© 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. 18964
    Prev
    Next
    INDEX
    Explanations

    once you have

    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
    夺åıĸ
    -0.09
    她æĺ¯
    -0.09
    æīĢæıIJä¾ĽçļĦ
    -0.08
    ä¸īå±Ĥ
    -0.08
    å¦Ļ
    -0.08
    sid
    -0.08
    çłĶ究æĬ¥åijĬ
    -0.08
    çĶŁäº§çļĦ
    -0.08
    çļĦ人çĶŁ
    -0.08
    人çĶŁ
    -0.08
    POSITIVE LOGITS
    å¥Ħ
    0.11
    æľīäºĨ
    0.11
    大èĩ´
    0.11
    éħįæĸĻ
    0.10
    令人
    0.10
    åĪĿæŃ¥
    0.10
     cục
    0.09
    大家
    0.09
    æĹħéĢĶ
    0.09
    æķķ
    0.09
    Activations Density 0.041%

    No Known Activations