synthsumm
Collection
generalist summarizers trained on curated long docs + synthetic LLM summaries • 5 items • Updated
How to use pszemraj/pegasus-x-large-book_synthsumm-bf16 with Transformers:
# Use a pipeline as a high-level helper
# Warning: Pipeline type "summarization" is no longer supported in transformers v5.
# You must load the model directly (see below) or downgrade to v4.x with:
# 'pip install "transformers<5.0.0'
from transformers import pipeline
pipe = pipeline("summarization", model="pszemraj/pegasus-x-large-book_synthsumm-bf16") # Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("pszemraj/pegasus-x-large-book_synthsumm-bf16")
model = AutoModelForSeq2SeqLM.from_pretrained("pszemraj/pegasus-x-large-book_synthsumm-bf16")this is just a new repo in bf16 (training precision). refer to original repo for details https://huggingface.co/pszemraj/pegasus-x-large-book_synthsumm
Fine-tuned on a synthetic dataset of curated long-context text and GPT-3.5-turbo-1106 summaries spanning multiple domains + "random" long-context examples from pretraining datasets
Try it: gradio demo | example outputs .md (gauntlet) | code for free HF inference api
Base model
google/pegasus-x-large