© 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. 51821
    Prev
    Next
    INDEX
    Explanations

    explain why

    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
    éĩįç»Ħ
    -0.09
     Ñĥзн
    -0.09
    ä¸Ģå¼ł
    -0.09
    åIJ¬å¾Ĺ
    -0.09
    onald
    -0.08
    å¼Ľ
    -0.08
    ãģ¿ãģŁãģĦãģª
    -0.08
    åIJ¬è§ģ
    -0.08
    说çĿĢ
    -0.08
    AJ
    -0.08
    POSITIVE LOGITS
    解éĩĬ
    0.28
    解éĩĭ
    0.22
     explanation
    0.21
     explanations
    0.20
     explain
    0.18
     explaining
    0.17
     Explanation
    0.16
     explan
    0.16
     explained
    0.16
    Explanation
    0.15
    Activations Density 0.185%

    No Known Activations