© 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. 17410
    Prev
    Next
    INDEX
    Explanations

    institute of

    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
    asure
    -0.08
    ison
    -0.08
    å±¥
    -0.08
    æŶèĬĤ
    -0.08
    禺
    -0.08
    åį«åģ¥å§Ķ
    -0.08
    heet
    -0.08
    اغ
    -0.08
    outh
    -0.08
     Lexington
    -0.07
    POSITIVE LOGITS
    ãģ¾ãģ¾
    0.11
    anni
    0.11
    è¿Ļåıªæĺ¯
    0.10
    ãģĹãģ¾ãģĨ
    0.10
    ãģ¦ãģĦ
    0.10
    ëį°
    0.09
    eam
    0.09
    颦
    0.09
    ytut
    0.09
     ngôn
    0.09
    Activations Density 0.029%

    No Known Activations