© 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. Qwen3-8B
    5. 18-RESID-BATCHTOPK-65K__L0-80
    7. 603
    Prev
    Next
    INDEX
    Explanations

    do it

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-8b-saes/saes_Qwen_Qwen3-8B_batch_top_k/resid_post_layer_18
    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
    broken
    -0.27
    è¯Ŀ
    -0.23
    æĬ±æĢ¨
    -0.22
    viz
    -0.22
     wording
    -0.22
    æĻļä¼ļ
    -0.22
    parer
    -0.22
    burgh
    -0.21
    /from
    -0.21
    åIJĥåĪ°
    -0.21
    POSITIVE LOGITS
    äºĽä»Ģä¹Ī
    0.34
    ä¸Ģå®ļçļĦ
    0.33
    åħ¶ä»ĸ
    0.29
    åħ¶æīĢ
    0.29
     something
    0.29
    æĽ´å¤ļçļĦ
    0.28
    ä¸Ģå®ļ
    0.28
    æĽ´å¤ļ
    0.27
    ä¸ĢåĪĩ
    0.27
     things
    0.27
    Activations Density 0.060%

    No Known Activations