -0.85
     referees
    -0.82
    شيخ
    -0.73
     پزشک
    -0.72
    tapete
    -0.71
    intermediate
    -0.70
     vídeo
    -0.69
    getDevice
    -0.69
    Libr
    -0.69
    辅助
    -0.68
    POSITIVE LOGITS
     variables
    0.91
     instance
    0.88
    variables
    0.86
    Starter
    0.84
    InstanceState
    0.82
    实例
    0.81
    Variables
    0.79
    INST
    0.75
    unfinished
    0.75
     Instances
    0.75
    Activations Density 0.010%

    No Known Activations