© 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. 10684
    Prev
    Next
    INDEX
    Explanations

    public use or access

    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
    æĤ¬æĮĤ
    -0.09
    çek
    -0.09
    æĥ³çĿĢ
    -0.09
    poke
    -0.09
    æĺİçĻ½
    -0.09
    ä½łçľĭ
    -0.09
    æĹ¥åĨħ
    -0.09
    room
    -0.08
     ÑģвÑıзи
    -0.08
    roe
    -0.08
    POSITIVE LOGITS
    osto
    0.10
    ØŃر
    0.08
    zones
    0.08
    erner
    0.08
    _ORIENTATION
    0.08
    ä¼ī
    0.07
    æ¯ıä¸Ģ个人
    0.07
    uren
    0.07
    atar
    0.07
    ibrary
    0.07
    Activations Density 0.007%

    No Known Activations