© 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. 27016
    Prev
    Next
    INDEX
    Explanations

    building features and spaces

    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
    æĶ¹å»º
    -0.10
    è£ħä¿®
    -0.09
    çĶŁçĶŁ
    -0.08
    楼å±Ĥ
    -0.08
    åij
    -0.08
    èµ°å»Ĭ
    -0.08
    天èĬ±æĿ¿
    -0.08
     home
    -0.08
    ceil
    -0.08
    æľīæĿ¡ä»¶çļĦ
    -0.08
    POSITIVE LOGITS
    侬
    0.11
    è°
    0.10
    launch
    0.10
    ä¿İ
    0.10
    å¸Ĥéķ¿
    0.09
    _sheet
    0.09
    æĭĪ
    0.09
    稽
    0.09
    鹤
    0.09
    çĶ¨åľ°
    0.09
    Activations Density 0.177%

    No Known Activations