© 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. 58336
    Prev
    Next
    INDEX
    Explanations

    must have or must read

    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
    ä¸Ģè§Ĵ
    -0.09
    åįĵ
    -0.09
    mus
    -0.09
    éĤ£åıª
    -0.08
     Gros
    -0.08
    .Obj
    -0.08
     Verde
    -0.08
    æīĢæľīæĿĥ
    -0.08
    deo
    -0.08
    ä¾į
    -0.08
    POSITIVE LOGITS
    åľ¨ä»ĸçļĦ
    0.10
    宫
    0.09
    æŃ¤äºº
    0.09
    æģĤ
    0.08
    ìĤ¬íķŃ
    0.08
    uum
    0.08
    urn
    0.08
     Fur
    0.08
    æĬ«éľ²
    0.08
    以人为
    0.08
    Activations Density 0.020%

    No Known Activations