© 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. 4666
    Prev
    Next
    INDEX
    Explanations

    at the same time

    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
    ä¸Ģæĺ¯
    -0.10
     ÑģобÑģÑĤвенно
    -0.09
    quo
    -0.09
    åĩ¼
    -0.09
     пÑĢежде
    -0.08
    æĽ°
    -0.08
    then
    -0.08
     ÑĤогда
    -0.08
     next
    -0.08
    ä¸ĢæŶ
    -0.08
    POSITIVE LOGITS
    -sex
    0.11
    หมà¸Ķ
    0.11
    èĤ©è´Ł
    0.10
    -sama
    0.10
    sb
    0.10
    ATUS
    0.10
    aket
    0.09
    èĨĪ
    0.09
    aneously
    0.09
    -runner
    0.09
    Activations Density 0.015%

    No Known Activations