© 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. 63140
    Prev
    Next
    INDEX
    Explanations

    boxing and boxers

    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
    OfType
    -0.10
    åħ®
    -0.10
    ãĥ´ãĤ¡
    -0.09
    elry
    -0.08
    fulness
    -0.08
    abilities
    -0.08
    æĸ¼
    -0.08
    çĬģ
    -0.08
    и
    -0.08
    æŃ£æĸĩ
    -0.08
    POSITIVE LOGITS
    éĹ®éĹ®
    0.10
    ï¸
    0.10
    è¢Ĥ
    0.10
    æĭ¯
    0.10
    åľºéĿ¢
    0.09
    stars
    0.09
    åı²ä¸Ĭ
    0.09
    èŀĤ
    0.09
    æĭ³
    0.09
    éĴĬ
    0.08
    Activations Density 0.011%

    No Known Activations