© 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. 57146
    Prev
    Next
    INDEX
    Explanations

    study resources

    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
    åģ¥
    -0.09
    bull
    -0.09
    Lint
    -0.08
    elope
    -0.08
    Spread
    -0.08
    lest
    -0.08
    çİ«
    -0.08
    çģ¾
    -0.08
    erie
    -0.08
     Union
    -0.08
    POSITIVE LOGITS
    åħ¬åı¸å°Ĩ
    0.10
    æīĢæıIJä¾Ľ
    0.09
    æīĢæıIJä¾ĽçļĦ
    0.09
    æıIJä¾ĽçļĦ
    0.09
    é¢Ĩ导çıŃåŃIJ
    0.09
    åħ¬åı¸åľ¨
    0.09
    èij£äºĭä¼ļ
    0.09
    åģļå¾Ĺ
    0.09
    缼è¡Į
    0.09
    æ©Łæ§ĭ
    0.09
    Activations Density 0.015%

    No Known Activations