by ibm-research
Open source · 686k downloads · 18 likes
PowerMoE 3b is an advanced language model based on a sparse Mixture-of-Experts (sMoE) architecture with 3 billion parameters. Through its selective activation mechanism, it efficiently mobilizes 800 million parameters per token, delivering performance comparable to dense models twice its size. Trained on a blend of open-source and proprietary data, it excels in diverse tasks such as natural language multiple-choice, code generation, and mathematical reasoning. The model stands out for its computational efficiency and ability to compete with heavier architectures while reducing inference costs. Its innovative approach makes it a powerful tool for applications requiring both precision and speed.
PowerMoE-3B is a 3B sparse Mixture-of-Experts (sMoE) language model trained with the Power learning rate scheduler. It sparsely activates 800M parameters for each token. It is trained on a mix of open-source and proprietary datasets. PowerMoE-3B has shown promising results compared to other dense models with 2x activate parameters across various benchmarks, including natural language multi-choices, code generation, and math reasoning. Paper: https://arxiv.org/abs/2408.13359
Note: Requires installing HF transformers from source.
This is a simple example of how to use PowerMoE-3b model.
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # or "cpu"
model_path = "ibm/PowerMoE-3b"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
prompt = "Write a code to find the maximum value in a list of numbers."
# tokenize the text
input_tokens = tokenizer(prompt, return_tensors="pt")
# transfer tokenized inputs to the device
for i in input_tokens:
input_tokens[i] = input_tokens[i].to(device)
# generate output tokens
output = model.generate(**input_tokens, max_new_tokens=100)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# loop over the batch to print, in this example the batch size is 1
for i in output:
print(i)