metadata
language: en
license: mit
tags:
- pruning
- sparse
- atropos
- code-generation
gpt2-xl-structured_pruning
This is a pruned version of gpt2-xl generated as part of the Atropos pruning exercise.
Pruning Details
- Strategy: structured_pruning
- Method: PyTorch magnitude-based unstructured pruning
- Generated by: Atropos pruning exercise scripts
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("gpt2-xl-structured_pruning")
tokenizer = AutoTokenizer.from_pretrained("gpt2-xl-structured_pruning")
About Atropos
Atropos estimates ROI for pruning and quantization optimizations in LLM deployments.
This model was pruned to validate Atropos projections against actual results.
Citation
If you use this model, please cite the original model and Atropos:
@software{atropos,
title = {Atropos: ROI Estimation for LLM Pruning},
year = {2026},
}