© 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. 1297
    Prev
    Next
    INDEX
    Explanations

    lists of logs or problems

    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
    sWith
    -0.10
    ãģ°ãģĭãĤĬ
    -0.09
    rig
    -0.09
     themselves
    -0.08
    ÑĮÑı
    -0.08
    aires
    -0.08
     Rope
    -0.08
    ä¸Ģ个个
    -0.08
    们çļĦ
    -0.08
    çĩ®
    -0.08
    POSITIVE LOGITS
    迹
    0.09
    å®ĥçļĦ
    0.09
    ymbol
    0.09
    éģĹæ¼ı
    0.09
    æ¼ı
    0.09
    ä½łè§īå¾Ĺ
    0.09
    ourced
    0.09
    olve
    0.08
    æĽ²çº¿
    0.08
    cape
    0.08
    Activations Density 0.477%

    No Known Activations