© 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. 27663
    Prev
    Next
    INDEX
    Explanations

    Union

    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
    æ´»çĿĢ
    -0.09
    åIJĪä¸Ģ
    -0.08
    ç¬ĥ
    -0.08
    æ½´
    -0.08
    cott
    -0.08
    å¾Īé«ĺçļĦ
    -0.08
    éĻ¢éķ¿
    -0.08
     íķŃ
    -0.08
    roscope
    -0.08
    stuff
    -0.08
    POSITIVE LOGITS
    union
    0.13
    .union
    0.12
     Union
    0.11
     union
    0.10
    Union
    0.10
    OWN
    0.10
    lux
    0.10
    èļģ
    0.09
     unions
    0.09
    estre
    0.09
    Activations Density 0.026%

    No Known Activations