© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29732
    Prev
    Next
    INDEX
    Explanations

    purpose and investigate

    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
     truth
    -0.09
    éĩijéĵ¶
    -0.09
    truth
    -0.09
    è·¯è¿ĩ
    -0.09
    å°±éľĢè¦ģ
    -0.08
    que
    -0.08
    认è¯ĨåĪ°
    -0.08
    etched
    -0.08
    real
    -0.08
     quar
    -0.08
    POSITIVE LOGITS
    ä¸įåIJĪçIJĨ
    0.11
    æĦİ
    0.11
    ä¸į顾
    0.09
    麻çĹ¹
    0.09
    好åĥıæĺ¯
    0.09
     GRE
    0.09
    æľªç»ı
    0.09
     Ñħа
    0.09
    ÑĦан
    0.09
    俨
    0.08
    Activations Density 0.299%

    No Known Activations