© 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. 1898
    Prev
    Next
    INDEX
    Explanations

    You are followed by 'are', 'will', 'Save', 'scored'

    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
    éĵĽ
    -0.09
    æĦļ
    -0.09
    æīª
    -0.09
    人æ°ijæĶ¿åºľ
    -0.09
    HECK
    -0.09
    neck
    -0.08
    ses
    -0.08
    -election
    -0.08
    å±ģ
    -0.08
    ustral
    -0.08
    POSITIVE LOGITS
    oretical
    0.12
    orem
    0.10
    itat
    0.09
    itre
    0.09
    er
    0.09
    ograd
    0.09
    åĢ¼å¾Ĺåħ³æ³¨
    0.08
    æľªæĿ¥çļĦ
    0.08
    ä¸ªå·¥ä½ľ
    0.08
    ify
    0.08
    Activations Density 0.071%

    No Known Activations