|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- Open-Orca/SlimOrca |
|
- jondurbin/airoboros-3.1 |
|
- riddle_sense |
|
language: |
|
- en |
|
library_name: transformers |
|
--- |
|
|
|
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl) |
|
|
|
# SlimOrcaBoros |
|
|
|
A Mistral 7B finetuned model using SlimOrca, Auroboros 3.1 and RiddleSense. |
|
|
|
### Training |
|
|
|
Trained for 4 epochs, but released @ epoch 3. |
|
# [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_openaccess-ai-collective__mistral-7b-slimorcaboros) |
|
|
|
| Metric | Value | |
|
|-----------------------|---------------------------| |
|
| Avg. | 54.1 | |
|
| ARC (25-shot) | 63.65 | |
|
| HellaSwag (10-shot) | 83.7 | |
|
| MMLU (5-shot) | 63.46 | |
|
| TruthfulQA (0-shot) | 55.81 | |
|
| Winogrande (5-shot) | 77.03 | |
|
| GSM8K (5-shot) | 23.43 | |
|
| DROP (3-shot) | 11.62 | |
|
|
