© 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. 24963
    Prev
    Next
    INDEX
    Explanations

    transitioning from digression

    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
     them
    -1.05
    *
    -1.01
     "
    -0.99
    ‘
    -0.98
     as
    -0.97
    ’
    -0.95
     Y
    -0.94
    '
    -0.94
     X
    -0.93
     Applicability
    -0.93
    POSITIVE LOGITS
    Anyway
    1.95
     Anyway
    1.80
    Anyways
    1.58
    さて
    1.52
     Anyways
    1.49
     anyway
    1.47
    Итак
    1.45
    Anyhow
    1.38
     digress
    1.30
    Törté
    1.23
    Activations Density 0.050%

    No Known Activations