© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36828
    Prev
    Next
    INDEX
    Explanations

    backwoods, backdoor, backwater, backpage, backbench, backlinks, backlog

    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
    arez
    -0.10
    -sided
    -0.10
    ipel
    -0.09
    ashtra
    -0.09
    odem
    -0.09
    辩论
    -0.09
    енÑĤ
    -0.09
    -automatic
    -0.09
    ä¸Ĭå²Ĺ
    -0.09
    DRV
    -0.09
    POSITIVE LOGITS
    stant
    0.09
     ÑģебÑı
    0.08
    迳
    0.08
    çĪµ
    0.08
    轩è¾ķ
    0.08
     naï
    0.08
    æĬĢæľ¯åĴĮ
    0.08
     bol
    0.08
    éģĵ
    0.08
    趾
    0.08
    Activations Density 0.013%

    No Known Activations