© 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. 48885
    Prev
    Next
    INDEX
    Explanations

    authors and writers

    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
    0
    -0.10
    åIJĦéĥ¨éŨ
    -0.10
    6
    -0.10
    3
    -0.09
    æĭ
    -0.09
    5
    -0.09
    4
    -0.09
    æĸ¹æĸ¹éĿ¢éĿ¢
    -0.09
    cour
    -0.09
    9
    -0.08
    POSITIVE LOGITS
    pus
    0.11
    æĻļå¹´
    0.10
    åĨĻçļĦ
    0.10
    ç»Ī身
    0.10
    åĨįè§ģ
    0.09
    主è§Ĵ
    0.09
    igg
    0.09
    碧
    0.09
    æķħ乡
    0.08
    ê»ĺ
    0.08
    Activations Density 0.043%

    No Known Activations