© 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. 56211
    Prev
    Next
    INDEX
    Explanations

    instead, concluding with a comma or contrast

    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
     correspondent
    -0.11
    mpr
    -0.10
    zing
    -0.09
    èĢĮåIJİ
    -0.09
    bay
    -0.09
     Incorporated
    -0.09
    ä¸įæĺ¯ä¸Ģ个
    -0.09
    ESS
    -0.09
    ä¸įéĶĪ
    -0.08
    æĬ¥åijĬæĺ¾ç¤º
    -0.08
    POSITIVE LOGITS
     вмеÑģÑĤ
    0.10
    æĪª
    0.10
    åıijçĶŁäºĨ
    0.09
    åıijèµ·
    0.09
    UpdateTime
    0.08
    æĦıå¤ĸ
    0.08
     instead
    0.08
    instead
    0.08
    Nos
    0.08
    irk
    0.08
    Activations Density 0.041%

    No Known Activations