© 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. 18815
    Prev
    Next
    INDEX
    Explanations

    emerald or jade objects

    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
    ility
    -0.10
     Harrison
    -0.09
    ÑĥÑĢ
    -0.09
    ihan
    -0.09
    临
    -0.08
    inator
    -0.08
    ä»ĵ
    -0.08
    erge
    -0.08
    åĦ¡
    -0.08
     sola
    -0.08
    POSITIVE LOGITS
    leaf
    0.11
    çĦ¶æĺ¯
    0.11
    itus
    0.10
    æĪij以为
    0.10
    semb
    0.09
    漪
    0.09
    ä½łä¸į
    0.09
    æīĴ
    0.09
    baum
    0.09
    tery
    0.09
    Activations Density 0.002%

    No Known Activations