© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52545
    Prev
    Next
    INDEX
    Explanations

    slowing down

    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
    æģļ
    -0.11
    ÑģÑĤаÑĤ
    -0.09
    æ¼Ĥ亮
    -0.09
     возд
    -0.09
    éĿ¢åħ·
    -0.09
    èµ°å»Ĭ
    -0.09
     ná»iji
    -0.09
    ilen
    -0.09
    缼å¼Ģ
    -0.08
    .opend
    -0.08
    POSITIVE LOGITS
    èĦļæŃ¥
    0.13
    ä¸ĭä¾Ĩ
    0.12
    ä¸ĭæĿ¥
    0.12
    æĶ¾ç¼ĵ
    0.11
    lessly
    0.11
     down
    0.10
     pace
    0.10
    -down
    0.09
    ä¸ĭæĿ¥çļĦ
    0.09
    éļ½
    0.09
    Activations Density 0.027%

    No Known Activations