© 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. 47771
    Prev
    Next
    INDEX
    Explanations

    would recommend or would guess

    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
     bro
    -0.09
    åŀ®
    -0.08
    hoe
    -0.08
    æ¯Ķè¾ĥå¤ļ
    -0.08
    æݨåĩºäºĨ
    -0.08
    ió
    -0.08
    agraph
    -0.08
     наÑĤÑĥÑĢалÑĮн
    -0.08
    aret
    -0.07
    _HI
    -0.07
    POSITIVE LOGITS
    ä¸įæĥľ
    0.11
    毫ä¸įçĬ¹è±«
    0.11
    亲æīĭ
    0.10
    çİ¢
    0.10
    ainless
    0.10
    æ¿Ł
    0.10
    绪
    0.10
    åİ»æİī
    0.09
    é¡ĺæĦı
    0.09
    encers
    0.09
    Activations Density 0.023%

    No Known Activations