© 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. 29825
    Prev
    Next
    INDEX
    Explanations

    exact cause or mechanism

    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
    VICE
    -0.09
    èµĦæ·±
    -0.09
    ä¸į失
    -0.09
    ä»İåīį
    -0.09
    usa
    -0.09
    ased
    -0.08
    8
    -0.08
    5
    -0.08
    aise
    -0.08
    å¾Īé«ĺçļĦ
    -0.08
    POSITIVE LOGITS
    ä¸İåIJ¦
    0.15
    å°ļ
    0.13
    å¦Ĥä½ķ
    0.13
    æľªçŁ¥
    0.12
    è¿Ļä¸ĢçĤ¹
    0.12
    è¿ĺä¸į
    0.11
    æ½ľåĬĽ
    0.10
    ä½ķ
    0.10
    ç»ĨèĬĤ
    0.10
    åIJĥä»Ģä¹Ī
    0.10
    Activations Density 0.477%

    No Known Activations