© 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. 32918
    Prev
    Next
    INDEX
    Explanations

    how to approach

    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
    侬
    -0.09
    ä¹¾åĿ¤
    -0.09
    ç°ĭ
    -0.09
    astro
    -0.08
    jer
    -0.08
    ç§ģç«ĭ
    -0.08
    éĹ¾
    -0.08
    /std
    -0.08
    çĭ·
    -0.08
    èŀ³
    -0.08
    POSITIVE LOGITS
    sWith
    0.10
    atto
    0.10
    ../
    0.10
    able
    0.10
    baugh
    0.09
    san
    0.09
    ego
    0.09
    à¸ģารà¸Ĺำ
    0.09
    à¹Ħหà¸Ļ
    0.09
    geist
    0.09
    Activations Density 0.035%

    No Known Activations