© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127792
    Prev
    Next
    INDEX
    Explanations

    causal or temporal connectors

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Negative Logits
     says
    -1.27
    Menurut
    -1.12
    !
    -1.08
    申し上げます
    -1.05
    =[]
    -1.03
    でしたね
    -1.02
     statunitense
    -1.01
    +')
    -1.01
     stated
    -1.00
    According
    -1.00
    POSITIVE LOGITS
     pihaknya
    1.91
    recently
    1.43
     «
    1.37
     “
    1.30
     "
    1.29
     his
    1.25
     '"
    1.24
     recente
    1.15
     ± 
    1.13
    wasn
    1.12
    Activations Density 0.032%

    No Known Activations