© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29688
    Prev
    Next
    INDEX
    Explanations

    Chinese characters

    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
    DSA
    -0.11
    avit
    -0.09
    ç¼ł
    -0.09
    tlement
    -0.08
    Scoped
    -0.08
    åĩłåįĥ
    -0.08
     коÑĤоÑĢ
    -0.08
    ighth
    -0.08
    .drawable
    -0.08
     Soda
    -0.08
    POSITIVE LOGITS
    shed
    0.09
    çŃĭ
    0.09
    åĪĿæģĭ
    0.09
    æĭ¯æķij
    0.08
    ä¸ĢåĽ¢
    0.08
    è¦ģæĬĬ
    0.08
    æŀģéĢŁ
    0.08
    åĵģç±»
    0.08
    æľīå¿ħè¦ģ
    0.08
    ç¥ĸ
    0.08
    Activations Density 0.030%

    No Known Activations