© 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. 51044
    Prev
    Next
    INDEX
    Explanations

    chair and cigar

    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
    é»Ķ
    -0.10
    abras
    -0.10
    渼
    -0.09
    纯æ´ģ
    -0.09
    verbs
    -0.09
    å¿Ļ
    -0.08
    åĬŀ
    -0.08
    害ç¾ŀ
    -0.08
    èĤļåŃIJ
    -0.08
     Griffin
    -0.08
    POSITIVE LOGITS
    æ¤ħåŃIJ
    0.19
    æ¤ħ
    0.15
     chairs
    0.14
     chair
    0.13
    座æ¤ħ
    0.13
    æ²Ļåıij
    0.13
    ä¸ĢæĬĬ
    0.11
     Chair
    0.10
    ä¸ĢåĪ»
    0.10
    åĿIJåľ¨
    0.10
    Activations Density 0.050%

    No Known Activations