© 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. 56041
    Prev
    Next
    INDEX
    Explanations

    overturned, over, -proof suffixes, met, hit

    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
    backs
    -0.12
    åĻĻ
    -0.11
     Bulg
    -0.10
    åºĵ
    -0.09
     backs
    -0.09
    æݾ
    -0.09
    æĥĭ
    -0.09
    åıŃ
    -0.09
    creativecommons
    -0.09
    Oil
    -0.09
    POSITIVE LOGITS
    åĢ¾
    0.12
    éĩįå¿ĥ
    0.11
     overturn
    0.11
    éĩįéĩį
    0.10
    严èĤĥ
    0.10
    ç¿»
    0.10
    ä¸į平衡
    0.09
     Äijá»ķ
    0.09
    åĤ¾
    0.09
    æĶ¶æĶ¯
    0.09
    Activations Density 0.040%

    No Known Activations