© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37840
    Prev
    Next
    INDEX
    Explanations

    together with

    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
    ä¸Ģéģį
    -0.10
     Bos
    -0.09
     tons
    -0.08
    ur
    -0.08
    ä¸Ģ头
    -0.08
    erie
    -0.08
    urd
    -0.08
    PEC
    -0.08
    ...
    -0.08
     deb
    -0.08
    POSITIVE LOGITS
    大家ä¸Ģèµ·
    0.10
    åıĸåĨ³
    0.10
    é¢Ĩ
    0.10
    æĪij们ä¸Ģèµ·
    0.09
    婺
    0.09
    razione
    0.09
    æŁ¬
    0.09
    æݸ
    0.09
    -sama
    0.08
     sá»ķ
    0.08
    Activations Density 0.012%

    No Known Activations