© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6852
    Prev
    Next
    INDEX
    Explanations

    code comments

    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
    ä½ķåĨµ
    -0.10
    å°±æĺ¯åĽłä¸º
    -0.09
    å¾Ĺ太
    -0.09
    æĽ´ä½ķåĨµ
    -0.09
    shine
    -0.09
    0
    -0.08
     Continued
    -0.08
    9
    -0.08
    8
    -0.08
    5
    -0.08
    POSITIVE LOGITS
    æľī人
    0.11
    æĪĸå¤ļ
    0.10
    æľĢ好æĺ¯
    0.10
    åľ¨æĪijçļĦ
    0.09
    æľīçļĦ
    0.09
    è¿ĻéĩĮ
    0.09
    楼主
    0.09
    æľĢ好
    0.08
    è¿ĻçĤ¹
    0.08
    æĪijä¹Ł
    0.08
    Activations Density 0.027%

    No Known Activations