© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36107
    Prev
    Next
    INDEX
    Explanations

    And /or

    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
    енÑĮ
    -0.09
    槿
    -0.08
    åĺ´éĩĮ
    -0.08
    /gin
    -0.08
    å§ļæĺİ
    -0.08
    inate
    -0.08
     ÐĵоÑĢ
    -0.08
    ote
    -0.08
     hass
    -0.08
    æѤæŶ
    -0.08
    POSITIVE LOGITS
     And
    0.10
    /or
    0.10
    æĪĸèĢħ说
    0.10
    æĪĸå¤ļ
    0.09
    AndView
    0.09
    ä¹ĥèĩ³
    0.08
    AndGet
    0.08
     {}.
    0.08
    æĪĸèĢħåħ¶ä»ĸ
    0.08
    ighton
    0.08
    Activations Density 0.090%

    No Known Activations