© 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. 62631
    Prev
    Next
    INDEX
    Explanations

    Affiliated, affinity, firm, affair, affront

    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
    ë°¤
    -0.11
     xét
    -0.10
     Highlander
    -0.09
    åŀĽ
    -0.09
    ditor
    -0.09
    .utc
    -0.09
    æ»ĵ
    -0.09
    ä»ĸ认为
    -0.09
    infinity
    -0.09
    产ä¸ļæī¶è´«
    -0.09
    POSITIVE LOGITS
    iliated
    0.11
    ordable
    0.11
    ł
    0.09
     aff
    0.09
    ancy
    0.09
    teen
    0.09
    ication
    0.09
    ichage
    0.09
    ulary
    0.09
    ¹
    0.09
    Activations Density 0.041%

    No Known Activations