ChaoticNeutrals/Kool-Aid_7B
Kool-Aid_7B is a 7 billion parameter language model created by ChaoticNeutrals, formed by merging ErosEris and CookieNexus using the SLERP method. This model achieves an average score of 69.69 on the Open LLM Leaderboard, demonstrating capabilities across various reasoning and language understanding tasks. With a 4096 token context length, it is suitable for general-purpose applications requiring robust language generation and comprehension.
Loading preview...
Kool-Aid_7B Model Overview
Kool-Aid_7B is a 7 billion parameter language model developed by ChaoticNeutrals, created through a strategic merge of the pre-trained models ErosEris and CookieNexus. This model leverages the SLERP (Spherical Linear Interpolation) merge method, combining the strengths of its constituent models to enhance overall performance.
Key Capabilities & Performance
Evaluated on the Open LLM Leaderboard, Kool-Aid_7B demonstrates solid performance across a range of benchmarks, achieving an average score of 69.69. Specific benchmark results include:
- AI2 Reasoning Challenge (25-Shot): 67.49
- HellaSwag (10-Shot): 86.13
- MMLU (5-Shot): 63.82
- TruthfulQA (0-shot): 65.12
- Winogrande (5-shot): 81.37
- GSM8k (5-shot): 54.21
These scores indicate proficiency in reasoning, common sense, factual recall, and mathematical problem-solving. The model was configured using a specific YAML setup, applying varying interpolation values across self-attention and MLP layers to optimize the merge.
Use Cases
Given its balanced performance across multiple benchmarks and 7B parameter size, Kool-Aid_7B is well-suited for:
- General text generation and comprehension tasks.
- Applications requiring moderate reasoning and factual understanding.
- Scenarios where a 4096 token context length is sufficient for processing inputs.