© 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. 36537
    Prev
    Next
    INDEX
    Explanations

    our community

    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
    åĿŀ
    -0.10
    æĭ¼åij½
    -0.09
    etur
    -0.09
    å¸ı
    -0.09
    çłģ头
    -0.09
    没人
    -0.08
    沦
    -0.08
    (argument
    -0.08
    马åħĭæĢĿ
    -0.08
    泡
    -0.08
    POSITIVE LOGITS
    .ic
    0.09
    ocity
    0.09
     society
    0.08
    ÑĪедÑĪ
    0.08
    ibold
    0.08
    æŀĦ
    0.08
    _ti
    0.08
    обÑīа
    0.08
    icher
    0.08
    çĤ¹å¤ļ
    0.08
    Activations Density 0.070%

    No Known Activations