43 lines
1.1 KiB
Markdown
43 lines
1.1 KiB
Markdown
---
|
|
license: apache-2.0
|
|
---
|
|
|
|
# Model Card
|
|
|
|
### Model Description
|
|
|
|
OpenLlama 3B fine-tuned using ShareGPT datasets for multi-turn conversations.
|
|
|
|
- **Developed by:** l3utterfly
|
|
- **Funded by:** Layla Network
|
|
- **Model type:** OpenLlama
|
|
- **Language(s) (NLP):** English
|
|
- **License:** Llama2
|
|
- **Finetuned from model:** OpenLlama 3B
|
|
|
|
## Uses
|
|
|
|
Base model used by Layla - the offline personal assistant: https://www.layla-network.ai
|
|
|
|
Help & support: https://discord.gg/x546YJ6nYC
|
|
|
|
Prompt:
|
|
```
|
|
User:
|
|
Assistant:
|
|
```
|
|
|
|
# [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_l3utterfly__open-llama-3b-v2-layla)
|
|
|
|
| Metric | Value |
|
|
|-----------------------|---------------------------|
|
|
| Avg. | 35.63 |
|
|
| ARC (25-shot) | 38.23 |
|
|
| HellaSwag (10-shot) | 66.43 |
|
|
| MMLU (5-shot) | 28.56 |
|
|
| TruthfulQA (0-shot) | 44.4 |
|
|
| Winogrande (5-shot) | 62.83 |
|
|
| GSM8K (5-shot) | 1.06 |
|
|
| DROP (3-shot) | 7.88 |
|