© 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. 84433
    Prev
    Next
    INDEX
    Explanations

    verb ending in -ing

    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
     would
    -1.39
     do
    -1.27
     find
    -1.23
     will
    -1.23
     by
    -1.23
     чтобы
    -1.18
     when
    -1.16
     people
    -1.15
     of
    -1.13
     but
    -1.09
    POSITIVE LOGITS
     the
    1.34
     Não
    1.27
     poils
    1.26
     вдруг
    1.25
     tenues
    1.21
     héros
    1.20
     Trabalho
    1.19
     époux
    1.18
     durant
    1.16
     sinds
    1.15
    Activations Density 0.023%

    No Known Activations