© 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. 10-GEMMASCOPE-RES-131K
    4. 119654
    Prev
    Next
    INDEX
    Explanations

    specific words following noun-like tokens

    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_10/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
     trám
    -2.23
    tedy
    -2.23
     tedy
    -2.08
     lámparas
    -2.03
     trhu
    -2.03
     cráneo
    -2.02
     estómago
    -2.02
     aguj
    -1.99
     moverse
    -1.96
     ozdob
    -1.95
    POSITIVE LOGITS
     will
    2.59
     this
    2.41
     of
    2.34
     even
    2.09
     if
    2.06
     in
    2.05
     it
    2.03
    並且
    1.95
    !
    1.95
     from
    1.88
    Activations Density 0.042%

    No Known Activations