© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8359
    Prev
    Next
    INDEX
    Explanations

    "to" followed by verb/noun

    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
    åİļéĩį
    -0.08
    rium
    -0.07
    holm
    -0.07
    éĵ¢
    -0.07
    履约
    -0.07
    fest
    -0.07
     Bauer
    -0.07
    (tb
    -0.07
    urate
    -0.07
    åĽ´çĿĢ
    -0.07
    POSITIVE LOGITS
    æīĭèĩĤ
    0.10
    ounds
    0.09
    annon
    0.09
    serter
    0.09
    uez
    0.09
    _Variable
    0.09
    èĥĮåIJİ
    0.09
    人家
    0.09
    轩
    0.09
    à¹Ģà¸ģร
    0.09
    Activations Density 0.313%

    No Known Activations