© 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. 25107
    Prev
    Next
    INDEX
    Explanations

    differences and comparisons

    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
     positiva
    -0.84
     herhangi
    -0.84
    ⇄
    -0.83
    AuthProvider
    -0.82
     fizik
    -0.81
     iyi
    -0.81
     it
    -0.81
     kendi
    -0.79
     editar
    -0.78
     |=
    -0.77
    POSITIVE LOGITS
     differences
    2.08
     difference
    1.67
     both
    1.65
     одина
    1.59
     Differences
    1.48
    Differences
    1.47
    の違い
    1.46
    differences
    1.44
    Both
    1.44
     differ
    1.37
    Activations Density 0.182%

    No Known Activations