© 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. 60630
    Prev
    Next
    INDEX
    Explanations

    first inn

    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
    踵
    -0.10
    æ²ī
    -0.10
    åĢĻéĢī人
    -0.09
    å·¥ä½ľå®¤
    -0.09
    髯
    -0.09
    åłĤ
    -0.09
    èijĹ
    -0.08
    èľķ
    -0.08
    让人们
    -0.08
    éĽį
    -0.08
    POSITIVE LOGITS
    reon
    0.09
    shit
    0.09
    quez
    0.08
    orget
    0.08
    forme
    0.08
    еÑħ
    0.08
     anthrop
    0.08
    شر
    0.08
     Sheffield
    0.08
    浦
    0.08
    Activations Density 0.133%

    No Known Activations