Update README.md

README.md

@@ -13,7 +13,50 @@ tags:
 - stable-diffusion
 - sdxl
 datasets:
-- KavinduHansaka/prompt-gen-8k-flux-sdxl
+- KavinduHansaka/prompt-gen-10k-flux-sdxl
+
+model-index:
+- name: Llama-3.2-1B-ImageGen
+  results:
+  - task:
+      type: text-generation
+      name: Prompt Generation (Dev)
+    dataset:
+      type: json
+      name: prompt_gen_refined_dev500
+    metrics:
+    - name: eval_loss
+      type: loss
+      value: 0.7515
+      verified: false
+    - name: perplexity
+      type: perplexity
+      value: 2.12
+      verified: false
+    - name: avg_target_words
+      type: length
+      value: 106.8
+      verified: false
+
+  - task:
+      type: text-generation
+      name: Prompt Generation (Test)
+    dataset:
+      type: json
+      name: prompt_gen_refined_test500
+    metrics:
+    - name: eval_loss
+      type: loss
+      value: 0.7483
+      verified: false
+    - name: perplexity
+      type: perplexity
+      value: 2.11
+      verified: false
+    - name: avg_target_words
+      type: length
+      value: 106.7
+      verified: false
 ---
 
 # Llama-3.2-1B — Image Prompt Generation (LoRA Merged)
@@ -33,12 +76,12 @@ It is designed to create **cinematic, detailed, and structured prompts** for tex
 - **Languages:** English (prompt tags, stylistic descriptors)  
 - **License:** MIT  
 - **Finetuned with:** LoRA adapters, then merged  
-- **Training dataset:** [prompt-gen-8k-flux-sdxl](https://huggingface.co/datasets/KavinduHansaka/prompt-gen-8k-flux-sdxl)  
+- **Training dataset:** [prompt-gen-10k-flux-sdxl](https://huggingface.co/datasets/KavinduHansaka/prompt-gen-10k-flux-sdxl)  
 
 ### Model Sources
 - **Merged model repo:** https://huggingface.co/KavinduHansaka/Llama-3.2-1B-ImageGen  
 - **LoRA adapter repo:** https://huggingface.co/KavinduHansaka/Llama-3.2-1B-ImageGen-LoRA 
-- **Training dataset:** https://huggingface.co/datasets/KavinduHansaka/prompt-gen-8k-flux-sdxl  
+- **Training dataset:** https://huggingface.co/datasets/KavinduHansaka/prompt-gen-10k-flux-sdxl  
 
 ---
 
@@ -89,7 +132,7 @@ print(tok.decode(out[0], skip_special_tokens=True))
 
 ## Training Details
 
-- **Training data:** [prompt-gen-8k-flux-sdxl](https://huggingface.co/datasets/KavinduHansaka/prompt-gen-8k-flux-sdxl)  
+- **Training data:** [prompt-gen-8k-flux-sdxl](https://huggingface.co/datasets/KavinduHansaka/prompt-gen-10k-flux-sdxl)  
 - **Training method:** LoRA with PEFT, adapters merged into base model.  
 - **Precision:** bfloat16/float16 during training.  
 
@@ -111,7 +154,6 @@ title = {LLaMA 3.2 (1B)},
 author = {Meta AI},
 year = {2024},
 url = {https://huggingface.co/meta-llama/Llama-3.2-1B}
-
 }
 
 @misc{llama3.2-1b-promptgen,