Update README.md

README.md

@@ -73,10 +73,10 @@ These benchmark are current benchmark and not final benchmark, due to recursive
 
 hf (pretrained=EpistemeAI/metatune-gpt20b-R0,parallelize=True,dtype=bfloat16), gen_kwargs: (temperature=1,top_p=1,max_new_tokens=1000), limit: 30.0, num_fewshot: 5, batch_size: 1
 
-  |             Tasks             |Version|     Filter     |n-shot|  Metric   |   |metatune|MiniMax M1 80k|Llama 4 Maverick|
-|-------------------------------|------:|----------------|-----:|-----------|---|-----:|-----:|-----:|
-|gsm8k_cot                      |      3|flexible-extract|     5|exact_match|↑  |0.91  |   -  |   -  |
-|gpqa_diamond_cot_n_shot        |      2|flexible-extract|     0|exact_match|↑  |0.722 |0.70  |0.67|
+|             Tasks           |metatune|MiniMax M1 80k|Llama 4 Maverick|
+|:----------------------------|:-----|:-----|:----- |
+|gsm8k_cot                    |0.91  |   -  |   -  |
+|gpqa_diamond_cot_n_shot      |0.722 |0.70  |0.67|
 
 ## Thank you
 - OpenAI