© 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. 21896
    Prev
    Next
    INDEX
    Explanations

    pronoun followed by need

    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±ĭåŃIJ
    -0.09
    奢ä¾Īåĵģ
    -0.09
    æĭ·
    -0.09
    uptools
    -0.09
    æĥłåıĬ
    -0.08
    泡泡
    -0.08
    åIJĪä¸Ģ
    -0.08
    боÑĤ
    -0.08
    chal
    -0.08
    奢
    -0.08
    POSITIVE LOGITS
     wer
    0.10
    jac
    0.09
    èĭŁ
    0.09
    å¤ļå¹´æĿ¥
    0.09
    CUS
    0.08
    å®ĥ们
    0.08
    Wer
    0.08
    驸
    0.08
    ecs
    0.08
    æīĢéľĢçļĦ
    0.08
    Activations Density 0.023%

    No Known Activations