© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7216
    Prev
    Next
    INDEX
    Explanations

    clip or clippers

    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
    Not in Any Lists

    No Comments

    Negative Logits
    棰
    -0.10
    æĭį
    -0.10
    azzo
    -0.09
     slab
    -0.09
    vos
    -0.09
    éĥijéĩį
    -0.09
    绣
    -0.09
    sla
    -0.09
    (training
    -0.09
    autop
    -0.09
    POSITIVE LOGITS
    amb
    0.08
    ja
    0.08
    å¼ı
    0.08
    å±IJ
    0.08
    åĢ¦
    0.08
    3
    0.08
     onto
    0.07
    åıįè¿ĩæĿ¥
    0.07
    åĿļæĮģä¸įæĩĪ
    0.07
     res
    0.07
    Activations Density 0.023%

    No Known Activations