© 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. 63625
    Prev
    Next
    INDEX
    Explanations

    allow or disallow

    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
    ailable
    -0.09
    侥幸
    -0.09
    æĿ¿åĿĹ
    -0.09
    ä¸ĵåľº
    -0.09
    èĻı
    -0.09
    Esc
    -0.08
    éºĴ
    -0.08
    èĩĨ
    -0.08
    yte
    -0.08
    estival
    -0.08
    POSITIVE LOGITS
    allows
    0.10
     permits
    0.10
     permitting
    0.10
    åĴĮæĶ¯æĮģ
    0.09
    é¼ĵåĬ±
    0.09
    åľ¨ä¸ĢæĹģ
    0.09
    æĮĩ导æĦıè§ģ
    0.09
    让ä»ĸ们
    0.09
     permit
    0.09
    allow
    0.09
    Activations Density 0.067%

    No Known Activations