© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73888
    Prev
    Next
    INDEX
    Explanations

    highway disruptions and closures

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

    The neuron is picking out mentions of transportation infrastructure—roads, highways, streets, tunnels, 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_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
     galinha
    -1.33
    lateinit
    -1.30
    /}.
    -1.14
     leão
    -1.13
    lında
    -1.13
    ènements
    -1.12
     coelho
    -1.11
    imagenes
    -1.10
     quoque
    -1.10
     сахара
    -1.05
    POSITIVE LOGITS
    FFECT
    1.54
    確實
    1.49
     Thursday
    1.49
    鎳
    1.48
    під
    1.33
    
    1.33
     تريد
    1.30
     yesterday
    1.30
     informó
    1.24
    三個
    1.23
    Activations Density 0.012%

    No Known Activations