Model Description
This is model is a finetune of Llama-2-7b on the PRM8000K dataset. It's a POC, the model was training on a subset of our processed dataset. The model can answer math questions in a step-by-step manner.
How to Get Started with the Model
Use the code below to get started with the model.
import transformers
import torch
model_id = "FranckArmand/llama-2-7b-chat-hf-math-step-by-step"
# Run text generation pipeline with our next model
prompt = "Solve over the integers: $$2^a+4^b+8^c=328.$$"
pipe = transformers.pipeline(task="text-generation", model=model_id, model_kwargs={"torch_dtype": torch.bfloat16}, device_map="auto")
result = pipe(f"<s>[INST] {prompt} [/INST]")
print(result[0]['generated_text'])
- Downloads last month
- 27