© 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. 104406
    Prev
    Next
    INDEX
    Explanations

    programming code or comments

    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
     after
    -2.06
     because
    -1.86
     as
    -1.81
     by
    -1.81
     ostensibly
    -1.80
     undeniably
    -1.77
     with
    -1.75
     from
    -1.72
     get
    -1.70
     epitom
    -1.65
    POSITIVE LOGITS
     astounding
    1.84
     vijf
    1.77
     Dinas
    1.73
     horrid
    1.70
     luscious
    1.68
     unn
    1.67
     quintessential
    1.66
     tumultuous
    1.66
     insanely
    1.65
     .......
    1.63
    Activations Density 0.010%

    No Known Activations