© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37509
    Prev
    Next
    INDEX
    Explanations

    always exists

    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
    佯
    -0.10
    æĹıèĩªæ²»åİ¿
    -0.09
     Wax
    -0.08
    uard
    -0.08
    çĤī
    -0.08
    lia
    -0.08
    ä»ħ代表
    -0.08
    å¾Īéĩįè¦ģçļĦ
    -0.08
    æľīå¿ħè¦ģ
    -0.08
    pc
    -0.08
    POSITIVE LOGITS
    æĢ»æľī
    0.13
    æŁIJ个
    0.11
    æĪĸå¤ļæĪĸå°ij
    0.11
    æŁIJä¸Ģ
    0.10
    ething
    0.10
    éĺ´éĺ³
    0.10
    someone
    0.10
     either
    0.10
    æºIJæºIJä¸įæĸŃ
    0.10
    æŁIJ
    0.10
    Activations Density 0.028%

    No Known Activations