© 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. 32795
    Prev
    Next
    INDEX
    Explanations

    commitment and progress

    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
    弥补
    -0.08
    ä»ĭ
    -0.08
    é»Ħæĺı
    -0.08
    unch
    -0.08
    æ£ł
    -0.08
    Hop
    -0.08
    TEX
    -0.08
    HostName
    -0.08
     kite
    -0.08
     Strait
    -0.08
    POSITIVE LOGITS
    åĪ¶è®¢
    0.11
    为èĩªå·±
    0.11
    æĦıå¿Ĺ
    0.11
    åħĥæŦ
    0.11
    åĨ³å¿ĥ
    0.10
    计åĪĴ
    0.10
    ç»Ļèĩªå·±
    0.10
    ä¸ŃéĢĶ
    0.10
    åģĩæĹ¥
    0.10
     commitment
    0.10
    Activations Density 0.066%

    No Known Activations