© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7321
    Prev
    Next
    INDEX
    Explanations

    family relations: aunt, uncle, niece, nephew

    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
    æĸĻ
    -0.11
    rc
    -0.10
    è·Łå¥¹
    -0.10
     Mush
    -0.09
    anga
    -0.09
    èĢģåħ¬
    -0.09
    缣
    -0.09
     глаз
    -0.09
    èĢģçĻ¾å§ĵ
    -0.09
    æĹ¥æĬ¥è®°èĢħ
    -0.09
    POSITIVE LOGITS
    们çļĦ
    0.11
    gram
    0.11
    stry
    0.09
    ties
    0.08
    们
    0.08
    coh
    0.08
    ä¸Ģåı°
    0.08
    à¸Ĭา
    0.08
     borderline
    0.08
    egin
    0.08
    Activations Density 0.015%

    No Known Activations