© 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. 11040
    Prev
    Next
    INDEX
    Explanations

    parts and replacements

    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
     Schultz
    -0.09
    æ§ģ
    -0.09
    éİı
    -0.08
    brid
    -0.08
    主线
    -0.08
    è§īå¾Ĺå¾Ī
    -0.08
    诸侯
    -0.08
    .TIM
    -0.08
    çªĥ
    -0.08
    åįĹçĵľ
    -0.08
    POSITIVE LOGITS
    æĽ¿æį¢
    0.13
    éĥ¨ä»¶
    0.12
    éħį件
    0.12
    è¡¥
    0.11
    åİŁä»¶
    0.11
    è¡¥é½IJ
    0.11
    ä¿®å¤į
    0.11
    replacement
    0.11
    æĤłæĤł
    0.11
    ä¿®çIJĨ
    0.10
    Activations Density 0.083%

    No Known Activations