© 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. 6980
    Prev
    Next
    INDEX
    Explanations

    single most

    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
    å½ĵåľ°çļĦ
    -0.10
    éĢīæĭ
    -0.08
    æīĢåľ¨çļĦ
    -0.08
    åĬłåħ¥äºĨ
    -0.08
    oyo
    -0.08
    nger
    -0.08
    oner
    -0.08
    ä¸įéĢĤåIJĪ
    -0.08
    为æĪij们
    -0.08
     Haut
    -0.07
    POSITIVE LOGITS
    æľĢ
    0.25
    æľĢæľī
    0.23
    æľĢåħ·
    0.22
    æľĢ为
    0.22
    æľĢãĤĤ
    0.22
     most
    0.22
     ê°Ģìŀ¥
    0.22
    æľĢ容æĺĵ
    0.21
     наиболее
    0.21
    æľĢéļ¾
    0.20
    Activations Density 0.129%

    No Known Activations