© 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. 39250
    Prev
    Next
    INDEX
    Explanations

    Stevenson

    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
    åĥļ
    -0.10
    æIJĶ
    -0.09
    âĻ¢
    -0.09
    обоÑĢоÑĤ
    -0.09
    çĭİ
    -0.08
    allel
    -0.08
     ÑĤÑĭÑģÑı
    -0.08
    åĭĺ
    -0.08
    zens
    -0.08
    èµ·æĿ¥
    -0.08
    POSITIVE LOGITS
    积
    0.10
     Clarkson
    0.09
    大åıij
    0.09
    енÑĮ
    0.09
    çIJĨç§ij
    0.09
    nob
    0.09
    lor
    0.09
    tube
    0.08
    讵
    0.08
    æĥļ
    0.08
    Activations Density 0.001%

    No Known Activations