© 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. 3554
    Prev
    Next
    INDEX
    Explanations

    comprehension and reasoning

    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
    æĵį
    -0.09
    ogne
    -0.09
    ãģĤãģŁãĤĬ
    -0.09
    带ä¸Ĭ
    -0.09
    æĤ¼
    -0.08
    æĦŁè§¦
    -0.08
    滥çĶ¨
    -0.08
    ulaire
    -0.08
    å²±
    -0.08
    注åĨĮèµĦæľ¬
    -0.08
    POSITIVE LOGITS
    aight
    0.10
    ATAR
    0.09
     Woodward
    0.09
    ä¸įåĩĨ
    0.08
    è¹Ħ
    0.08
     Acc
    0.08
    ä¸ĩäºĭ
    0.08
    odo
    0.08
    æĬĬæı¡
    0.08
    满æĢĢ
    0.08
    Activations Density 0.087%

    No Known Activations