© 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. 56451
    Prev
    Next
    INDEX
    Explanations

    handcuffs, wrists, hands, tape, guard

    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
    延
    -0.11
    æĭĶ
    -0.11
    å¡Į
    -0.10
     наÑĢ
    -0.10
    è·ĥ
    -0.10
    è¯ķæİ¢
    -0.10
    æĿ¥ä¸´
    -0.09
    åĬłéĢŁ
    -0.09
    _dispatcher
    -0.09
    å¿Ĩ
    -0.09
    POSITIVE LOGITS
    gio
    0.09
    åİļåİļ
    0.08
    ä¸Ĭ线
    0.08
    衬
    0.08
    éĻIJåĪ¶
    0.08
    ç¼ļ
    0.08
    éħįå¥Ĺ
    0.07
     Situation
    0.07
    æĬĬä»ĸ
    0.07
    åĪĽéĢłåĬĽ
    0.07
    Activations Density 0.029%

    No Known Activations