© 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. 48060
    Prev
    Next
    INDEX
    Explanations

    extension sites or agent of

    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
     Ñĥме
    -0.10
    å®Ī
    -0.09
    æį
    -0.09
    heel
    -0.09
    çıĻ
    -0.09
    被æĬĵ
    -0.09
    ç©°
    -0.08
    ucking
    -0.08
    oyer
    -0.08
    quet
    -0.08
    POSITIVE LOGITS
    å¼ĢåĬŀ
    0.10
    åĬŀåѦ
    0.10
    庶
    0.09
    ¾ç¤º
    0.09
    对å¤ĸ
    0.09
    é¢ģåıij
    0.09
    轩è¾ķ
    0.09
    å©ī
    0.09
     Juda
    0.08
    åĬŀ
    0.08
    Activations Density 0.120%

    No Known Activations