© 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. 11090
    Prev
    Next
    INDEX
    Explanations

    python executables and versions

    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
    æĭ³å¤´
    -0.10
    çļĦéĴ±
    -0.10
    ä¸įæĢķ
    -0.09
    ä»ħ次äºİ
    -0.09
    é«®
    -0.09
    该项
    -0.09
    ä¸ĵéĹ¨ä¸º
    -0.08
    æķ°åįĥ
    -0.08
    åIJİæĤĶ
    -0.08
    ä¸Ģç¬ij
    -0.08
    POSITIVE LOGITS
    entially
    0.11
     Uint
    0.09
    idual
    0.09
    Uint
    0.09
    ément
    0.08
    liner
    0.08
    æij¸
    0.08
    ential
    0.08
    Ub
    0.08
    eward
    0.08
    Activations Density 0.006%

    No Known Activations