© 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. 38723
    Prev
    Next
    INDEX
    Explanations

    said in

    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
    angu
    -0.09
    è¿Ļç§įäºĭæĥħ
    -0.09
    usk
    -0.09
    çĽħ
    -0.09
    é¢ī
    -0.09
     Titles
    -0.09
    åħ¨æł¡
    -0.08
    èĸ®
    -0.08
    书记
    -0.08
    æŀī
    -0.08
    POSITIVE LOGITS
    cpy
    0.08
    ishops
    0.08
    èµ·åĪĿ
    0.08
     ÑĢанее
    0.08
    pone
    0.08
    íĴį
    0.08
    两个æľĪ
    0.08
    -ref
    0.07
    cro
    0.07
     Variation
    0.07
    Activations Density 0.018%

    No Known Activations