© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8204
    Prev
    Next
    INDEX
    Explanations

    work for or well

    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
    ä¸Ģéĥ¨åĪĨ
    -0.10
    /from
    -0.10
    åłĥ
    -0.10
    åŀĤ
    -0.09
    burgh
    -0.09
     gamm
    -0.09
    æ¸ħè¿ľ
    -0.08
    coh
    -0.08
    ting
    -0.08
     conquest
    -0.08
    POSITIVE LOGITS
     далÑĮÑĪе
    0.10
    uate
    0.09
    ä¸ĭåİ»
    0.09
    /is
    0.09
    ä¹ĭåľ°
    0.08
    çļĦæĹ¥
    0.08
    eced
    0.08
    /do
    0.08
    ument
    0.08
     vite
    0.08
    Activations Density 0.065%

    No Known Activations