© 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. 19093
    Prev
    Next
    INDEX
    Explanations

    partnering with

    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
    æŀī
    -0.09
    ær
    -0.09
    çĶ¥
    -0.09
    çľ¼éĩĮ
    -0.08
    æľ¬äºº
    -0.08
    anz
    -0.08
    åįļè§Ī
    -0.08
    è´¢ç»ı
    -0.08
    åĪ¨
    -0.08
    sehen
    -0.08
    POSITIVE LOGITS
     Trap
    0.09
    ULO
    0.08
    Harness
    0.08
    æŶå°ļ
    0.08
     Ree
    0.08
    .Formatter
    0.08
    èĢħçļĦ
    0.08
    Trap
    0.08
    人éĥ½
    0.08
    éĺĻ
    0.08
    Activations Density 0.006%

    No Known Activations