README.md

metadata

license: mit
language:
  - en

Introduction

MoMo-70B is trained via Supervised Fine-Tuning (SFT) using LoRA, with the QWEN-72B model as its base-model.
This is a Direct Preference Optimization(DPO) version of v1.8.6 , with several optimizations in hyperparameters.
Note that we did not exploit any form of weight merge.
For leaderboard submission, the trained weight is realigned for compatibility with llama.
MoMo-70B is trained using Moreh's MoAI platform, which simplifies the training of large-scale models, and AMD's MI250 GPU.

Details

Used Librarys

• torch
• peft

Used Datasets

• slimorca
• truthy
• orca_dpo_pairs
• No other dataset was used
• No benchmark test set or the training set are used
  • data contamination check result

Model	ARC	MMLU	TruthfulQA	GSM8K
V1.8.6(result < 0.1, %)	TBU	TBU	0.73	TBU

Used Environments

• AMD MI250 & MoAI platform
• Please visit https://moreh.io/product for more information about MoAI platform
• Or, contact us directly [email protected]

How to use

# pip install transformers==4.35.2
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("moreh/MoMo-70B-LoRA-V1.8.6")
model = AutoModelForCausalLM.from_pretrained(
    "moreh/MoMo-70B-LoRA-V1.8.6"
)