| model_type: llama | |
| pipeline_tag: text-generation | |
| datasets: | |
| - togethercomputer/RedPajama-Data-1T-Sample | |
| tags: | |
| - llama | |
| This is [Llama 2 13b](https://huggingface.co/meta-llama/Llama-2-13b-hf) with some additional attention heads from original-flavor Llama 33b frankensteined on. | |
| Fine-tuned on ~10M tokens from RedPajama to settle in the transplants a little. | |
| Not intended for use as-is - this model is meant to serve as a base for further tuning, hopefully with a greater capacity for learning than 13b. | |
| # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) | |
| Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chargoddard__llama2-22b) | |
| | Metric | Value | | |
| |-----------------------|---------------------------| | |
| | Avg. | 46.85 | | |
| | ARC (25-shot) | 58.53 | | |
| | HellaSwag (10-shot) | 82.55 | | |
| | MMLU (5-shot) | 54.68 | | |
| | TruthfulQA (0-shot) | 39.84 | | |
| | Winogrande (5-shot) | 76.32 | | |
| | GSM8K (5-shot) | 9.93 | | |
| | DROP (3-shot) | 6.08 | | |