© 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. 48863
    Prev
    Next
    INDEX
    Explanations

    far more

    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
    å·®ä¸įå¤ļ
    -0.09
     ÑħоÑĤÑı
    -0.09
    ä¸ĩä¸Ģ
    -0.08
    åįģ足
    -0.08
    ýt
    -0.08
     maybe
    -0.08
    ensed
    -0.08
    æĢ»è§īå¾Ĺ
    -0.08
     ÑĤо
    -0.07
     Collider
    -0.07
    POSITIVE LOGITS
    åĵ®
    0.12
    oles
    0.10
    thest
    0.10
    mente
    0.10
    æĺ¯ä¸Ģéĥ¨
    0.09
    åħ¶ä»ĸåĽ½å®¶
    0.09
    raf
    0.09
    èĩ³äºİ
    0.09
    makt
    0.09
    legen
    0.09
    Activations Density 0.007%

    No Known Activations