© 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. 1873
    Prev
    Next
    INDEX
    Explanations

    illustrated by

    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
    inel
    -0.09
     Almighty
    -0.09
    Ĥæķ°
    -0.08
    kel
    -0.08
    oral
    -0.08
    chair
    -0.08
    åĩłå¤©
    -0.08
     Nag
    -0.08
    -0.08
    污
    -0.08
    POSITIVE LOGITS
    lık
    0.09
    æݧèĤ¡éĽĨåĽ¢
    0.09
    ׼×Ķ
    0.09
    èĢħçļĦ
    0.09
    érie
    0.09
    degree
    0.09
     giú
    0.08
    讲究
    0.08
    æĪijè®°å¾Ĺ
    0.08
    -su
    0.08
    Activations Density 0.013%

    No Known Activations