© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 71592
    Prev
    Next
    INDEX
    Explanations

    numbers and calculations

    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
     before
    -1.13
     during
    -1.13
     just
    -1.11
     stolet
    -1.09
     only
    -1.07
     people
    -1.06
    水墨
    -1.05
     Because
    -1.03
     over
    -1.00
     gets
    -0.98
    POSITIVE LOGITS
    えっ
    1.38
    すっ
    1.21
     ridiculously
    1.19
     homoto
    1.19
     réaliser
    1.14
     insanely
    1.13
     wonderfully
    1.10
    あー
    1.09
     sagit
    1.09
    あぁ
    1.06
    Activations Density 0.068%

    No Known Activations