© 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. 54417
    Prev
    Next
    INDEX
    Explanations

    nothing has been done

    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
     Consum
    -0.09
     asc
    -0.08
    第äºĮ次
    -0.08
    åĽłæŀľ
    -0.08
    åĩĨ
    -0.08
    _blk
    -0.08
    mel
    -0.08
    æīĵåħ¥
    -0.08
    çīĪæĿĥå½Ĵ
    -0.08
    andbox
    -0.08
    POSITIVE LOGITS
    è¯ļæĦı
    0.12
    å®ŀéĻħè¡ĮåĬ¨
    0.11
    羣å¿ĥ
    0.11
    åģľçķĻåľ¨
    0.11
    èIJ½å®ŀ
    0.11
    åĬĽåº¦
    0.11
    æݪæĸ½
    0.11
    举æݪ
    0.11
    åĪ°ä½į
    0.11
    ä¸įåĪ°ä½į
    0.11
    Activations Density 0.054%

    No Known Activations