© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48992
    Prev
    Next
    INDEX
    Explanations

    numerical method or technique

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    دÙĬØ©
    -0.09
    å®£ä¼łæ´»åĬ¨
    -0.09
     Mahar
    -0.09
    ibox
    -0.09
    æĮº
    -0.08
    ulled
    -0.08
    ideal
    -0.08
    pread
    -0.08
    ÃŃdo
    -0.08
    pte
    -0.08
    POSITIVE LOGITS
    解
    0.13
     ÑĢеÑĪениÑı
    0.10
    solution
    0.10
     solution
    0.10
     solutions
    0.09
    æĸ¹æ³ķ
    0.09
    solve
    0.09
    讥
    0.09
     ÑĢеÑĪение
    0.09
     solve
    0.09
    Activations Density 0.459%

    No Known Activations