60 lines
1.5 KiB
Markdown
60 lines
1.5 KiB
Markdown
|
---
|
||
|
|
language:
|
||
|
|
- ko
|
||
|
|
datasets:
|
||
|
|
- instruction
|
||
|
|
library_name: transformers
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
license: mit
|
||
|
|
---
|
||
|
|
# **etri-ones-solar**
|
||
|
|
|
||
|
|
## Model Details
|
||
|
|
**Model Developers**
|
||
|
|
- the model is fine-tuned by open instruction dataset
|
||
|
|
|
||
|
|
**Model Architecture**
|
||
|
|
- this model is an auto-regressive language model based on the solar transformer architecture.
|
||
|
|
|
||
|
|
**Base Model**
|
||
|
|
- solar https://huggingface.co/upstage/SOLAR-10.7B-v1.0
|
||
|
|
|
||
|
|
**Training Dataset**
|
||
|
|
-
|
||
|
|
|
||
|
|
---
|
||
|
|
# Model comparisons1
|
||
|
|
> comming soon
|
||
|
|
|
||
|
|
| Model | Average | Ko-ARC | Ko-HellaSwag | Ko-MMLU | Ko-TruthfulQA | Ko-CommonGen V2 |
|
||
|
|
| --- | --- | --- | --- | --- | --- | --- |
|
||
|
|
| **[...your_model_name...]** | NaN | NaN | NaN | NaN | NaN | NaN |
|
||
|
|
|
||
|
|
---
|
||
|
|
# Model comparisons2
|
||
|
|
> AI-Harness evaluation; [link](https://github.com/Beomi/ko-lm-evaluation-harness)
|
||
|
|
|
||
|
|
| Model | Copa | Copa | HellaSwag | HellaSwag | BoolQ | BoolQ | Sentineg | Sentineg |
|
||
|
|
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
||
|
|
| | 0-shot | 5-shot | 0-shot | 5-shot | 0-shot | 5-shot | 0-shot | 5-shot |
|
||
|
|
| **[...your_model_name...]** | NaN | NaN | NaN | NaN | NaN | NaN | NaN | NaN |
|
||
|
|
|
||
|
|
---
|
||
|
|
# Implementation Code
|
||
|
|
```python
|
||
|
|
### KO-Platypus
|
||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||
|
|
import torch
|
||
|
|
|
||
|
|
repo = "[...your_model_repo...]"
|
||
|
|
OpenOrca = AutoModelForCausalLM.from_pretrained(
|
||
|
|
repo,
|
||
|
|
return_dict=True,
|
||
|
|
torch_dtype=torch.float16,
|
||
|
|
device_map='auto'
|
||
|
|
)
|
||
|
|
OpenOrca_tokenizer = AutoTokenizer.from_pretrained(repo)
|
||
|
|
```
|
||
|
|
|
||
|
|
---
|