This model uses a context window of 8k. It was trained with the ChatML template.
Thanks to bartowski, solidrust, and LoneStriker for the quantized models.
OrpoLlama-4-8B outperforms Llama-3-8B-Instruct on the GPT4All and TruthfulQA datasets.
Evaluation performed using
LLM AutoEval, see the entire leaderboard
here.
mlabonne/OrpoLlama-3-8B-1k corresponds to a version of this model trained on 1K samples (you can see the parameters in
this article). The current version was trained on a full epoch.
You can find the experiment on W&B at
this address.
1!pip install -qU transformers accelerate
2
3from transformers import AutoTokenizer
4import transformers
5import torch
6
7model = "mlabonne/OrpoLlama-3-8B"
8messages = [{"role": "user", "content": "What is a large language model?"}]
9
10tokenizer = AutoTokenizer.from_pretrained(model)
11prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
12pipeline = transformers.pipeline(
13 "text-generation",
14 model=model,
15 torch_dtype=torch.float16,
16 device_map="auto",
17)
18
19outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
20print(outputs[0]["generated_text"])
