© 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. 2375
    Prev
    Next
    INDEX
    Explanations

    algorithms, charts, and flow diagrams

    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
    iterated
    -0.09
    elle
    -0.09
    eland
    -0.09
    iad
    -0.08
    æķ¦
    -0.08
    ertia
    -0.08
    Ace
    -0.08
    ä¸Ńåħ³
    -0.08
    .uc
    -0.08
    (ib
    -0.08
    POSITIVE LOGITS
    åĪĨæĶ¯
    0.11
    æµģç¨ĭ
    0.11
    flow
    0.10
     flow
    0.10
    线索
    0.10
    åħ¨æµģç¨ĭ
    0.10
    FLOW
    0.09
    สร
    0.09
    ä¸ĢæŃ¥æŃ¥
    0.08
    循
    0.08
    Activations Density 0.104%

    No Known Activations