© 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. 50825
    Prev
    Next
    INDEX
    Explanations

    just made, found, or released

    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
    æľīå¤ļ
    -0.10
    TRY
    -0.09
    inine
    -0.08
    ä¹¾
    -0.08
    (base
    -0.08
    ÂŃi
    -0.08
    å¥ĭ
    -0.08
     Past
    -0.08
     Flesh
    -0.08
    å¹Į
    -0.08
    POSITIVE LOGITS
    äºĨä¸Ģ个
    0.11
    æĮł
    0.11
    èĢ½è¯¯
    0.10
    æĦıä¹īä¸ĬçļĦ
    0.10
    éĻį
    0.10
    çĤ¹äºĨ
    0.09
    ç¼ģ
    0.09
    å®ŀåľ¨æĺ¯
    0.09
     something
    0.09
    è¹Ī
    0.09
    Activations Density 0.078%

    No Known Activations