© 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. 11057
    Prev
    Next
    INDEX
    Explanations

    long time, a while

    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
    åıįèĢĮ
    -0.08
    麻å°Ĩ
    -0.08
    éĥ¸
    -0.08
    ink
    -0.08
    amin
    -0.08
    ergy
    -0.08
    ify
    -0.07
    áĥª
    -0.07
    oret
    -0.07
    ote
    -0.07
    POSITIVE LOGITS
    æݬ
    0.09
    èĩªçĦ¶æĺ¯
    0.08
    éĺĸ
    0.08
    é¡·
    0.08
    è¹ij
    0.08
     bum
    0.08
    转载
    0.08
    /******/
    0.08
    å°±æĺ¯è¿Ļæł·
    0.08
    èĮ«èĮ«
    0.07
    Activations Density 0.442%

    No Known Activations