RetNet-1B-Hybrid-XATL

Hybrid RetNet

This is a hybrid architecture between self-attention based Transformer and RetNet, where only the 2nd and middle layer is multi-head attention, and otherwise RetNet.

This is the model weight accompanying the paper Cross-Architecture Transfer Learning for Linear-Cost Inference Transformers, in which new Linear-Cost Inference models (e.g. RetNet) are not trained from scratch but transfer shared weight components from other PTLMs. The model's input/output embeddings, MLP weights, & Layer Norms has been transferred from pythia-1B. For more detail, please refer to the paper.

Model Details

Model Description

• Developed by: NucleusAI, Sehyun Choi
• Model type: RetNet & Transformer Hybrid

Model Sources

• Repository: RetNet-XATL
• Paper: Cross-Architecture Transfer Learning for Linear-Cost Inference Transformers

How to Get Started with the Model

Use the code below to get started with the model.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

torch.set_default_device("cuda")

model = AutoModelForCausalLM.from_pretrained("NucleusAI/RetNet-1B-Hybrid-XATL", torch_dtype="auto", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("NucleusAI/RetNet-1B-Hybrid-XATL", trust_remote_code=True)  # same as EleutherAI/pythia-1B

inputs = tokenizer("Hi there!", return_tensors="pt", return_attention_mask=False)

outputs = model.generate(**inputs, max_length=200)
text = tokenizer.batch_decode(outputs)[0]
print(text)

Training Data

The model has been trained with pile_dedup dataset, in favor of comparison with the same sized pythia models.