© 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. 62722
    Prev
    Next
    INDEX
    Explanations

    awards and distinctions

    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
    å¼ķãģį
    -0.10
    çĤ½
    -0.09
    å¥ĩ
    -0.09
    ä¸įäºĨ
    -0.09
    ocks
    -0.09
    éªĪ
    -0.09
    å·½
    -0.09
    延å®ī
    -0.08
    å·´åŁº
    -0.08
    aving
    -0.08
    POSITIVE LOGITS
    æIJĵ
    0.09
    éĢ¢
    0.08
    usan
    0.08
    oky
    0.08
    kw
    0.08
    æ³ķå¾ĭè§Ħå®ļ
    0.08
    æĪķ
    0.08
    Gate
    0.07
    æķķ
    0.07
     Dish
    0.07
    Activations Density 0.013%

    No Known Activations