© 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. 49460
    Prev
    Next
    INDEX
    Explanations

    interested in

    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
    é¦ĸè¦ģ
    -0.09
    çĤİçĥŃ
    -0.09
    äºij端
    -0.08
    iw
    -0.08
    ãģ¨ãģĵãĤį
    -0.08
    满æĦı度
    -0.08
    缺ä¹ı
    -0.08
    ä¸Ģæŀ¶
    -0.08
    æľīæŶ
    -0.08
     iw
    -0.08
    POSITIVE LOGITS
    äºĨå¾Īä¹ħ
    0.10
     ways
    0.10
    æѤäºĭ
    0.10
     how
    0.10
    ä½ĵè´´
    0.09
    ä¹Łä¸įæĥ³
    0.09
    å·²ä¹ħçļĦ
    0.09
    tron
    0.09
    aret
    0.09
    hoe
    0.09
    Activations Density 0.013%

    No Known Activations