© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20134
    Prev
    Next
    INDEX
    Explanations

    pristine, untouched, frontier, undeveloped

    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
    æIJ¦
    -0.10
    讲究
    -0.09
    å¿Ĩ
    -0.09
     *__
    -0.09
    roc
    -0.09
    ÙĪÙĨØ©
    -0.08
    渺
    -0.08
    macro
    -0.08
     Beg
    -0.08
    Andrew
    -0.08
    POSITIVE LOGITS
     untouched
    0.13
    å¼Ģåıij
    0.11
    åħ¬åħ±èµĦæºIJ
    0.11
    çł£
    0.11
    çł´åĿı
    0.10
    æĮĸæİĺ
    0.09
    å°ļæľª
    0.09
    åıijçİ°
    0.09
    æĹħ游èµĦæºIJ
    0.09
    éĸĭçĻ¼
    0.09
    Activations Density 0.035%

    No Known Activations