© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 50043
    Prev
    Next
    INDEX
    Explanations

    bel/belief/belt/belgium

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_65k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    信頼
    0.79
     trust
    0.79
     trusts
    0.79
    Reli
    0.77
     믿
    0.73
     reliability
    0.69
     percaya
    0.69
     Trust
    0.68
    trust
    0.68
    Trust
    0.67
    POSITIVE LOGITS
    риа
    0.49
     тут
    0.44
    рад
    0.42
    rea
    0.42
     quedando
    0.41
    <0xA5>
    0.40
     கேட்க
    0.39
    ening
    0.39
    <0xAB>
    0.38
     Ком
    0.38
    Activations Density 0.016%

    No Known Activations