|
--- |
|
license: other |
|
tags: |
|
- llama |
|
- llama-2 |
|
- pytorch |
|
- chatbot |
|
- storywriting |
|
- generalist-model |
|
--- |
|
# chronos-hermes-13b-v2 |
|
([chronos-13b-v2](https://huggingface.co/elinas/chronos-13b-v2) + [Nous-Hermes-Llama2-13b](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b)) 75/25 merge |
|
|
|
This offers the imaginative writing style of chronos while still retaining coherency and being capable. Outputs are long and utilize exceptional prose. |
|
Supports a maxium context length of 4096. |
|
|
|
- [GPTQ Quantized Weights](https://huggingface.co/Austism/chronos-hermes-13b-v2-GPTQ) |
|
|
|
## Prompt Format |
|
|
|
The model follows the Alpaca prompt format: |
|
``` |
|
### Instruction: |
|
<prompt> |
|
|
|
### Response: |
|
|
|
``` |
|
This is an adaption of [chronos-hermes-13b](https://huggingface.co/Austism/chronos-hermes-13b) for llama-2. |
|
# [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_Austism__chronos-hermes-13b-v2) |
|
|
|
| Metric | Value | |
|
|-----------------------|---------------------------| |
|
| Avg. | 49.34 | |
|
| ARC (25-shot) | 60.32 | |
|
| HellaSwag (10-shot) | 83.21 | |
|
| MMLU (5-shot) | 55.05 | |
|
| TruthfulQA (0-shot) | 50.91 | |
|
| Winogrande (5-shot) | 75.37 | |
|
| GSM8K (5-shot) | 11.75 | |
|
| DROP (3-shot) | 8.73 | |
|
|
