© 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. 4704
    Prev
    Next
    INDEX
    Explanations

    not followed by verb

    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
    ä¸Ńæĸĩ
    -0.10
    _GF
    -0.09
    积水
    -0.09
    ä¸įåĪ°
    -0.09
    erus
    -0.08
    幸
    -0.08
    æĪijæīį
    -0.08
    å¾Īæ£Ĵ
    -0.08
    åŨ
    -0.08
    åĶı
    -0.08
    POSITIVE LOGITS
    åıªè¦ģæľī
    0.09
    ä»İå°ı
    0.09
    微信åħ¬ä¼Ĺ
    0.09
    زاÙĦ
    0.09
     Cald
    0.08
    ẩ
    0.08
    èĥŃ
    0.08
     Äijủ
    0.08
    ä¸įåłª
    0.08
    endant
    0.08
    Activations Density 0.032%

    No Known Activations