© 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. 19735
    Prev
    Next
    INDEX
    Explanations

    straight ahead or lie down

    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
    èŀºæĹĭ
    -0.10
    ä¸Ģçľ¼
    -0.09
    -mails
    -0.09
    ervas
    -0.09
     tph
    -0.09
    BASE
    -0.09
    èIJ¥ä¸ļé¢Ŀ
    -0.08
    åIJĿ
    -0.08
    ServiceProvider
    -0.08
     elt
    -0.08
    POSITIVE LOGITS
    诱人
    0.09
    åĬłå¿«åıijå±ķ
    0.09
    åıijèĤ²
    0.08
    hart
    0.08
    èij¡
    0.08
     Jerome
    0.08
    Ãį
    0.08
    upal
    0.08
    èĬŃ
    0.08
    æ¤ħ
    0.08
    Activations Density 0.039%

    No Known Activations