© 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. 31708
    Prev
    Next
    INDEX
    Explanations

    Gary followed by a name

    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
    est
    -0.12
    -minded
    -0.12
    sWith
    -0.11
    оÑģÑĤ
    -0.11
    ãģĵãģĿ
    -0.10
    ×ķת
    -0.10
    çѵ
    -0.10
    éĿ¥
    -0.10
    enance
    -0.10
    åºŀ大çļĦ
    -0.10
    POSITIVE LOGITS
    å¤ł
    0.11
    ----------------------------------------------------------------
    0.11
    iting
    0.11
    ¯
    0.10
    ãĤĪãģĨãģª
    0.10
    éné
    0.09
    çĦ¶æĺ¯
    0.09
    ÑĤив
    0.09
    ysis
    0.09
    ãĥ³ãĤ°
    0.09
    Activations Density 0.272%

    No Known Activations