© 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. 38538
    Prev
    Next
    INDEX
    Explanations

    types of buildings

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

    The neuron fires strongest on words naming or describing man‐made dwellings and structures (e.g. huts, houses, barns, cabins, cottages).

    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
    шні
    -1.08
    ednes
    -1.05
    人數
    -1.03
    躋
    -1.02
    GRAPHY
    -1.02
     квіт
    -0.99
    ших
    -0.98
    зывают
    -0.97
    orescence
    -0.96
    irmanship
    -0.96
    POSITIVE LOGITS
     construida
    1.42
     buildings
    1.37
     silueta
    1.27
    搭建
    1.23
     built
    1.23
     structures
    1.21
     structure
    1.20
     construction
    1.19
     housing
    1.15
     rústica
    1.15
    Activations Density 0.068%

    No Known Activations