© 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. 1615
    Prev
    Next
    INDEX
    Explanations

    number or letter followed by closing parenthesis

    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
    æĮĩ示
    -0.09
    mos
    -0.09
    çł£
    -0.09
    said
    -0.09
    èĶ¬
    -0.09
    addon
    -0.09
    åĺīå¹´
    -0.08
    uar
    -0.08
    å¾Ļ
    -0.08
    éĢŁ
    -0.08
    POSITIVE LOGITS
    ä¸Ģæĺ¯
    0.10
    å¼ĭ
    0.09
    èĬ¡
    0.08
    æ´»çĿĢ
    0.08
    åı¯æĥ³
    0.08
     налиÑĩие
    0.08
    ATUS
    0.08
     closures
    0.08
    人们对
    0.08
     yapt
    0.07
    Activations Density 0.104%

    No Known Activations