© 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. 57864
    Prev
    Next
    INDEX
    Explanations

    stylized

    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
    ining
    -0.09
    voy
    -0.09
    dux
    -0.09
    iqu
    -0.09
    ido
    -0.08
    ajar
    -0.08
    opoly
    -0.08
    èĤ²
    -0.08
    	TokenName
    -0.08
    iqué
    -0.08
    POSITIVE LOGITS
    è°ģèĥ½
    0.10
    åŃ
    0.10
    0.09
     Kaiser
    0.09
    ä¸Ģå®ļçļĦ
    0.09
    é¢Ĩ导人
    0.09
    çľĩ
    0.09
    å±İ
    0.09
    çIJ¢ç£¨
    0.08
    åķľ
    0.08
    Activations Density 0.004%

    No Known Activations