© 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. 50995
    Prev
    Next
    INDEX
    Explanations

    did / done / doing it

    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
    èĬŃ
    -0.08
    kich
    -0.08
    è¡Ĺä¸Ĭ
    -0.08
    xin
    -0.08
     Holland
    -0.08
     Alla
    -0.08
    ame
    -0.08
    å®Įå¤ĩ
    -0.08
    对æĪij们
    -0.08
    arn
    -0.08
    POSITIVE LOGITS
    è¿Ļä¹Īåģļ
    0.21
    è¿Ļæł·åģļ
    0.20
    æŃ¤ä¸¾
    0.17
    æѤäºĭ
    0.17
     done
    0.16
    åģļçļĦäºĭ
    0.15
    æĢİä¹Īåģļ
    0.14
     doing
    0.14
    åģļçļĦ
    0.13
    åģļçļĦäºĭæĥħ
    0.13
    Activations Density 0.170%

    No Known Activations