© 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. 8107
    Prev
    Next
    INDEX
    Explanations

    able, edible, ible, medicinal, drinking

    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
    flutter
    -0.10
    enne
    -0.10
     Dust
    -0.09
    обÑĢаз
    -0.08
    orc
    -0.08
    pers
    -0.08
    specs
    -0.08
    å°ıä¼Ļ
    -0.08
    åİŁæłĩé¢ĺ
    -0.08
    oran
    -0.08
    POSITIVE LOGITS
    able
    0.13
    ible
    0.11
    ABLE
    0.11
    æĢ§å¼º
    0.11
    ä¸İåIJ¦
    0.10
    åĴĮæĶ¯æĮģ
    0.09
    åĮĸçļĦ
    0.09
    èī¯å¥½
    0.09
    elta
    0.09
    IBLE
    0.09
    Activations Density 0.020%

    No Known Activations