metadata
library_name: transformers
license: llama3
datasets: mlabonne/chatml-OpenHermes2.5-dpo-binarized-alpha
tags:
- DPO
- Llama3-8B
This is an ExLlamaV2 quantized model in 4bpw of Muhammad2003/Llama3-8B-OpenHermes-DPO using the default calibration dataset.
Original Model card:
Llama3-8B-OpenHermes-DPO
Llama3-8B-OpenHermes-DPO is DPO-Finetuned model of Llama3-8B, on the OpenHermes-2.5 preference dataset using QLoRA.
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Muhammad2003/Llama3-8B-OpenHermes-DPO"
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"])