w601sxs/pythia-70m-instruct-orca-chkpt-64000

To use, do:

from peft import PeftModel, PeftConfig
from transformers import AutoTokenizer
ref_model = AutoModelForCausalLM.from_pretrained("EleutherAI/pythia-70m-deduped-v0", torch_dtype=torch.bfloat16)
peft_model_id = "w601sxs/pythia-70m-instruct-orca-chkpt-64000"

config = PeftConfig.from_pretrained(peft_model_id)
model = PeftModel.from_pretrained(ref_model, peft_model_id)
tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)

model = model.to('cuda:0')
model.eval()


inputs = tokenizer(prompt, return_tensors="pt")

with torch.no_grad():
    outputs = model.generate(input_ids=inputs["input_ids"].to("cuda"), max_new_tokens=10)
    print(tokenizer.batch_decode(outputs.detach().cpu().numpy(), skip_special_tokens=True)[0]

Prompt format

context: < ... >
question: < ... >
answer: < ... >

For e.g.

context: <You are an AI assistant. User will you give you a task. Your goal is to complete the task as faithfully as you can. While performing the task think step-by-step and justify your steps.>
 question: <Here is some data: The Rice Boat eatType restaurant; The Rice Boat food Fast food; The Rice Boat familyFriendly yes; The Rice Boat near Express by Holiday Inn.

Write a sentence that describes this data:>
 answer: <