© 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. 33433
    Prev
    Next
    INDEX
    Explanations

    first big well full good meaningful formal

    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
    utra
    -0.10
    łĢ
    -0.09
    ria
    -0.09
    å°ķ
    -0.09
    许åı¯
    -0.09
    æīķ
    -0.09
    æĴIJ
    -0.08
    WillAppear
    -0.08
    è²ī
    -0.08
    èĻº
    -0.08
    POSITIVE LOGITS
     thing
    0.11
     Thing
    0.09
    æĬŁ
    0.08
    æĪijèĥ½
    0.08
    ç¡
    0.08
    full
    0.08
    çݯèĬĤ
    0.08
    æ»ĭåij³
    0.08
    溢
    0.08
    1
    0.08
    Activations Density 0.165%

    No Known Activations