© 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. 54728
    Prev
    Next
    INDEX
    Explanations

    Em followed by specific suffixes

    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
    itos
    -0.10
    à¹Į
    -0.09
    аÑĤоÑĢ
    -0.09
    arian
    -0.09
     Institutes
    -0.09
    sWith
    -0.08
     Excell
    -0.08
    ä¸įæĺİ
    -0.08
    ¶
    -0.08
    iddles
    -0.08
    POSITIVE LOGITS
    ding
    0.12
    oyer
    0.11
    minent
    0.10
    agine
    0.10
    ders
    0.09
    ryo
    0.09
    ÑĥÑīеÑģÑĤв
    0.09
    çĦ¶æĺ¯
    0.09
    ponder
    0.09
    otions
    0.09
    Activations Density 0.046%

    No Known Activations