© 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. 27322
    Prev
    Next
    INDEX
    Explanations

    Roland or Achilles

    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
    worthy
    -0.10
     dÃłng
    -0.10
     ÐŁÐµÑĤÑĢ
    -0.09
    ëģĶ
    -0.09
    ìļĶ
    -0.09
    oral
    -0.09
    æŶèĬĤ
    -0.09
    ãĤĤãģ£ãģ¨
    -0.09
    scape
    -0.09
    ious
    -0.09
    POSITIVE LOGITS
     latter
    0.10
    uted
    0.09
    ewan
    0.08
    åįĴ
    0.08
    wards
    0.08
     largest
    0.08
    åĺī宾
    0.08
    amina
    0.08
    é¦ģ
    0.08
    éļĻ
    0.08
    Activations Density 0.089%

    No Known Activations