© 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. 10611
    Prev
    Next
    INDEX
    Explanations

    changes reflected

    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
    åºĨ幸
    -0.09
     Äijá»īnh
    -0.09
    使人
    -0.09
    ä¸ĵ项èµĦéĩij
    -0.09
    æ®ĩ
    -0.09
     Rhe
    -0.09
    handleRequest
    -0.09
     continuation
    -0.09
    æĪĸå¤ļ
    -0.09
    çı¥
    -0.09
    POSITIVE LOGITS
     changes
    0.13
    çĶŁæķĪ
    0.12
    changes
    0.11
    åıĺåĮĸ
    0.11
     Changes
    0.11
    _changes
    0.10
    Changes
    0.10
     chances
    0.10
     изменениÑı
    0.10
    æĽ´æĶ¹
    0.10
    Activations Density 0.339%

    No Known Activations