© 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. 4714
    Prev
    Next
    INDEX
    Explanations

    parents

    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
    ç¥
    -0.10
    ÑĢаÐ
    -0.10
    åĵĪå°Ķ
    -0.10
    WOOD
    -0.10
    hausen
    -0.10
    _hal
    -0.09
    mans
    -0.09
    æĶ¿åºľéĥ¨éŨ
    -0.09
    inand
    -0.09
     Nunes
    -0.09
    POSITIVE LOGITS
     themselves
    0.14
    åıĮåıĮ
    0.11
    俱
    0.10
    alled
    0.09
    åĪĨåĪ«æĺ¯
    0.09
    åķĨ
    0.09
    é½IJåħ¨
    0.08
    á»ijt
    0.08
    éĤ£éĩĮ
    0.08
    zt
    0.08
    Activations Density 0.027%

    No Known Activations