---
license: mit
datasets:
- nlpai-lab/kullm-v2
base_model:
  - LDCC/LDCC-SOLAR-10.7B
  - realPCH/kosolra-kullm
language:
- ko
- en
tags:
- merge
- mergekit
---

### Model Strategy
We merged following two model with SLERP method using [mergekit](https://github.com/arcee-ai/mergekit) library.
- [realPCH/kosolra-kullm](https://huggingface.co/realPCH/kosolra-kullm)
- [LDCC/LDCC-SOLAR-10.7B](https://huggingface.co/LDCC/LDCC-SOLAR-10.7B)

which are both based on [yanolja/KoSOLAR-10.7B-v0.1](https://huggingface.co/yanolja/KoSOLAR-10.7B-v0.1-deprecated).

### Run the model
```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "c1park/kosolra-kullm-LDCC-merge"
tokenizer = AutoTokenizer.from_pretrained(model_id)

model = AutoModelForCausalLM.from_pretrained(model_id)

text = "[INST] Put instruction here. [/INST]"
inputs = tokenizer(text, return_tensors="pt")

outputs = model.generate(**inputs, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

```