© 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. 52868
    Prev
    Next
    INDEX
    Explanations

    important to note

    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
    髯
    -0.10
    åºŀ
    -0.10
    ä¸ĢåľĪ
    -0.10
    æ¡Ģ
    -0.09
    .Cmd
    -0.09
    é¡»
    -0.09
    TRAN
    -0.08
    éłĪ
    -0.08
    åIJ¬çĿĢ
    -0.08
    衾
    -0.08
    POSITIVE LOGITS
    éĹ¾
    0.09
    èĨº
    0.09
    elay
    0.09
    etsy
    0.08
    çļĦæĺ¯
    0.08
    è½Ń
    0.08
    æ¯ı个人çļĦ
    0.08
    ate
    0.08
    jit
    0.08
     enf
    0.08
    Activations Density 0.019%

    No Known Activations