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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38361
    Prev
    Next
    INDEX
    Explanations

    tokens and colon separators

    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
     on
    -1.12
     from
    -1.01
     not
    -0.90
     Very
    -0.89
    hamos
    -0.89
     or
    -0.88
     незавер
    -0.85
     no
    -0.84
     by
    -0.82
    Although
    -0.82
    POSITIVE LOGITS
    captures
    1.05
    📂
    1.04
    ᾷ
    1.02
    GRUPO
    0.99
    škin
    0.98
    yecek
    0.98
     insanely
    0.96
    lasting
    0.95
     exhibición
    0.94
    ranging
    0.93
    Activations Density 0.005%

    No Known Activations