© 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. 11332
    Prev
    Next
    INDEX
    Explanations

    auxiliary verbs followed by participles or adjectives

    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
    ç¡®
    -0.09
    人工
    -0.09
    reds
    -0.08
    飧
    -0.08
    辩
    -0.08
    ÑĢон
    -0.08
    à¹ģล
    -0.08
    ä¼ĹæīĢ
    -0.08
    uhl
    -0.08
    _tl
    -0.08
    POSITIVE LOGITS
    éºĵ
    0.11
    åĴĮ个人
    0.10
    arus
    0.09
    çĮľæµĭ
    0.09
    çĿ£
    0.09
    失åĪ©
    0.09
    澡
    0.09
    ivent
    0.09
    åħ¬ç«ĭ
    0.09
    InnerHTML
    0.08
    Activations Density 0.332%

    No Known Activations