cjiao/goldengoose-divsweep_goose_n512_grouporc_tau0.50_seed100-7grp

Model Overview

This model, goldengoose-divsweep_goose_n512_grouporc_tau0.50_seed100-7grp, is a 1.5 billion parameter language model fine-tuned from the Qwen/Qwen2.5-1.5B-Instruct base. It leverages a substantial 32768-token context window, making it suitable for processing longer inputs and generating more extensive responses.

Key Capabilities

• Enhanced Mathematical Reasoning: The model was specifically trained using the GRPO (Grouped Reinforcement Learning with Policy Optimization) method, as introduced in the DeepSeekMath research. This training approach aims to significantly improve its performance on mathematical and logical reasoning tasks.
• Instruction Following: As a fine-tuned instruction model, it is designed to understand and execute user prompts effectively, generating relevant and coherent text based on given instructions.
• Efficient Performance: With 1.5 billion parameters, it offers a balance between performance and computational efficiency, making it accessible for various applications.

Training Details

The fine-tuning process utilized the TRL (Transformer Reinforcement Learning) library. The application of the GRPO method, detailed in the paper "DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models" arXiv:2402.03300, is central to its specialized capabilities.

When to Use This Model

This model is particularly well-suited for applications requiring:

• Solving mathematical problems or generating mathematical explanations.
• Tasks that benefit from improved logical reasoning and structured output.
• Instruction-following scenarios where a smaller, yet capable, model is preferred for efficiency.