© 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. 6019
    Prev
    Next
    INDEX
    Explanations

    middle / mid

    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
    æĪĸå¤ļæĪĸå°ij
    -0.11
     /*#__
    -0.10
    è¦ģ好好
    -0.10
    anst
    -0.09
    è¿Ķ
    -0.09
    è¿Ļä¸ĢåĪ»
    -0.09
    åį°è®°
    -0.09
    èł²
    -0.09
    å¤įæŁ¥
    -0.09
    çĶĺ
    -0.09
    POSITIVE LOGITS
    .middle
    0.12
     mid
    0.11
    (mid
    0.10
    mid
    0.10
    æīĵæĸŃ
    0.10
    ä¸ŃéĢĶ
    0.09
    ä¸ŃæĸŃ
    0.09
     middle
    0.09
     ongoing
    0.09
    ä¸ŃæľŁ
    0.09
    Activations Density 0.243%

    No Known Activations