© 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. 60820
    Prev
    Next
    INDEX
    Explanations

    unique, special, different

    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
    åĮĢ
    -0.09
    åħĢ
    -0.09
    é¡¿
    -0.08
    åĢŁæѤ
    -0.08
    æĭīå¼Ģ
    -0.08
    stantiate
    -0.08
    礼æĭľ
    -0.08
    鸿
    -0.08
    ILA
    -0.08
     Rath
    -0.08
    POSITIVE LOGITS
    ä¹ĭå¤Ħ
    0.09
    æĸ½è¡Į
    0.09
    boro
    0.09
    UGC
    0.09
     SO
    0.08
    ÐľÐĺ
    0.08
    -Series
    0.08
    éģĹ产
    0.08
    éĢ¼
    0.08
     мн
    0.08
    Activations Density 0.020%

    No Known Activations