© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 8394
    Prev
    Next
    INDEX
    Explanations

    could say what they might do

    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_262k_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
     résulte
    0.43
    ง
    0.42
    れい
    0.40
     resulta
    0.39
     کوډ
    0.39
    b
    0.39
     llamas
    0.39
    䒨
    0.39
     ensue
    0.39
    ју
    0.38
    POSITIVE LOGITS
     могла
    0.66
     rightly
    0.63
     COULD
    0.63
     owes
    0.62
     certainly
    0.62
     rightfully
    0.59
     deserves
    0.58
     могло
    0.58
     deserve
    0.58
    未必
    0.58
    Activations Density 0.008%

    No Known Activations