© 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. 56937
    Prev
    Next
    INDEX
    Explanations

    cured, set, solidified

    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
    æĭĸ
    -0.09
    mor
    -0.09
     estr
    -0.09
    icens
    -0.08
    çĦ¦çĤ¹
    -0.08
    æĤłéĹ²
    -0.08
    RIA
    -0.08
     Razor
    -0.08
    ç¼ĵåĨ²
    -0.08
    åķľ
    -0.08
    POSITIVE LOGITS
    åĽºåĮĸ
    0.13
    æĪIJåŀĭ
    0.11
    硬åĮĸ
    0.11
    å·©åĽº
    0.11
    çī¢åĽº
    0.10
    åİĭå®ŀ
    0.09
    èĪ¶
    0.09
    orks
    0.09
     tack
    0.09
    åĬłåĽº
    0.09
    Activations Density 0.155%

    No Known Activations