© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35864
    Prev
    Next
    INDEX
    Explanations

    subject pronouns and "is"

    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
     réfrig
    -1.20
     überweisen
    -1.17
    ciąż
    -1.17
    bereiche
    -1.16
     gekomen
    -1.16
    chetti
    -1.13
    今でも
    -1.11
     kemeja
    -1.10
    甙
    -1.09
    σκε
    -1.09
    POSITIVE LOGITS
     queijo
    1.23
     later
    1.16
    介绍
    1.13
    Dette
    1.11
    Sziasztok
    1.09
     but
    1.09
     is
    1.08
     ambienti
    1.08
     sereia
    1.05
     habt
    1.05
    Activations Density 0.021%

    No Known Activations