© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9405
    Prev
    Next
    INDEX
    Explanations

    inspiration and reference

    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
    ç½²
    -0.10
    æĪij说
    -0.09
    婵
    -0.09
    éĻ¬
    -0.09
    åĥ®
    -0.09
    å°¤åħ¶
    -0.08
    åŃIJå¼Ł
    -0.08
    æ²®
    -0.08
    毫
    -0.08
    ä¸ĵå±ŀ
    -0.08
    POSITIVE LOGITS
    åıĤèĢĥ
    0.17
     inspiration
    0.16
    åĢŁéī´
    0.15
    åıĤçħ§
    0.15
    çģµæĦŁ
    0.14
     inspir
    0.14
     reference
    0.13
    åıĥèĢĥ
    0.13
     Inspiration
    0.13
    Insp
    0.12
    Activations Density 0.095%

    No Known Activations