| license: llama2 | |
| datasets: | |
| - WizardLM/WizardLM_evol_instruct_V2_196k | |
| language: | |
| - en | |
| pipeline_tag: conversational | |
| # LLaMA 2 Wizard 70B QLoRA | |
| Fine tuned on WizardLM/WizardLM_evol_instruct_V2_196k dataset. | |
| # [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_v2ray__LLaMA-2-Wizard-70B-QLoRA) | |
| | Metric | Value | | |
| |-----------------------|---------------------------| | |
| | Avg. | 65.4 | | |
| | ARC (25-shot) | 67.58 | | |
| | HellaSwag (10-shot) | 87.52 | | |
| | MMLU (5-shot) | 69.11 | | |
| | TruthfulQA (0-shot) | 61.79 | | |
| | Winogrande (5-shot) | 82.32 | | |
| | GSM8K (5-shot) | 30.48 | | |
| | DROP (3-shot) | 59.03 | | |