Improve model card: Add pipeline tag, library, paper, code, and usage

README.md

@@ -1,29 +1,35 @@
 ---
 license: apache-2.0
+pipeline_tag: text-generation
+library_name: transformers
 ---
+
 # HRWKV7-hxa079-Qwen3-8B
 
+**Paper:** [RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale](https://huggingface.co/papers/2505.03005)
+**Code:** [https://github.com/recursal/RADLADS](https://github.com/recursal/RADLADS)
+
 ### Model Description
 
 HRWKV7-Qwen3-8N-Preview is an RNN hybrid architecture model that combines RWKV v7's linear attention mechanism with Group Query Attention (GQA) layers. Built upon the Qwen3-8B foundation, this model replaces most Transformer attention blocks with RWKV blocks while strategically maintaining some GQA layers to enhance performance on specific tasks.
 
-- **Developed by:** OpenMOSE
-- **Model type:** Hybrid Linear-Attention Language Model
-- **Language(s):** Multilingual (inherited from Qwen3-8B)
-- **License:** Apache-2.0
-- **Base Model:** Qwen3-8B
-- **Year:** 2025
+-   **Developed by:** OpenMOSE
+-   **Model type:** Hybrid Linear-Attention Language Model
+-   **Language(s):** Multilingual (inherited from Qwen3-8B)
+-   **License:** Apache-2.0
+-   **Base Model:** Qwen3-8B
+-   **Year:** 2025
 
 ### Architecture Specifications
 
-- **Architecture:** RWKV v7 based "hxa079" Architecture + Group Query Attention Hybrid
-- **Total Layers:** 36 layers (L36D4096)
-  - 32 RWKV layers (with Rope)
-  - 4 GQA layers (No Rope, No Position Embeddings)
-- **Hidden Dimension:** 4096
-- **Training Context Window:** 4096 tokens
-- **Inference Context Window** 16384+
-- **Training Strategy** Following RADLADS method based knowledge distillation
+-   **Architecture:** RWKV v7 based "hxa079" Architecture + Group Query Attention Hybrid
+-   **Total Layers:** 36 layers (L36D4096)
+    -   32 RWKV layers (with Rope)
+    -   4 GQA layers (No Rope, No Position Embeddings)
+-   **Hidden Dimension:** 4096
+-   **Training Context Window:** 4096 tokens
+-   **Inference Context Window** 16384+
+-   **Training Strategy** Following RADLADS method based knowledge distillation
 
 ## Technical Innovation
 
@@ -31,54 +37,91 @@ HRWKV7-Qwen3-8N-Preview is an RNN hybrid architecture model that combines RWKV v
 
 The model implements several key improvements over original RWKV architectures:
 
-1. **Token Shift Removal**: In order to effectively inherit the teacher model weights, we removed the residual connection one token ago.
-2. **GroupNorm Removal**: Helps improve training stability issues
-3. **k_first Introduction**: Experimentally adopted the approach of residually connecting k layers in layer 0.
+1.  **Token Shift Removal**: In order to effectively inherit the teacher model weights, we removed the residual connection one token ago.
+2.  **GroupNorm Removal**: Helps improve training stability issues
+3.  **k_first Introduction**: Experimentally adopted the approach of residually connecting k layers in layer 0.
 
 ### Hybrid Design Benefits
 
-- **Linear Attention Inference**: RWKV blocks enable O(1) memory complexity during inference, and the hybrid approach reduces the KVCache to 1/9 of full GQA.
-- **Enhanced Needle Tasks**: Strategic placement of GQA layers significantly improves performance on needle-in-haystack retrieval tasks, addressing a known limitation of pure linear attention models
-- **Implicit Position Encoding**: Interestingly, the model achieves better performance when RoPE (Rotary Position Embedding) is not applied to GQA layers, suggesting that RWKV blocks provide implicit positional encoding capabilities
+-   **Linear Attention Inference**: RWKV blocks enable O(1) memory complexity during inference, and the hybrid approach reduces the KVCache to 1/9 of full GQA.
+-   **Enhanced Needle Tasks**: Strategic placement of GQA layers significantly improves performance on needle-in-haystack retrieval tasks, addressing a known limitation of pure linear attention models
+-   **Implicit Position Encoding**: Interestingly, the model achieves better performance when RoPE (Rotary Position Embedding) is not applied to GQA layers, suggesting that RWKV blocks provide implicit positional encoding capabilities
 
