© 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. 30185
    Prev
    Next
    INDEX
    Explanations

    website categories

    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
    åı¦ä¸Ģæĸ¹éĿ¢
    -0.12
    åıĪä¸Ģ次
    -0.10
    对éĿ¢
    -0.10
    adows
    -0.10
    aires
    -0.09
    æľĢ
    -0.09
    ä½ľèĢħæľ¬äºº
    -0.09
    æĹ¢æľī
    -0.09
    abis
    -0.09
    æĸ°ä¸Ģ代
    -0.08
    POSITIVE LOGITS
    -than
    0.13
    类似çļĦ
    0.12
     than
    0.12
    world
    0.11
    类似
    0.11
    纷纷
    0.10
    éĻĨç»Ń
    0.10
     besides
    0.10
    than
    0.09
     niż
    0.09
    Activations Density 0.285%

    No Known Activations