© 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. 26587
    Prev
    Next
    INDEX
    Explanations

    diva, il divo, fashionista

    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
     Jug
    -0.08
     horizon
    -0.08
     hor
    -0.08
    erset
    -0.08
    çľĭ好
    -0.08
    çĿ¦
    -0.08
    modifiers
    -0.07
    hor
    -0.07
    un
    -0.07
    æĢĢ
    -0.07
    POSITIVE LOGITS
    å¨ĺå¨ĺ
    0.12
    们
    0.12
    æ°Ķè´¨
    0.12
    åĢij
    0.10
    强æĤį
    0.10
    ä¸į容
    0.10
    ACING
    0.10
    们çļĦ
    0.10
    ì§ĵ
    0.10
    央ä¼ģ
    0.10
    Activations Density 0.003%

    No Known Activations