accuracy-maker/Llama-3.2-1B-GRPO-gsm8k

LLaMA-3.2-1B-Instruct Post-training by GRPO from DeepSeek

This model is a post-trained version of LLaMA-3.2-1B-Instruct.

Model Details

• Base Model: LLaMA-3.2-1B
• Training Data: openai/gsm8k
• Post-training Steps: 1000
• Checkpoint: checkpoint-1000/
• Framework: Hugging Face transformers
• Usage: Mathematical Reasoning.

How to Use

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "accuracy-maker/Llama-3.2-1B-GRPO-gsm8k"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

input_text = "What is the capital of France?"
generate_with_stream(input_text)