© 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. 40384
    Prev
    Next
    INDEX
    Explanations

    cutting, slicing, splitting

    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
    骥
    -0.09
    åįķ纯
    -0.08
    uard
    -0.08
    æŁIJç§įç¨ĭ度
    -0.08
    oulos
    -0.08
    opa
    -0.08
     Jacqu
    -0.08
    graphics
    -0.08
    rama
    -0.08
    å®Ī
    -0.08
    POSITIVE LOGITS
    åĪĩãĤĬ
    0.12
    ä¸ĢåĪĢ
    0.11
    åĪĩæĸŃ
    0.10
    åĪĩ
    0.10
    夫
    0.10
    åĪĩåī²
    0.10
     cuts
    0.10
     slicing
    0.09
    åī²
    0.09
     scoop
    0.09
    Activations Density 0.065%

    No Known Activations