© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67048
    Prev
    Next
    INDEX
    Explanations

    be verb + modifier

    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
     hierfür
    -1.32
    ウェット
    -1.30
     considéré
    -1.29
     utilisé
    -1.27
     этому
    -1.25
    ようで
    -1.21
    撂
    -1.20
    鵙
    -1.20
     représenter
    -1.20
    Kritik
    -1.20
    POSITIVE LOGITS
     with
    1.62
     some
    1.59
     its
    1.48
     had
    1.39
     itself
    1.38
    鹸
    1.35
     from
    1.35
     hvis
    1.34
    чему
    1.34
    цент
    1.33
    Activations Density 0.111%

    No Known Activations