© 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. 23554
    Prev
    Next
    INDEX
    Explanations

    facility upgrades and technology improvements

    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
    åħ¬å¼Ģåıij
    -0.10
    æ¯Ĥ
    -0.09
    ixel
    -0.09
    æĤ¨çļĦåŃ©åŃIJ
    -0.09
    å°Ĩèĩªå·±çļĦ
    -0.08
    åħ¬çĦ¶
    -0.08
    æķĻä½ł
    -0.08
    éĹ®ä»ĸ
    -0.08
    çŁŃ线
    -0.08
    壳
    -0.08
    POSITIVE LOGITS
    æĶ¹åĸĦ
    0.12
    ä¸Ģæī¹
    0.10
     поÑıви
    0.10
    æĽ´å¤ļçļĦ
    0.09
    å®ĮåĸĦ
    0.09
    æĸ°çļĦ
    0.09
    çŃīä¸Ģæī¹
    0.09
    æĢ¥éľĢ
    0.09
    æĽ´æĸ°
    0.09
     Needed
    0.09
    Activations Density 0.054%

    No Known Activations