© 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. 26976
    Prev
    Next
    INDEX
    Explanations

    awkes, Kalivas, Bremer, Strahl, Walsh

    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
    /libs
    -0.08
    robe
    -0.08
    à¹ģà¸ģ
    -0.08
    家å±ŀ
    -0.08
    褪
    -0.08
    DATES
    -0.07
     representing
    -0.07
    çĶļèĩ³æĺ¯
    -0.07
    代表
    -0.07
     Commerce
    -0.07
    POSITIVE LOGITS
    é«ĺæ½®
    0.10
    缸éĢļ
    0.09
    ncy
    0.09
    isse
    0.09
    itto
    0.08
    esar
    0.08
    å¾®è§Ĥ
    0.08
    elled
    0.08
    nemonic
    0.08
    uxtap
    0.08
    Activations Density 0.215%

    No Known Activations