© 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. 2553
    Prev
    Next
    INDEX
    Explanations

    front, patch, auto, R, WE, Fab, N, Google

    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
     воÑĤ
    -0.11
    oug
    -0.11
    æĪijè§īå¾Ĺ
    -0.09
    antas
    -0.09
    nton
    -0.08
    æĥŃ
    -0.08
    arring
    -0.08
    ä¹Ŀé¾Ļ
    -0.08
    ims
    -0.08
     admissions
    -0.08
    POSITIVE LOGITS
    ihan
    0.10
    æĬ¤èĤ¤
    0.09
    ç»ĦåĽ¾
    0.09
    æºIJæºIJä¸įæĸŃ
    0.08
    说åĩºæĿ¥
    0.08
    åºĸ
    0.08
    èĶ
    0.08
    arged
    0.08
    éĿ¢åŃIJ
    0.08
    ä¸īæĿ¡
    0.08
    Activations Density 0.138%

    No Known Activations