© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 20633
    Prev
    Next
    INDEX
    Explanations

    reversal, reversing, change, direction

    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
    å½ĵåīįä½įç½®
    -0.11
    æĹłæĦı
    -0.09
    çľ¶
    -0.09
    èĥ¼
    -0.09
    erner
    -0.09
    é¨
    -0.09
    é¢Ŀ
    -0.09
    roman
    -0.08
    åıªåī©
    -0.08
    涨å¹ħ
    -0.08
    POSITIVE LOGITS
    转åIJij
    0.16
    转身
    0.16
    转æĬĺ
    0.15
    åıį转
    0.15
    .Change
    0.15
    éĢĨ转
    0.14
    åĽŀ头
    0.14
     reversal
    0.14
    转åıĺ
    0.13
    æĶ¹åıĺ
    0.13
    Activations Density 0.064%

    No Known Activations