© 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. 1086
    Prev
    Next
    INDEX
    Explanations

    bag features

    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
    çĤ¹çĩĥ
    -0.11
    ä¼IJ
    -0.10
    æ±°
    -0.09
    æī«
    -0.09
    ynchronize
    -0.08
    ä½ĥ
    -0.08
     Coch
    -0.08
    anova
    -0.08
     hobby
    -0.08
    éĵ§
    -0.08
    POSITIVE LOGITS
    è¿Ľåĩº
    0.12
    ä¿Ŀ管
    0.10
     Bags
    0.10
    è¢ĭåŃIJ
    0.09
    æĶ¶çº³
    0.09
    Bag
    0.09
    æ©IJ
    0.09
    æķŀå¼Ģ
    0.09
    è¡Ķæİ¥
    0.09
    æĵĺ
    0.09
    Activations Density 0.096%

    No Known Activations