© 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. 32873
    Prev
    Next
    INDEX
    Explanations

    instead, rather, but

    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
    Not in Any Lists

    No Comments

    Negative Logits
    liśmy
    -0.96
    فق
    -0.91
     besonders
    -0.85
    だけは
    -0.83
     Tiere
    -0.81
     similaires
    -0.81
    哪怕
    -0.80
    naph
    -0.80
     semoga
    -0.79
     hoffen
    -0.77
    POSITIVE LOGITS
     instead
    1.99
     вместо
    1.54
    instead
    1.50
     rather
    1.48
     zamiast
    1.40
     but
    1.38
     invece
    1.38
    而不是
    1.36
    Instead
    1.34
     chứ
    1.28
    Activations Density 0.014%

    No Known Activations