© 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. 26264
    Prev
    Next
    INDEX
    Explanations

    positive evaluation

    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
    4
    -0.08
    çĽ¸å¯¹æĿ¥è¯´
    -0.08
    acy
    -0.08
    либо
    -0.08
    åIJĦç§į
    -0.08
    bast
    -0.08
     relative
    -0.08
    inch
    -0.07
    ads
    -0.07
    6
    -0.07
    POSITIVE LOGITS
    羣æŃ£
    0.13
    好çļĦ
    0.12
    èĢħãģ¯
    0.12
    ä¸İåIJ¦
    0.11
    æĥħåĨµä¸ĭ
    0.11
    ä¸Ģ个好çļĦ
    0.11
    æĪIJåĬŁçļĦ
    0.11
    DCALL
    0.10
    羣æŃ£çļĦ
    0.10
    æĪIJçĨŁçļĦ
    0.10
    Activations Density 0.086%

    No Known Activations