© 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. 52624
    Prev
    Next
    INDEX
    Explanations

    protected or resistant

    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
    姬
    -0.09
    士
    -0.09
    æĿ¥çľĭçľĭ
    -0.09
    èIJ¥éĢł
    -0.09
    çĨŁæĤī
    -0.09
    æī®æ¼Ķ
    -0.09
    æľªèĥ½
    -0.09
    被è§Ĩ为
    -0.08
    åĭŁ
    -0.08
    æĬĦ
    -0.08
    POSITIVE LOGITS
    -proof
    0.11
    -resistant
    0.11
     resistant
    0.10
    -safe
    0.10
     fool
    0.10
    æĹłå¿§
    0.10
    ä¸Ģéģĵ
    0.10
    proof
    0.09
    æĬµæĬĹ
    0.09
    æĮł
    0.09
    Activations Density 0.130%

    No Known Activations