© 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. 18633
    Prev
    Next
    INDEX
    Explanations

    problem identification

    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
    æĶ¾åѦ
    -0.11
    强åĬ²
    -0.10
    æijĺ
    -0.10
    ä¿Ŀ驾
    -0.09
    æľīåĬ©
    -0.09
    亮çĤ¹
    -0.09
    çļĩåIJİ
    -0.09
    itz
    -0.08
    æIJŃéħį
    -0.08
    è¹ī
    -0.08
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.10
    以人æ°ij为
    0.10
    人æĢ§åĮĸ
    0.09
    èī¯å¿ĥ
    0.09
    æĤĦæĤĦ
    0.09
    为广大
    0.09
    åĸij
    0.09
     customer
    0.09
    å·¥ç¨ĭå¸Ī
    0.09
    ä¸į顾
    0.09
    Activations Density 0.129%

    No Known Activations