© 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. 38244
    Prev
    Next
    INDEX
    Explanations

    at least one

    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
    äºķ
    -0.10
    èĩ³å°ij
    -0.10
    è¿Ħä»Ĭ为æŃ¢
    -0.09
    æĥ¹
    -0.08
    perator
    -0.08
    Busy
    -0.08
     Alma
    -0.08
    éķ¿è¿ľ
    -0.08
    ervlet
    -0.08
    æŃ£åĽłä¸º
    -0.08
    POSITIVE LOGITS
     jaki
    0.09
    éĤ£åıª
    0.09
    chie
    0.09
    à¹ģห
    0.08
    ä¸Ģ次æĢ§
    0.08
     få
    0.08
    ä¸Ģåıªæīĭ
    0.08
    zu
    0.08
    ly
    0.08
    ä¸Ģå¹´
    0.08
    Activations Density 0.028%

    No Known Activations