© 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. 60685
    Prev
    Next
    INDEX
    Explanations

    without doing specified action

    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
     Okay
    -0.08
    è¢ħ
    -0.08
    çŁ¥éģĵäºĨ
    -0.08
    æĹłè§Ĩ
    -0.08
    rate
    -0.08
    ctions
    -0.07
     already
    -0.07
    ction
    -0.07
    rates
    -0.07
    ç£Ĭ
    -0.07
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.11
    á»ĩm
    0.10
    æĺµ
    0.10
    大æłij
    0.09
    éĵį
    0.09
    .prof
    0.09
    èĢĥè¯ģ
    0.09
    .har
    0.09
    NECT
    0.09
    äºĮåŃĹ
    0.09
    Activations Density 0.456%

    No Known Activations