Phind CodeLlama 34B V2

🚀 Phind-CodeLlama-34B-v2

We've fine-tuned Phind-CodeLlama-34B-v1 on high-quality programming data, achieving 73.8% pass@1 on HumanEval. It's the current state-of-the-art among open-source models and is instruction-tuned for easy use.

📚 Documentation

Model Details

This model is fine-tuned from Phind-CodeLlama-34B-v1 and attains 73.8% pass@1 on HumanEval. Phind-CodeLlama-34B-v2 is multi-lingual, proficient in Python, C/C++, TypeScript, Java, and more.

Dataset Details

We fine-tuned on a proprietary dataset of 1.5B tokens of high-quality programming problems and solutions. This dataset consists of instruction-answer pairs rather than code completion examples, making it structurally different from HumanEval. LoRA was not used; both models are a native finetune. We used DeepSpeed ZeRO 3 and Flash Attention 2 to train these models in 15 hours on 32 A100 - 80GB GPUs, with a sequence length of 4096 tokens.

📦 Installation

Make sure to install Transformers from the main git branch:

pip install git+https://github.com/huggingface/transformers.git

💻 Usage Examples

Basic Usage

This model accepts the Alpaca/Vicuna instruction format. For example:

### System Prompt
You are an intelligent programming assistant.

### User Message
Implement a linked list in C++

### Assistant
...

Advanced Usage

To reproduce our HumanEval results:

from transformers import AutoTokenizer, LlamaForCausalLM
from human_eval.data import write_jsonl, read_problems
from tqdm import tqdm

# initialize the model

model_path = "Phind/Phind-CodeLlama-34B-v2"
model = LlamaForCausalLM.from_pretrained(model_path, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_path)

# HumanEval helper

def generate_one_completion(prompt: str):
    tokenizer.pad_token = tokenizer.eos_token
    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=4096)

    # Generate
    generate_ids = model.generate(inputs.input_ids.to("cuda"), max_new_tokens=384, do_sample=True, top_p=0.75, top_k=40, temperature=0.1)
    completion = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
    completion = completion.replace(prompt, "").split("\n\n\n")[0]

    return completion

# perform HumanEval
problems = read_problems()

num_samples_per_task = 1
samples = [
    dict(task_id=task_id, completion=generate_one_completion(problems[task_id]["prompt"]))
    for task_id in tqdm(problems)
    for _ in range(num_samples_per_task)
]
write_jsonl("samples.jsonl", samples)

# run `evaluate_functional_correctness samples.jsonl` in your HumanEval code sandbox

🔧 Technical Details

• Hardware Type: 32x A100 - 80GB
• Hours used: 480 GPU - hours
• Cloud Provider: AWS
• Compute Region: us - east - 1

📄 License

The license for this model is llama2.

📊 Model Index

⚠️ Important Note

This model has undergone very limited testing. Additional safety testing should be performed before any real - world deployments.