 ## Intended Use
 
 This is an **experimental research model** designed to explore hybrid architectures combining linear and quadratic attention mechanisms. It is intended for:
 
-- Research into efficient attention mechanisms
-- Benchmarking hybrid architecture performance
-- Exploring linear attention limitations and solutions
-- Academic and industrial R&D purposes
+-   Research into efficient attention mechanisms
+-   Benchmarking hybrid architecture performance
+-   Exploring linear attention limitations and solutions
+-   Academic and industrial R&D purposes
 
 ## Limitations
 
-- **Experimental Status**: This model is in experimental stages and may exhibit unexpected behaviors
-- **Context Window**: Limited to 4096 tokens during training, though RWKV architecture theoretically supports longer sequences
-- **Performance Variability**: As a hybrid model, performance may vary significantly across different task types
+-   **Experimental Status**: This model is in experimental stages and may exhibit unexpected behaviors
+-   **Context Window**: Limited to 4096 tokens during training, though RWKV architecture theoretically supports longer sequences
+-   **Performance Variability**: As a hybrid model, performance may vary significantly across different task types
 
 ## Training Details
 
-- **Training Context Window:** 4096 tokens
-- **Training GPU** AMD MI300X x 1(takes 80hrs) Runpod
-- **Training Strategy** 8bit MLP Quant, frozen emb,mlp,head, Deepspeed Stage1, Stage1 100M, Stage2 360M
-- **Base Model Initialization:** Weights initialized from Qwen3-8B
-- **Architecture Conversion:** Transformer attention blocks systematically replaced with RWKV blocks, except for 6 strategically placed GQA layers
+-   **Training Context Window:** 4096 tokens
+-   **Training GPU** AMD MI300X x 1(takes 80hrs) Runpod
+-   **Training Strategy** 8bit MLP Quant, frozen emb,mlp,head, Deepspeed Stage1, Stage1 100M, Stage2 360M
+-   **Base Model Initialization:** Weights initialized from Qwen3-8B
+-   **Architecture Conversion:** Transformer attention blocks systematically replaced with RWKV blocks, except for 6 strategically placed GQA layers
 
 ## Evaluation
 
 Performance evaluation is ongoing. The model shows promising results in:
-- Maintaining base model capabilities while achieving linear attention efficiency
-- Significantly improved needle-in-haystack task performance compared to pure RWKV architectures
-- Competitive performance on standard language modeling benchmarks
+-   Maintaining base model capabilities while achieving linear attention efficiency
+-   Significantly improved needle-in-haystack task performance compared to pure RWKV architectures
+-   Competitive performance on standard language modeling benchmarks
+
+## Sample Usage
+
+You can use this model with the Hugging Face `transformers` library. Ensure you have `trust_remote_code=True` set, as it uses a custom architecture.
 
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+model_name = "OpenMOSE/HRWKV7-hxa079-Qwen3-8B" # The name of this model
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype=torch.bfloat16, # Or torch.float16, depending on your system and model precision
+    device_map="auto",
+    trust_remote_code=True,
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+
+prompt = "Tell me a short story about a brave knight named Sir Reginald."
+inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True, top_k=50, top_p=0.95, temperature=0.7)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
 
 ## Thank you for Big help :)
-- SmerkyG Inspired by RADLADS (https://arxiv.org/abs/2505.03005)
-- https://github.com/recursal/RADLADS-paper
+-   SmerkyG Inspired by RADLADS ([https://arxiv.org/abs/2505.03005](https://arxiv.org/abs/2505.03005))
+-   [https://github.com/recursal/RADLADS-paper](https://github.com/recursal/RADLADS-paper) (This is the primary repository for the RADLADS paper's code)
 
 ## Training Code
-- https://github.com/OpenMOSE/RWKVInside (still buggy)
-
+-   [https://github.com/OpenMOSE/RWKVInside](https://github.com/OpenMOSE/RWKVInside) (This repository contains training code specifically for this model variant, still buggy)
+
+## Citation
+
+If you use this model or find our work valuable, please consider citing the RADLADS paper:
+
+```bibtex
+@misc{goldstein2025radladsrapidattentiondistillation,
+      title={RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale},
+      author={Daniel Goldstein and Eric Alcaide and Janna Lu and Eugene Cheah},
+      year={2025},
+      eprint={2505.03005},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2505.03005},
+}
+```
 
 ## Model Card Contact