© 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. 34524
    Prev
    Next
    INDEX
    Explanations

    okay with

    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
    rys
    -0.10
    еÑĢÑĮ
    -0.10
    _segments
    -0.09
    ry
    -0.09
    rites
    -0.09
    ä¸įå¾Ĺå·²
    -0.08
    çĨ¬
    -0.08
     полÑĮз
    -0.08
    aland
    -0.08
    InOut
    -0.08
    POSITIVE LOGITS
    æĬīæĭ©
    0.09
    ä¸įå¾Ĺ转载
    0.09
    è¿Ļä¸ĢçĤ¹
    0.09
    åĮ¯
    0.08
    .hw
    0.08
    ãĥĢãĥ¼
    0.08
    çªľ
    0.08
    ãģĭãģ¤
    0.08
    ä¾ĥ
    0.08
    ibe
    0.08
    Activations Density 0.149%

    No Known Activations