Update README.md

README.md

@@ -6,11 +6,13 @@ datasets:
 - Intel/orca_dpo_pairs
 pipeline_tag: text-generation
 ---
+# Orca-Llama-3-8B-Instruct-DPO
 
 Finetuned [Llama 3 8B Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs) using a single 3090 24GB.
 
 GGUF can be found here [RDson/Orca-Llama-3-8B-Instruct-DPO-GGUF](https://huggingface.co/RDson/Orca-Llama-3-8B-Instruct-DPO-GGUF)
 
+
 ORPOConfig:
 
 ```
@@ -24,19 +26,20 @@ ORPOConfig:
     per_device_eval_batch_size=2,
     gradient_accumulation_steps=4,
     optim="paged_adamw_8bit",
-    num_train_epochs=3,
+    num_train_epochs=1,
     evaluation_strategy="steps",
     eval_steps=0.2,
     logging_steps=1,
-    warmup_steps=10,
+    warmup_steps=35,
     report_to="wandb",
     output_dir="./results/",
-    fp16=True
+    fp16=True,
+    save_steps=50
 ```
 
 <div style="text-align: center;">
-  <img src="https://i.imgur.com/3ObAZpO.png" style="width: 100%; margin: 0 auto; display: inline-block;"/>
-  <img src="https://i.imgur.com/9dInJrS.png" style="width: 100%; margin: 0 auto; display: inline-block;"/>
+  <img src="https://i.imgur.com/vQ4RzSl.png" style="width: 100%; margin: 0 auto; display: inline-block;"/>
+  <img src="https://i.imgur.com/9H75ijW.png" style="width: 100%; margin: 0 auto; display: inline-block;"/>
 </div>
 
 # Model Card for Model ID