© 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
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37052
    Prev
    Next
    INDEX
    Explanations

    unacceptable statements

    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
    ANCED
    -0.10
    èĥĦ
    -0.09
     LAB
    -0.09
    ué
    -0.09
    å¿Į
    -0.09
    kÄħ
    -0.08
    åİ¢
    -0.08
    象
    -0.08
    æ¦ĨæŀĹ
    -0.08
    Polygon
    -0.08
    POSITIVE LOGITS
    pone
    0.11
    容å¿į
    0.10
    è¿ŁåĪ°
    0.09
    å²Ĺ
    0.09
    irth
    0.09
     anymore
    0.08
    ä»»ä½ķä¸Ģ个
    0.08
    轻微
    0.08
    æľīä»»ä½ķ
    0.08
    éĹ®è´£
    0.08
    Activations Density 0.009%

    No Known Activations