© 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. 3658
    Prev
    Next
    INDEX
    Explanations

    Arizona State University

    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
    antry
    -0.10
    æ§Į
    -0.09
    缯
    -0.09
    åºķçĽĺ
    -0.09
    åĩŃ
    -0.09
    Chance
    -0.09
    .bmp
    -0.08
    éĻĦåĬłåĢ¼
    -0.08
    æ»
    -0.08
    endi
    -0.08
    POSITIVE LOGITS
    plevel
    0.11
    lei
    0.09
     apost
    0.09
     Ñĩел
    0.09
     Barber
    0.09
    èĪĢ
    0.09
    èĢĢçľ¼
    0.09
    çIJĩ
    0.09
    igel
    0.08
     commend
    0.08
    Activations Density 0.001%

    No Known Activations