© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 38869
    Prev
    Next
    INDEX
    Explanations

    list delimiters, punctuation, code 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
     in
    -4.56
     as
    -4.03
     to
    -3.89
     with
    -3.89
     at
    -3.64
     and
    -3.63
     by
    -3.50
     on
    -3.39
     from
    -3.36
     for
    -3.31
    POSITIVE LOGITS
     maniere
    3.39
     diriger
    3.34
     délais
    3.28
     enthou
    3.02
    쐈
    3.00
     refroid
    2.97
     bascul
    2.94
     corrom
    2.91
     rafraî
    2.89
     mété
    2.88
    Activations Density 0.190%

    No Known Activations