Update README.md

README.md

@@ -62,10 +62,11 @@ print(tokenizer.decode(tokens[0], skip_special_tokens=True))
 The dataset is comprised of a mixture of open datasets large-scale datasets available on the [HuggingFace Hub](https://huggingface.co/datasets):
 - HuggingFaceH4/ultrachat_200k
 - HuggingFaceH4/ultrafeedback_binarized
+- Intel/orca_dpo_pairs
 - meta-math/MetaMathQA
-- Capybara
 - Instruct Code Dataset (Internal)
 - Wizard Dataset
+- Open-Orca/SlimOrca
 
 ### Training Procedure
 
@@ -77,7 +78,7 @@ The dataset is comprised of a mixture of open datasets large-scale datasets avai
 
 | Model | Size | Alignment | MT-Bench (score) | AlpacaEval (win rate %) |
 |-------------|-----|----|---------------|--------------|
-| **Stable Zephyr 3B** 🪁 | 3B | DPO | 6.86 | 75.19 |
+| **Stable Zephyr 3B** 🪁 | 3B | DPO | 6.64 | 76.00 |
 | Stable Zephyr (SFT only) | 3B | SFT | 7.12 | 71.15 | 
 | MPT-Chat |  7B |dSFT |5.42| -|
 | Xwin-LMv0.1 | 7B| dPPO| 6.19| 87.83|