BERT-Base (dynamic_rank) — NeuralOps Compressed & LoRA-Healed Model Zoo

This repository provides a physically compressed and LoRA-healed checkpoint of google-bert/bert-base-uncased fine-tuned on GLUE SST-2, compressed using the dynamic_rank (dynamic_entropy_svd_factory) engine from the NeuralOps Library at Pareto-selected rank k=128 (2.39x compression ratio, -58.2% parameter reduction).

Compression & Architecture Profile

dynamic_entropy_svd_factory: Dynamic Entropy SVD factory function.

Property	Specification
Registry Key (`COMPRESSOR_REGISTRY`)	`dynamic_rank`
NeuralOps Class & Family	`dynamic_entropy_svd_factory` (`svd`)
Source Implementation	`neuralops/compression/svd/dynamic_rank.py`
Base Architecture	`google-bert/bert-base-uncased` (`109,483,778` parameters, 12 layers, $D=768$)
Showcase Rank (`k`)	`128` (`LoRA Recovery Applied: True`)
Compressed Parameter Count	`45,782,786` (`-58.18%` reduction, `2.391x` physical compression ratio)
Teacher Baseline SST-2 Acc	`91.63%`
Showcase Final SST-2 Acc	`73.62%` (Zero-Shot: `52.41%`, Parity Delta: `-18.01%`)

Multi-Rank Pareto Evaluation (`k=256, 128, 64, 32`)

Evaluated physically on NVIDIA GeForce RTX 3090 (24GB) on the GLUE SST-2 validation split (872 samples). Whenever zero-shot compression drops accuracy by > 2.0% alongside > 1.10x reduction, NeuralOps applies targeted r=8 LoRA healing (50 steps, lr=2e-4, 45s budget):

Rank (`k`)	Physical Ratio	Compressed Params (Reduction)	Zero-Shot Acc	Healed Acc	Parity vs Teacher (`91.63%`)
`256`	`2.39x`	`45,782,786` (`-58.2%`)	`52.41%`	`73.62%` (LoRA)	`-18.01%`
`128` ⭐	`2.39x`	`45,782,786` (`-58.2%`)	`52.41%`	`73.62%` (LoRA)	`-18.01%`
`64`	`3.08x`	`35,500,034` (`-67.6%`)	`51.15%`	`50.92%` (LoRA)	`-40.71%`
`32`	`3.65x`	`30,023,426` (`-72.6%`)	`50.11%`	`61.70%` (LoRA)	`-29.93%`

(⭐ indicates the serialized checkpoint packaged in this Hugging Face repository.)

Packaged Artifacts

pytorch_model.bin: Full serialized compressed BERT state dictionary (with low-rank factors / quantization buffers / LoRA healing adapters baked in).
config.json: Model configuration tagged with compression_method: "dynamic_rank" and compression_rank: 128.
compression_summary.json: Complete machine-readable telemetry across all evaluated ranks [256, 128, 64, 32].
Tokenizer Assets: Standard bert-base-uncased vocabulary (vocab.txt, tokenizer.json, tokenizer_config.json).

Quick Start (Reconstruction & Inference)

import torch
from huggingface_hub import hf_hub_download
from transformers import AutoTokenizer, BertForSequenceClassification
from neuralops.core.compression_pipeline import compress_model

repo_id = "tahamajs/bert-comp-dynamic-rank-sst2"
tokenizer = AutoTokenizer.from_pretrained(repo_id)

# Reconstruct compressed NeuralOps topology and load serialized compressed + healed weights
model = BertForSequenceClassification.from_pretrained("google-bert/bert-base-uncased", num_labels=2)
model = compress_model(model, {"compressors": [{"method": "dynamic_rank", "rank": 128}]})

ckpt_path = hf_hub_download(repo_id=repo_id, filename="pytorch_model.bin")
state_dict = torch.load(ckpt_path, map_location="cpu")
model.load_state_dict(state_dict, strict=False)
model.eval()

inputs = tokenizer("NeuralOps achieves strong compression without sacrificing downstream accuracy.", return_tensors="pt")
with torch.no_grad():
    logits = model(**inputs).logits
    pred = logits.argmax(dim=-1).item()
print("Predicted SST-2 Sentiment (0=Negative, 1=Positive):", pred)

Citation

@software{neuralops2026compressorzoo,
  title   = {NeuralOps: Multi-Rank Compressed BERT-SST2 Zoo with Pareto LoRA Recovery},
  author  = {Majlesi, Taha},
  year    = {2026},
  url     = {https://github.com/tahamajs/NeuralOps}
}