© 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. 35693
    Prev
    Next
    INDEX
    Explanations

    lord

    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
    umin
    -0.11
    æĤį
    -0.10
    éĤ¨
    -0.10
    åĺı
    -0.10
    oba
    -0.09
    èĬĿ麻
    -0.09
    线索
    -0.09
     comb
    -0.09
    -times
    -0.09
    ä»·
    -0.09
    POSITIVE LOGITS
    iales
    0.11
    oload
    0.09
    IOD
    0.08
     تعاÙĦÙī
    0.08
    ниÑĨ
    0.08
    lify
    0.08
    åķ¼
    0.08
     vid
    0.08
    \Contracts
    0.08
    ìĬµ
    0.08
    Activations Density 0.015%

    No Known Activations