© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21334
    Prev
    Next
    INDEX
    Explanations

    used to do something

    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
    çĶ¨æīĭ
    -0.10
    æĪIJ为
    -0.09
    inery
    -0.09
    鸾
    -0.09
    çĶ¨
    -0.08
     вÑĭÑĢажен
    -0.08
     Geneva
    -0.08
    åĩºæīĭ
    -0.08
    æĪIJ为ä¸ĢåIJį
    -0.08
    游泳
    -0.08
    POSITIVE LOGITS
    æģ¯
    0.10
    èģĶ绾
    0.09
    è¡¡éĩı
    0.09
    买车
    0.09
    审è§Ĩ
    0.09
    æĴ¬
    0.09
    éĺ²æŃ¢
    0.09
    æīĵçł´äºĨ
    0.09
    lett
    0.09
    çļ´
    0.09
    Activations Density 0.121%

    No Known Activations