© 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. 18689
    Prev
    Next
    INDEX
    Explanations

    bread and sandwich items

    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
    çıł
    -0.10
    ä¸įåĪĨ
    -0.09
    eldom
    -0.09
    çµĤ
    -0.09
    çĿ¹
    -0.08
     sed
    -0.08
    çļĵ
    -0.08
    ÑĢак
    -0.08
     tal
    -0.08
     Chair
    -0.08
    POSITIVE LOGITS
    é¹Ħ
    0.10
    çīĩéĿ¢
    0.10
     electr
    0.09
    ä¸įæĪIJ
    0.09
    æĢĢ
    0.09
    é¥
    0.09
    é¦ĸéĥ½
    0.09
     Serif
    0.09
    çģµ
    0.09
    责任åĪ¶
    0.09
    Activations Density 0.039%

    No Known Activations