© 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. 56969
    Prev
    Next
    INDEX
    Explanations

    Mill, Mills, MacMillan, McMillan, Carmilla, Camilla

    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
    çĶ¨è½¦
    -0.08
    ä¸ĥ大
    -0.08
    絡
    -0.08
    ual
    -0.08
    oor
    -0.08
    æŁ¢
    -0.08
    atron
    -0.08
    ouve
    -0.08
    è²Į
    -0.08
    entions
    -0.08
    POSITIVE LOGITS
    jee
    0.11
    æ»ĩ
    0.10
    常å§Ķä¼ļ
    0.10
    mond
    0.10
    alker
    0.09
     thiá»ĥu
    0.09
    onto
    0.09
    SEE
    0.09
     dÃ¼ÅŁ
    0.09
    ием
    0.09
    Activations Density 0.005%

    No Known Activations