© 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. 20201
    Prev
    Next
    INDEX
    Explanations

    increase threefold or 4-fold

    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
    ä¹łæĥ¯
    -0.10
    ä¸ĭæĿ¥
    -0.10
    SWG
    -0.09
    .portal
    -0.09
    æĽ²
    -0.08
    uur
    -0.08
    _NT
    -0.08
    æĺ¶
    -0.08
    ä¸Ĭä¸ĭ
    -0.08
    çĩĥçħ¤
    -0.08
    POSITIVE LOGITS
    æĬ«
    0.10
    azon
    0.09
     Hung
    0.09
    ãģªãģĬ
    0.09
    æīĵçł´
    0.09
    çļĦ人æĿ¥è¯´
    0.08
    orget
    0.08
     Thin
    0.08
    å¸ĥå±Ģ
    0.08
    æīĶ
    0.08
    Activations Density 0.231%

    No Known Activations