© 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. 31563
    Prev
    Next
    INDEX
    Explanations

    disclaimers and licensing

    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
    缸åħ³
    -0.25
    à¹Ģà¸Ĥา
    -0.20
    缸åħ³äººåijĺ
    -0.17
    缸åħ³è§Ħå®ļ
    -0.17
    à¹Ģรา
    -0.17
    缸åħ³æĶ¿çŃĸ
    -0.15
    以ä¸Ĭ
    -0.13
    èµĦéĩij
    -0.12
    缸åħ³ä¿¡æģ¯
    -0.12
    形象
    -0.12
    POSITIVE LOGITS
     sorts
    0.20
    /from
    0.17
    vido
    0.17
    ï¬ģ
    0.13
    0.13
    /about
    0.13
    course
    0.13
    icina
    0.12
     interest
    0.12
    åħ¶å®ŀ
    0.12
    Activations Density 2.138%

    No Known Activations