© 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. 61655
    Prev
    Next
    INDEX
    Explanations

    lots of

    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
    éĢĴ
    -0.09
    wor
    -0.09
    èĮ¹
    -0.08
    UX
    -0.08
    Context
    -0.08
    UTOR
    -0.08
    è·Łä¸įä¸Ĭ
    -0.08
    è¨Ģ
    -0.08
     Brooks
    -0.07
     sworn
    -0.07
    POSITIVE LOGITS
    亮çĤ¹
    0.11
    åı¯ä¾Ľ
    0.10
    æľīçĽĬ
    0.10
    ç§įç±»
    0.10
    æijĨ
    0.10
    loff
    0.10
    ä¹IJ趣
    0.10
     useful
    0.10
    å°ĸ
    0.09
    éĢī项
    0.09
    Activations Density 0.050%

    No Known Activations