Update README.md

README.md

@@ -105,7 +105,7 @@ Please use "Please reason step by step, and put your final answer within \boxed{
 #### System‑Level Safety:
 - The model is designed to be deployed as part of a broader system that implements safety measures (e.g., Prompt Guard, Code Shield) to ensure outputs remain safe even under adversarial conditions.
 
----
+---s
 
 ### Safety Fine‑Tuning & Data Strategy
 
@@ -181,6 +181,11 @@ hf (pretrained=EpistemeAI/ReasoningCore-3B-R01), gen_kwargs: (None), limit: None
 |gpqa_diamond_zeroshot|      1|none  |     0|acc     |↑  |0.3182|±  |0.0332|
 |                     |       |none  |     0|acc_norm|↑  |0.3182|±  |0.0332|
 
+hf (pretrained=EpistemeAI/ReasoningCore-3B-R01), gen_kwargs: (None), limit: None, num_fewshot: 5, batch_size: 8
+|      Tasks       |Version|     Filter     |n-shot|  Metric   |   |Value |   |Stderr|
+|------------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
+|gsm8k_cot_zeroshot|      3|flexible-extract|     0|exact_match|↑  |0.3154|±  |0.0128|
+|                  |       |strict-match    |     0|exact_match|↑  |0.2873|±  |0.0125|
 
 # Uploaded  model