© 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. 26565
    Prev
    Next
    INDEX
    Explanations

    equinox and equator

    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
    åħ±
    -0.10
     беÑģп
    -0.10
    çĥ«
    -0.09
    åħļåĴĮ
    -0.09
    è¡Įä¸ļä¸Ń
    -0.09
     undecided
    -0.09
    pany
    -0.08
    幸
    -0.08
    аÑģÑģив
    -0.08
    ãĢħ
    -0.08
    POSITIVE LOGITS
    æļij
    0.12
    ç¥Ń
    0.10
    åħ¥ä¾µ
    0.10
    æŶèĬĤ
    0.09
    æģ¶
    0.09
    顶
    0.09
    èĮ§
    0.09
    ored
    0.09
    èĢħçļĦ
    0.09
    è°Ĵ
    0.09
    Activations Density 0.025%

    No Known Activations