© 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. 22620
    Prev
    Next
    INDEX
    Explanations

    billing and collections

    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
    å®Ĺ
    -0.09
    .rd
    -0.09
     Pur
    -0.09
    åIJ¦
    -0.08
    ENO
    -0.08
    ismatic
    -0.08
    横åIJij
    -0.08
    çļĦ带é¢Ĩ
    -0.08
     Franken
    -0.08
    æ¯Ĵ
    -0.08
    POSITIVE LOGITS
    æĶ¶åĽŀ
    0.17
    åĽŀæĶ¶
    0.14
    åĤ¬
    0.13
    æĶ¶æ¬¾
    0.13
    åºĶæĶ¶
    0.12
    æĶ¶åıĸ
    0.11
    追
    0.11
    æĶ¶å½ķ
    0.10
    æĮ½åĽŀ
    0.10
    lost
    0.10
    Activations Density 0.036%

    No Known Activations