© 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. 49874
    Prev
    Next
    INDEX
    Explanations

    link, tied, changes

    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
    à¹Ģà¸ľ
    -0.10
    ä¼Ĺ
    -0.10
    uario
    -0.09
    æī®
    -0.09
    æ¦Ľ
    -0.09
    indr
    -0.09
    辩è¯ģ
    -0.08
    çĭ¬è§Ĵ
    -0.08
    _IOC
    -0.08
    为åħ¶
    -0.08
    POSITIVE LOGITS
    å®ŀéĻħ
    0.11
     tied
    0.11
    浴室
    0.10
    çī¹å®ļ
    0.10
     ìĭ¤ìłľ
    0.09
    legate
    0.09
     tie
    0.09
    çļĦåĪ©çĽĬ
    0.09
    æŶæķĪ
    0.09
    實éļĽ
    0.09
    Activations Density 0.093%

    No Known Activations