© 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. 38887
    Prev
    Next
    INDEX
    Explanations

    spent [number] [time unit]

    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
    èĤª
    -0.09
    çī¹èī²
    -0.09
    åĬ¨äºº
    -0.09
    /from
    -0.09
    éĩįä¸Ńä¹ĭéĩį
    -0.08
    çħ
    -0.08
    ãģ¾ãģķ
    -0.08
    ä¸ĭ次
    -0.08
    æĸ¯çī¹
    -0.08
     ÄijÃŃch
    -0.08
    POSITIVE LOGITS
    è¿Ļ段æŶéĹ´
    0.10
    ãģĿãģĵãģ§
    0.09
    为èĩªå·±
    0.09
    大éĥ¨åĪĨ
    0.09
    å¾Īéķ¿æŶéĹ´
    0.09
    ties
    0.08
     Shirley
    0.08
    elters
    0.08
    ustom
    0.08
    PRI
    0.08
    Activations Density 0.031%

    No Known Activations