© 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. 47695
    Prev
    Next
    INDEX
    Explanations

    take

    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
    èĨ»
    -0.09
    æ¹ĸåĮº
    -0.09
    ç¤ģ
    -0.09
     taking
    -0.09
    Tho
    -0.08
    åĭĥåĭĥ
    -0.08
    åIJ¸çĥŁ
    -0.08
    obra
    -0.08
    èĥ¸åīį
    -0.08
    Ñij
    -0.08
    POSITIVE LOGITS
     advantage
    0.15
     seriously
    0.13
     ÑĥÑĩаÑģÑĤие
    0.10
     advant
    0.10
     ×ij×Ĺש×ij×ķף
    0.10
     plunge
    0.10
    adv
    0.10
     charge
    0.09
     Advantage
    0.09
     Seriously
    0.09
    Activations Density 0.251%

    No Known Activations