© 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. 55814
    Prev
    Next
    INDEX
    Explanations

    can followed by verbs

    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
    åĽŀäºĭ
    -0.10
    娩
    -0.09
    éļIJç§ģ
    -0.09
    éŃģ
    -0.09
    é¢ī
    -0.09
    æĭ¿äºĨ
    -0.09
    åĬ¨èį¡
    -0.09
    éļ¾åħ³
    -0.09
    vana
    -0.08
    åĭĭ
    -0.08
    POSITIVE LOGITS
    åģļä¸ĢäºĽ
    0.10
    éºĿ
    0.10
     Hil
    0.09
    èĩªæķij
    0.09
    修身
    0.09
    wash
    0.09
    度
    0.09
    ä¸Ģå®ļçļĦ
    0.09
     Ahead
    0.09
     try
    0.08
    Activations Density 0.023%

    No Known Activations