© 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. 9501
    Prev
    Next
    INDEX
    Explanations

    research methodologies, surveys, studies

    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
     миÑĢов
    -0.10
    éģĹ
    -0.10
    ä¸įæŃ»
    -0.10
    修补
    -0.09
    衬
    -0.09
     ÑĤоÑĢ
    -0.09
    è°¦
    -0.09
    .opensource
    -0.08
    à¸ļรร
    -0.08
    è·ij
    -0.08
    POSITIVE LOGITS
    è°ĥæŁ¥
    0.16
    æ°ijæĦı
    0.15
    è°ĥæŁ¥æĺ¾ç¤º
    0.14
    è°ĥçłĶ
    0.14
    å¿ĥçIJĨåѦ
    0.14
     survey
    0.13
    urvey
    0.13
    survey
    0.13
     surveys
    0.13
    éĹ®åį·
    0.12
    Activations Density 0.175%

    No Known Activations