nagayama0706/coding_model

1---2tags:3- merge4- mergekit5- lazymergekit6- huggingface/CodeBERTa-language-id7- Sharathhebbar24/code_gpt28base_model:9- huggingface/CodeBERTa-language-id10- Sharathhebbar24/code_gpt211license: apache-2.012pipeline_tag: question-answering13---14 15# coding_model16 17coding_model is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):18* [huggingface/CodeBERTa-language-id](https://huggingface.co/huggingface/CodeBERTa-language-id)19* [Sharathhebbar24/code_gpt2](https://huggingface.co/Sharathhebbar24/code_gpt2)20 21## 🧩 Configuration22 23```yaml24slices:25  - sources:26      - model: huggingface/CodeBERTa-language-id27        layer_range: [0, 32]28      - model: Sharathhebbar24/code_gpt229        layer_range: [0, 32]30merge_method: slerp31base_model: huggingface/CodeBERTa-language-id32parameters:33  t:34    - filter: self_attn35      value: [0, 0.5, 0.3, 0.7, 1]36    - filter: mlp37      value: [1, 0.5, 0.7, 0.3, 0]38    - value: 0.539dtype: bfloat1640```41 42## 💻 Usage43 44```python45!pip install -qU transformers accelerate46 47from transformers import AutoTokenizer48import transformers49import torch50 51model = "nagayama0706/coding_model"52messages = [{"role": "user", "content": "What is a large language model?"}]53 54tokenizer = AutoTokenizer.from_pretrained(model)55prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)56pipeline = transformers.pipeline(57    "text-generation",58    model=model,59    torch_dtype=torch.float16,60    device_map="auto",61)62 63outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)64print(outputs[0]["generated_text"])65```