© 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. 60584
    Prev
    Next
    INDEX
    Explanations

    habit, tendency, or affinity for

    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
    åĨ»
    -0.09
    第ä¸Ģ天
    -0.09
    åĨħç½®
    -0.09
    éļIJå½¢
    -0.09
    éĺµå®¹
    -0.08
    ç¼Ń
    -0.08
    æķ°çĻ¾
    -0.08
    èĮŃ
    -0.08
    å½ĵæĻļ
    -0.08
     Kurdistan
    -0.08
    POSITIVE LOGITS
     tendency
    0.11
    åĢ¾
    0.09
    urn
    0.09
     prone
    0.09
    åĢ¾åIJij
    0.09
    åıĺçİ°
    0.08
    ç¢İ
    0.08
     habit
    0.08
    åħĨ
    0.08
    reen
    0.08
    Activations Density 0.028%

    No Known Activations