© 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. 10992
    Prev
    Next
    INDEX
    Explanations

    when, given, would, needs

    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
    éĻ¤æѤ
    -0.09
    é£Łæ¬²
    -0.08
    ç¨įå¾®
    -0.08
    ounce
    -0.08
    åĩĭ
    -0.08
     FOX
    -0.08
    åı¸
    -0.08
    _cfg
    -0.08
    UDGE
    -0.08
    ï¼
    -0.08
    POSITIVE LOGITS
    äºĨåIJ§
    0.11
    äºĨåįĬ天
    0.10
    å¬
    0.10
    楼主
    0.10
    æĺ¯ä¸įåı¯èĥ½
    0.09
    äºĨèµ·æĿ¥
    0.09
    äºĨåĩºæĿ¥
    0.09
     Claus
    0.09
    äºĨåIJĹ
    0.09
    _emb
    0.09
    Activations Density 4.095%

    No Known Activations