© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37578
    Prev
    Next
    INDEX
    Explanations

    Mule, white, Bose, Jules

    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
    å®°
    -0.09
    é¢ī
    -0.09
    ä½Ĩä¸įéĻIJ
    -0.08
    stell
    -0.08
     dozen
    -0.08
    ening
    -0.08
    åħĪ导
    -0.08
    æĮ«
    -0.08
    ết
    -0.08
    oda
    -0.08
    POSITIVE LOGITS
    çķļ
    0.09
    insula
    0.09
    Jar
    0.09
    è¿Ŀ纪
    0.09
     trimest
    0.08
    åŃº
    0.08
    è¹Ģ
    0.08
     POSS
    0.08
    å®īåİ¿
    0.08
    å©Ĩ
    0.08
    Activations Density 0.003%

    No Known Activations