© 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. 22-GEMMASCOPE-RES-131K
    7. 27851
    Prev
    Next
    INDEX
    Explanations

    alarms, power outages, complaints

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

    The neuron activates on concrete, physical or sensory language—mentions of body parts, tangible objects, and actions involving touch or interaction with the environment.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     얻
    -0.91
     dava
    -0.87
    retes
    -0.86
    setPos
    -0.85
     給
    -0.85
     naturen
    -0.84
    essayer
    -0.82
    预览
    -0.82
    Вариант
    -0.81
    gidas
    -0.81
    POSITIVE LOGITS
     because
    0.97
     lack
    0.96
    ANNON
    0.95
     toallas
    0.92
     itinéraires
    0.92
    Reilly
    0.91
    こい
    0.90
     waited
    0.90
     klachten
    0.88
     system
    0.88
    Activations Density 0.058%

    No Known Activations