Update README.md

README.md

@@ -12,4 +12,69 @@ tags:
 - open
 - r1
 - math
----
+---
+# **Open-R1-Math-7B-Instruct**
+
+The *Open-R1-Math-7B-Instruct* is a fine-tuned language model designed for advanced reasoning and instruction‐following tasks. It leverages the Qwen2.5-7B base model and has been fine-tuned on a chain of thought reasoning dataset derived from [OpenR1-Math-220k](https://huggingface.co/datasets/open-r1/OpenR1-Math-220k). This model is optimized for tasks requiring logical reasoning, detailed explanations, and multi-step problem-solving, making it ideal for applications such as instruction-following, text generation, and complex reasoning tasks.
+
+# **Quickstart with Transformers**
+
+Below is a code snippet using `apply_chat_template` to show how to load the tokenizer and model and how to generate content:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_name = "Open-R1-Math-7B-Instruct"
+
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+prompt = "How many r in strawberry."
+messages = [
+    {"role": "system", "content": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."},
+    {"role": "user", "content": prompt}
+]
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True
+)
+model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+
+generated_ids = model.generate(
+    **model_inputs,
+    max_new_tokens=512
+)
+generated_ids = [
+    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
+]
+
+response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
+```
+
+# **Intended Use**
+
+The Open-R1-Math-7B-Instruct model is designed for advanced reasoning and instruction-following tasks, with specific applications including:
+
+1. **Instruction Following**: Providing detailed and step-by-step guidance for a wide range of user queries.
+2. **Logical Reasoning**: Solving problems requiring multi-step thought processes, such as math problems or complex logic-based scenarios.
+3. **Text Generation**: Crafting coherent, contextually relevant, and well-structured text in response to prompts.
+4. **Problem-Solving**: Analyzing and addressing tasks that require chain-of-thought (CoT) reasoning, making it ideal for education, tutoring, and technical support.
+5. **Knowledge Enhancement**: Leveraging reasoning datasets to offer deeper insights and explanations for a wide variety of topics.
+
+# **Limitations**
+
+1. **Data Bias**: As the model is fine-tuned on specific datasets, its outputs may reflect inherent biases from the training data.
+2. **Context Limitation**: Performance may degrade for tasks requiring knowledge or reasoning that significantly exceeds the model's pretraining or fine-tuning context.
+3. **Complexity Ceiling**: While optimized for multi-step reasoning, exceedingly complex or abstract problems may result in incomplete or incorrect outputs.
+4. **Dependency on Prompt Quality**: The quality and specificity of the user prompt heavily influence the model's responses.
+5. **Non-Factual Outputs**: Despite being fine-tuned for reasoning, the model can still generate hallucinated or factually inaccurate content, particularly for niche or unverified topics.
+6. **Computational Requirements**: Running the model effectively requires significant computational resources, particularly when generating long sequences or handling high-concurrency workloads.
+
+---
+
+This version reflects the new name *Open-R1-Math-7B-Instruct* and specifies that its fine-tuning data comes from the [OpenR1-Math-220k dataset](https://huggingface.co/datasets/open-r1/OpenR1-Math-220k).