© 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. 3446
    Prev
    Next
    INDEX
    Explanations

    examine the effect

    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
    çĤ¬
    -0.10
    ä¸ĢéĿ¢
    -0.10
     Morales
    -0.10
    åªļ
    -0.10
    çĶļ
    -0.09
    ç³ł
    -0.09
    èµŃ
    -0.09
     Gerald
    -0.09
    æĪij家
    -0.09
    询
    -0.09
    POSITIVE LOGITS
     whether
    0.11
    æĺ¯åIJ¦æľī
    0.10
    çļĦå½±åĵį
    0.10
     how
    0.10
    æĺ¯åIJ¦åŃĺåľ¨
    0.10
    æľī没æľī
    0.10
     Whether
    0.10
    å½±åĵį
    0.09
    æĺ¯åIJ¦
    0.09
    çĶµæ°Ķ
    0.09
    Activations Density 0.674%

    No Known Activations