© 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. 26272
    Prev
    Next
    INDEX
    Explanations

    end or final states

    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
    emer
    -0.09
     ÑĢаÑģпÑĢоÑģÑĤÑĢан
    -0.08
    inx
    -0.08
    ops
    -0.08
    æIJºå¸¦
    -0.08
     ops
    -0.08
    åĬłæ·±
    -0.08
     cy
    -0.08
    jis
    -0.08
    ConnectionFactory
    -0.08
    POSITIVE LOGITS
    æĪIJåĵģ
    0.22
     final
    0.19
    æľĢç»Ī
    0.18
    FINAL
    0.16
     finished
    0.16
    final
    0.15
    -final
    0.15
    Final
    0.14
     Final
    0.14
     FINAL
    0.14
    Activations Density 0.112%

    No Known Activations