© 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. 19889
    Prev
    Next
    INDEX
    Explanations

    cash and /bash scripts

    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
    åıijçĥ§
    -0.11
    esis
    -0.10
    ä¸Ģ头
    -0.09
    ä¸Ģ段
    -0.09
    orris
    -0.09
    æīĵçł´äºĨ
    -0.09
    hq
    -0.09
    å½ĵä¹ĭ
    -0.08
    ondo
    -0.08
    à¸ļาย
    -0.08
    POSITIVE LOGITS
    足迹
    0.09
    纽带
    0.09
    æµģåIJij
    0.09
    Ctr
    0.09
    ILI
    0.08
    æµģéĩı
    0.08
    uable
    0.08
    宫殿
    0.08
     Walnut
    0.08
    تÙĥ
    0.08
    Activations Density 0.006%

    No Known Activations