updated

README.md

@@ -1,3 +1,13 @@
+---
+license: other
+language:
+- en
+pipeline_tag: text-generation
+datasets:
+- LDJnr/Puffin
+- pvduy/rm_hh_helpful_only
+library_name: peft
+---
 <table>
 <tr>
 <td style="width: 30%; text-align: left; vertical-align: middle">
@@ -13,7 +23,7 @@ Using Microsoft's Phi 1.5 model like it was never intended.
 </table>
 
 # Main Procedure
-This model is an adapter on [puffin phi v2](https://huggingface.co/teknium/Puffin-Phi-v2) trained using QLoRA and DPO on 60,000 samples from the [anthropic helpful only](https://huggingface.co/datasets/pvduy/rm_hh_helpful_only) dataset.
+This model is an adapter on [puffin phi v2](https://huggingface.co/teknium/Puffin-Phi-v2) trained using [QLoRA](https://arxiv.org/pdf/2305.14314.pdf) and [DPO](https://arxiv.org/pdf/2305.18290.pdf) on 60,000 samples from the [anthropic helpful only](https://huggingface.co/datasets/pvduy/rm_hh_helpful_only) dataset.
 
 
 ---
@@ -36,4 +46,4 @@ The following `bitsandbytes` quantization config was used during training:
 ### Framework versions
 
 
-- PEFT 0.5.0
+- PEFT 0.5.0