© 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. 50253
    Prev
    Next
    INDEX
    Explanations

    Tim names

    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
    ç²¹
    -0.10
     Zus
    -0.09
    leine
    -0.09
    ä»Ĭ天æĪij们
    -0.09
    çŃ
    -0.08
    åĨĪ
    -0.08
    æĤ¦
    -0.08
    éħĨ
    -0.08
     WR
    -0.08
    umpt
    -0.08
    POSITIVE LOGITS
     Burr
    0.10
    å¦ĤæĦı
    0.10
    çªģ
    0.09
    ç½´
    0.09
    acy
    0.08
     Dak
    0.08
     resent
    0.08
    uctive
    0.08
    鼯
    0.08
     kindly
    0.08
    Activations Density 0.023%

    No Known Activations