© 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. 2197
    Prev
    Next
    INDEX
    Explanations

    iran nuclear

    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
    èĻļæŀĦ
    -0.10
    åĪĨæµģ
    -0.09
    æĮĩå°ĸ
    -0.09
    èĻĽ
    -0.08
    èģĶéĤ¦
    -0.08
     Companion
    -0.08
    说å®ŀ
    -0.08
    åĢ¬
    -0.08
     ancest
    -0.08
     Maced
    -0.08
    POSITIVE LOGITS
    峪
    0.10
    æĭ¥æĬ¤
    0.10
    æĶ¿æĿĥ
    0.10
    dia
    0.09
    èĨı
    0.09
    çĶ¨åľ°
    0.09
    ipur
    0.09
     dia
    0.09
    ISON
    0.09
    INGTON
    0.08
    Activations Density 0.141%

    No Known Activations