© 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. 63965
    Prev
    Next
    INDEX
    Explanations

    replace or upgrade parts

    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
    Negative Logits
    ç¨ĭ度
    -0.13
    æīĵåĩ»
    -0.10
    imat
    -0.10
    æIJĶ
    -0.10
    巨人
    -0.09
    严åİīæīĵåĩ»
    -0.09
    åıįå¤į
    -0.09
    åĩĢåĢ¼
    -0.08
    èĮĥ
    -0.08
    深度
    -0.08
    POSITIVE LOGITS
    æĽ¿æį¢
    0.18
    æĽ´æį¢
    0.18
     replace
    0.13
    è¡¥
    0.12
     replacement
    0.12
     replaced
    0.11
    æī©åħħ
    0.11
    replace
    0.11
     upgrade
    0.11
     Replace
    0.11
    Activations Density 0.143%

    No Known Activations