© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21657
    Prev
    Next
    INDEX
    Explanations

    not able to or unlikely

    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
     brav
    -0.09
    丼
    -0.09
    åİŁåĪĻä¸Ĭ
    -0.09
    LD
    -0.08
     LD
    -0.08
    èµĽåIJİ
    -0.08
    æĸ¹ãģ«
    -0.08
    俨
    -0.08
    交åıĭ
    -0.08
    æĬ¥éĢģ
    -0.08
    POSITIVE LOGITS
    unless
    0.14
    ä¹Łä¸įåı¯èĥ½
    0.14
    éĻ¤éĿŀ
    0.13
    çŁŃæľŁåĨħ
    0.13
     unless
    0.13
    æĺ¯ä¸įåı¯èĥ½
    0.12
    ä»įå°Ĩ
    0.12
     Unless
    0.12
    ä¸įåı¯èĥ½
    0.11
     unlikely
    0.11
    Activations Density 0.067%

    No Known Activations