© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37757
    Prev
    Next
    INDEX
    Explanations

    as being [adjective phrase]

    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
    èĪģ
    -0.10
    åĴĮ个人
    -0.10
    ugin
    -0.10
    禺
    -0.09
    odule
    -0.09
    åİŁæľīçļĦ
    -0.09
    utorials
    -0.09
    ç«ĭæ³ķ
    -0.09
    æIJ¬å®¶
    -0.09
    éĤ£å°±
    -0.09
    POSITIVE LOGITS
    åѦåİĨ
    0.09
    odie
    0.09
    æĺ¯æľĢ好çļĦ
    0.08
    æľīåĪ©äºİ
    0.08
    æIJŃ
    0.08
    ler
    0.08
     lub
    0.08
    pot
    0.08
     Nicholson
    0.08
    kie
    0.08
    Activations Density 0.122%

    No Known Activations