© 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. 1384
    Prev
    Next
    INDEX
    Explanations

    Protocol Buffers

    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
    æ³¥
    -0.08
     deported
    -0.08
    表çݰ为
    -0.08
    çªĪ
    -0.08
    åĪ·åį¡
    -0.08
    (argument
    -0.08
     lou
    -0.08
    å¹ħ度
    -0.08
    no
    -0.07
     ju
    -0.07
    POSITIVE LOGITS
    ãĥ¼ãĥł
    0.10
    éĥijéĩį
    0.10
    itol
    0.09
    èıĬèĬ±
    0.09
    ĵåIJį
    0.09
    WARN
    0.09
    åĩŃ
    0.09
    æľªç»ı
    0.09
    ç¥ŀç»ıç³»ç»Ł
    0.09
    åĪĨæŀIJå¸Ī
    0.08
    Activations Density 0.011%

    No Known Activations