© 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. 35354
    Prev
    Next
    INDEX
    Explanations

    mean, possible, help, useful

    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
     ??
    -0.11
    æīĵçĿĢ
    -0.09
     (?,
    -0.09
    æŃ£æĺ¯
    -0.08
    æĸ°çļĦä¸Ģ
    -0.08
    æŃ£åĽłä¸º
    -0.08
     ?????
    -0.08
     ???
    -0.08
    ured
    -0.08
     (?)
    -0.08
    POSITIVE LOGITS
    åIJĹ
    0.28
    äºĨåIJĹ
    0.27
    åĹİ
    0.24
    ãģ§ãģĻãģĭ
    0.22
    好åIJĹ
    0.21
    ãģ§ãģĹãĤĩãģĨãģĭ
    0.21
    éļ¾éģĵ
    0.21
    Õŀ
    0.20
    ãģ®ãģ§ãģĹãĤĩãģĨãģĭ
    0.17
    ØŁ
    0.17
    Activations Density 0.338%

    No Known Activations