© 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. 26231
    Prev
    Next
    INDEX
    Explanations

    alternative component to replacement

    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
    Negative Logits
    ä»ħ代表
    -0.09
    åIJ¸å¼ķæĽ´å¤ļ
    -0.09
    å½ķ
    -0.09
    æĶ¾æĿ¾
    -0.08
    éĽĩ主
    -0.08
    Close
    -0.08
    _RELEASE
    -0.08
    -close
    -0.08
     Reid
    -0.08
    åĩĢèµĦ产
    -0.08
    POSITIVE LOGITS
    转移åĪ°
    0.16
     alternative
    0.15
    åĪ¥ãģ®
    0.15
    转移
    0.13
    åıĹå½±åĵį
    0.12
    alternative
    0.12
     Alternative
    0.12
     elsewhere
    0.12
     alternatives
    0.12
     compensated
    0.12
    Activations Density 0.113%

    No Known Activations