| --- |
| tags: |
| - merge |
| - mergekit |
| - lazymergekit |
| - microsoft/codebert-base |
| - EleutherAI/gpt-neo-x-20b |
| - openai/codex |
| - bigscience/bloom |
| - google/jurassic-1-jumbo |
| - google/t5-v1_1-large |
| - facebook/bart-large |
| base_model: |
| - microsoft/codebert-base |
| - EleutherAI/gpt-neo-x-20b |
| - openai/codex |
| - bigscience/bloom |
| - google/jurassic-1-jumbo |
| - google/t5-v1_1-large |
| - facebook/bart-large |
| --- |
| |
| # code-slerp |
|
|
| code-slerp is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): |
| * [microsoft/codebert-base](https://huggingface.co/microsoft/codebert-base) |
| * [EleutherAI/gpt-neo-x-20b](https://huggingface.co/EleutherAI/gpt-neo-x-20b) |
| * [openai/codex](https://huggingface.co/openai/codex) |
| * [bigscience/bloom](https://huggingface.co/bigscience/bloom) |
| * [google/jurassic-1-jumbo](https://huggingface.co/google/jurassic-1-jumbo) |
| * [google/t5-v1_1-large](https://huggingface.co/google/t5-v1_1-large) |
| * [facebook/bart-large](https://huggingface.co/facebook/bart-large) |
|
|
| ## 🧩 Configuration |
|
|
| ```yaml |
| slices: |
| - sources: |
| - model: microsoft/codebert-base |
| layer_range: [0, 32] |
| - model: EleutherAI/gpt-neo-x-20b |
| layer_range: [0, 32] |
| - model: openai/codex |
| layer_range: [0, 32] |
| - model: bigscience/bloom |
| layer_range: [0, 32] |
| - model: google/jurassic-1-jumbo |
| layer_range: [0, 32] |
| - model: google/t5-v1_1-large |
| layer_range: [0, 32] |
| - model: facebook/bart-large |
| layer_range: [0, 32] |
| merge_method: slerp |
| base_model: microsoft/codebert-base |
| parameters: |
| t: |
| - filter: self_attn |
| value: [0, 0.5, 0.3, 0.7, 1] |
| - filter: mlp |
| value: [1, 0.5, 0.7, 0.3, 0] |
| - value: 0.5 |
| dtype: bfloat1 |
| ``` |
|
|
| ## 💻 Usage |
|
|
| ```python |
| !pip install -qU transformers accelerate |
| |
| from transformers import AutoTokenizer |
| import transformers |
| import torch |
| |
| model = "Or4cl3-1/code-slerp" |
| messages = [{"role": "user", "content": "What is a large language model?"}] |
| |
| tokenizer = AutoTokenizer.from_pretrained(model) |
| prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) |
| pipeline = transformers.pipeline( |
| "text-generation", |
| model=model, |
| torch_dtype=torch.float16, |
| device_map="auto", |
| ) |
| |
| outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95) |
| print(outputs[0]["generated_text"]) |
| ``` |
