© 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. 4993
    Prev
    Next
    INDEX
    Explanations

    parts or portions

    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
    itter
    -0.10
    venience
    -0.10
    åĬ©çIJĨ
    -0.09
    å·²ä¹ħçļĦ
    -0.09
    rov
    -0.09
    榷
    -0.09
    راس
    -0.09
    comed
    -0.09
    raf
    -0.08
     arrow
    -0.08
    POSITIVE LOGITS
    æ¯Ķéĩį
    0.10
    ä¸į失
    0.09
    飵
    0.09
    åĽ½æľīèµĦ产
    0.09
    åŁ
    0.08
    ä¸Ģå®ļæĺ¯
    0.08
    atics
    0.08
    èİ«è¿ĩäºİ
    0.08
    çıĤ
    0.08
    幺
    0.08
    Activations Density 0.244%

    No Known Activations