© 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. 35582
    Prev
    Next
    INDEX
    Explanations

    simplified management

    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
    çIJī
    -0.11
    å¨ĥå¨ĥ
    -0.10
    ê·¼
    -0.09
    æĹłæīĢè°ĵ
    -0.09
    åĬĪ
    -0.09
    DMI
    -0.09
    éŃĶçİĭ
    -0.08
    ç½ijåıĭè¯Ħ论
    -0.08
    ASURE
    -0.08
    åıĸæļĸ
    -0.08
    POSITIVE LOGITS
    èħ¾
    0.15
    ä¸ĵå¿ĥ
    0.12
    ä¸ĵ注äºİ
    0.12
    騰
    0.12
    纯æ´ģ
    0.12
     pur
    0.11
    Focus
    0.11
    æĽ´å®¹æĺĵ
    0.11
     focus
    0.11
    ä¸ĵ注
    0.10
    Activations Density 0.216%

    No Known Activations