© 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. 53327
    Prev
    Next
    INDEX
    Explanations

    names following punctuation

    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
    æľī许å¤ļ
    -0.09
    анÑĤ
    -0.09
    iba
    -0.08
    éļĶå£ģ
    -0.08
    destroy
    -0.08
    ä¸į认è¯Ĩ
    -0.08
    çĻ¾å®¶
    -0.08
    atrix
    -0.08
     nbytes
    -0.08
    Dot
    -0.08
    POSITIVE LOGITS
    åħ¼
    0.11
    ém
    0.10
     доб
    0.09
    主åĬ¨æĢ§
    0.09
    æѧè§Ĩ
    0.09
    ç¾ŀ
    0.09
    ä¿Ŀå®Ī
    0.09
    ä½Ĩä¸įéĻIJäºİ
    0.09
    ä½Ĩä¸įéĻIJ
    0.08
    è¾ĺ
    0.08
    Activations Density 0.046%

    No Known Activations