© 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. 21835
    Prev
    Next
    INDEX
    Explanations

    registering or downloading

    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
    åѦåĪ°
    -0.10
     SUMMARY
    -0.09
    æºIJ头
    -0.08
    ascar
    -0.08
    追éļı
    -0.08
    çIJ®
    -0.08
    çĻ¾åĪĨ
    -0.08
    辩论
    -0.08
    amu
    -0.08
    çĥģ
    -0.08
    POSITIVE LOGITS
    åĽŀå¿Ĩ
    0.10
    åľ¨æѤ
    0.09
    æĹłéľĢ
    0.09
    æѤå¤Ħ
    0.09
    æ¼ı
    0.09
    å¿ĺ
    0.09
    ä¹Ķ
    0.09
    é¢Ĩ
    0.09
    çĶ©
    0.09
    зд
    0.09
    Activations Density 0.017%

    No Known Activations