© 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. 27412
    Prev
    Next
    INDEX
    Explanations

    when

    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
    ä¹İ
    -0.11
    æīĢåľ¨
    -0.10
    amba
    -0.10
    åıĪ被
    -0.09
    æ¸´æľĽ
    -0.09
    ForObject
    -0.09
    ä¾Ŀçħ§
    -0.08
    缨
    -0.08
    ieux
    -0.08
    opleft
    -0.08
    POSITIVE LOGITS
     when
    0.17
    å½ĵæĪij们
    0.16
    when
    0.16
    å½ĵä½ł
    0.14
     khi
    0.13
     cuando
    0.13
    å½ĵæĪij
    0.13
    å½ĵä»ĸ
    0.13
    	when
    0.13
     gdy
    0.12
    Activations Density 0.534%

    No Known Activations