© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 47860
    Prev
    Next
    INDEX
    Explanations

    Sally Bedina, rust async json full

    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
     Gol
    0.39
     ออนไลน์
    0.39
    ꯡ
    0.37
    িন্দা
    0.36
    居然
    0.36
    ショップ
    0.35
    Gol
    0.35
     mat
    0.35
     Corinthians
    0.34
     Бай
    0.34
    POSITIVE LOGITS
    udah
    0.39
    śle
    0.38
    ாள
    0.38
    JsonResponse
    0.37
    ++,
    0.37
    wym
    0.36
     Macro
    0.36
    னையும்
    0.36
    ilir
    0.36
     Structures
    0.35
    Activations Density 0.005%

    No Known Activations