© 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. 1445
    Prev
    Next
    INDEX
    Explanations

    fight of the century

    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
    ities
    -0.18
    åĮĸçļĦ
    -0.10
    æ©IJ
    -0.10
    itÃł
    -0.09
    erator
    -0.09
    íŀĪ
    -0.09
    stry
    -0.09
    èĩ¾
    -0.08
    Ùij
    -0.08
    ized
    -0.08
    POSITIVE LOGITS
    болÑĮ
    0.10
    ress
    0.10
    erset
    0.09
    æĭ¢
    0.09
    rosse
    0.09
    å¾Ĺ好
    0.09
    çĦ¶æĺ¯
    0.09
    amous
    0.09
    unct
    0.09
    OLS
    0.08
    Activations Density 0.019%

    No Known Activations