© 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. 1431
    Prev
    Next
    INDEX
    Explanations

    forced to write

    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
    å¸ĮæľĽèĥ½å¤Ł
    -0.09
    าว
    -0.08
    称
    -0.08
    å°ĨæĿ¥
    -0.08
    itor
    -0.08
     Clips
    -0.08
    åĩij
    -0.07
    太大
    -0.07
    åįĥæĸ¹
    -0.07
    ç®ĹäºĨ
    -0.07
    POSITIVE LOGITS
    ä¸įå¾Ĺä¸į
    0.12
    çıįæĥľ
    0.11
    éĥ¨èIJ½
    0.11
    强迫
    0.10
    被迫
    0.10
     forced
    0.10
    éĶ¤
    0.10
    forced
    0.10
    åĨįä¸Ģ次
    0.10
    ä¸įç»ıæĦı
    0.10
    Activations Density 0.065%

    No Known Activations