© 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. 62629
    Prev
    Next
    INDEX
    Explanations

    debiasing, debloated, debonded, debauchery, Debtor, debater

    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
    ivity
    -0.11
    à¹Į
    -0.10
    认
    -0.10
    виÑĤ
    -0.09
    æ¯ĵ
    -0.09
    imer
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
    bread
    -0.09
    avanaugh
    -0.08
    Ãłng
    -0.08
    POSITIVE LOGITS
    inging
    0.10
    ewan
    0.09
    ends
    0.09
    çĦ¶æĺ¯
    0.09
    ew
    0.09
    ussy
    0.09
    ä¸įæ¸ħ
    0.09
    ewing
    0.09
    ilm
    0.08
    ../../../
    0.08
    Activations Density 0.020%

    No Known Activations