© 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. 10637
    Prev
    Next
    INDEX
    Explanations

    liberation and emancipation

    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
    ÙĬرة
    -0.09
    ook
    -0.08
    alled
    -0.08
    elves
    -0.08
    ç¼Ģ
    -0.08
    代价
    -0.08
    soever
    -0.08
    avig
    -0.08
    ness
    -0.08
    inta
    -0.08
    POSITIVE LOGITS
     khá»ıi
    0.13
    ä¸İåıijå±ķ
    0.12
    åĴĮåıijå±ķ
    0.11
    ä¸įäºĨ
    0.11
    åIJİçļĦ
    0.10
    ively
    0.10
    ë°
    0.10
    /close
    0.10
    iator
    0.09
    解æĶ¾
    0.09
    Activations Density 0.011%

    No Known Activations