© 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. 39724
    Prev
    Next
    INDEX
    Explanations

    contrasting conjunctions like whereas

    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
    vided
    -0.09
    泪
    -0.08
    icter
    -0.08
    款
    -0.08
     пÑĢедоÑģÑĤав
    -0.08
    èµĸ以
    -0.08
    åıĬåħ¶ä»ĸ
    -0.08
    è®¤çľŁèIJ½å®ŀ
    -0.08
    matic
    -0.08
    jest
    -0.08
    POSITIVE LOGITS
    è¿Ļå°±æĺ¯
    0.10
    è¿Ļä¸ĢçĤ¹
    0.09
    éĩı产
    0.09
    æ¯Ķå¦Ĥ说
    0.09
    oon
    0.09
    è¿ĻçĤ¹
    0.09
     Compatible
    0.09
    ego
    0.08
    oir
    0.08
    æ¯Ķå¦Ĥ
    0.08
    Activations Density 0.346%

    No Known Activations