© 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. 16805
    Prev
    Next
    INDEX
    Explanations

    first, my, their

    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
    è¿ħ
    -0.09
    glomer
    -0.09
    elon
    -0.09
    /rem
    -0.09
    ONY
    -0.09
    _reload
    -0.09
    overlap
    -0.09
    åģļçļĦäºĭ
    -0.09
    ué
    -0.09
    йÑĤ
    -0.09
    POSITIVE LOGITS
    第ä¸Ģ个
    0.25
    é¦ĸ个
    0.24
    çļĦ第ä¸Ģ个
    0.23
    第ä¸Ģéĥ¨
    0.22
    第ä¸Ģ家
    0.20
    第ä¸Ģ次
    0.20
     first
    0.20
    åĪĿãĤģãģ¦
    0.19
    é¦ĸ次
    0.19
    é¦ĸ款
    0.19
    Activations Density 0.150%

    No Known Activations