© 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. 57635
    Prev
    Next
    INDEX
    Explanations

    do the trick

    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
    íĺľ
    -0.09
    èģªæĺİ
    -0.09
    å¿į
    -0.09
    羯
    -0.09
    @update
    -0.08
     пÑĢием
    -0.08
    ãĥªãĥ³ãĤ¯
    -0.08
    å·²ç»ıæľī
    -0.08
    åİ»çľĭ
    -0.08
    trail
    -0.08
    POSITIVE LOGITS
    é¢ĺ主
    0.09
    趾
    0.09
    akter
    0.09
    åĩºè·¯
    0.08
    stan
    0.08
    à¸ĩาม
    0.08
    æĹ¢è¦ģ
    0.08
    åĵī
    0.08
    ATYPE
    0.08
    æı½
    0.08
    Activations Density 0.161%

    No Known Activations