Platypus2-70B-instruct
| Property | Value |
|---|---|
| Parameter Count | 70B |
| License | CC BY-NC 4.0 |
| Architecture | LLaMA 2 |
| Research Paper | Platypus Paper |
| Benchmark Score | 66.89 (Average) |
What is Platypus2-70B-instruct?
Platypus2-70B-instruct is an advanced language model that combines the strengths of Platypus2-70B and Llama-2-70b-instruct. Trained specifically for STEM and logic-based tasks using the Open-Platypus dataset, this model represents a significant advancement in specialized language modeling.
Implementation Details
The model was developed using LoRA fine-tuning on 8 A100 80GB GPUs, implementing the LLaMA 2 transformer architecture. It supports both FP16 and F32 tensor types and follows a specific prompt template for optimal performance.
- Trained on STEM and logic-based datasets
- Implements instruction-following capabilities
- Achieves strong performance on various benchmarks (ARC: 71.84, HellaSwag: 87.94, MMLU: 70.48)
Core Capabilities
- Enhanced STEM and logical reasoning abilities
- Strong performance in knowledge-intensive tasks
- Effective instruction following with structured input format
- Balanced performance across multiple evaluation metrics
Frequently Asked Questions
Q: What makes this model unique?
This model uniquely combines STEM-focused training from Platypus2-70B with the instruction-following capabilities of Llama-2-70b-instruct, creating a versatile model especially strong in technical and logical reasoning tasks.
Q: What are the recommended use cases?
The model is particularly well-suited for STEM applications, technical problem-solving, and logical reasoning tasks. It performs exceptionally well in academic and scientific contexts while maintaining strong general-purpose capabilities.
