tzervas/bwsk-gpt2-small

BWSK GPT-2 Small

GPT-2 Small (124M params) trained in 6 variants (3 BWSK modes x 2 experiments) on WikiText-2 with full convergence training and early stopping.

This repo contains all model weights, configs, and training results in a single consolidated repository.

What is BWSK?

BWSK is a framework that classifies every neural network operation as S-type (information-preserving, reversible, coordination-free) or K-type (information-erasing, synchronization point) using combinator logic. This classification enables reversible backpropagation through S-phases to save memory, and CALM-based parallelism analysis.

Model Overview

Property	Value
Base Model	openai-community/gpt2
Architecture	Transformer (causal_lm)
Parameters	124M
Dataset	WikiText-2
Eval Metric	Perplexity

S/K Classification

Type	Ratio
S-type (information-preserving)	60.8%
K-type (information-erasing)	39.2%

Fine-tune Results

Mode	Final Loss	Val Perplexity	Test Perplexity	Peak Memory	Time	Epochs
Conventional	2.8851	18.59	18.07	5.7 GB	8.6m	5
BWSK Analyzed	3.0312	18.59	18.09	5.7 GB	8.7m	5
BWSK Reversible	2.8376	18.58	18.09	3.6 GB	10.5m	5

Memory savings (reversible vs conventional): 36.8%

From Scratch Results

Mode	Final Loss	Val Perplexity	Test Perplexity	Peak Memory	Time	Epochs
Conventional	4.9907	291.30	296.78	5.7 GB	8.7m	5
BWSK Analyzed	4.9437	289.78	292.92	5.7 GB	8.8m	5
BWSK Reversible	4.7981	293.44	299.27	3.6 GB	10.3m	5

Memory savings (reversible vs conventional): 36.8%

Repository Structure

├── README.md
├── results.json
├── finetune-conventional/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── finetune-bwsk-analyzed/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── finetune-bwsk-reversible/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-conventional/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-bwsk-analyzed/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json
├── scratch-bwsk-reversible/
│   ├── model.safetensors
│   ├── config.json
│   └── training_results.json

Usage

Load a specific variant:

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load fine-tuned conventional variant
model = AutoModelForCausalLM.from_pretrained(
    "tzervas/bwsk-gpt2-small", subfolder="finetune-conventional"
)
tokenizer = AutoTokenizer.from_pretrained(
    "tzervas/bwsk-gpt2-small", subfolder="finetune-conventional"
)

# Load from-scratch BWSK reversible variant
model = AutoModelForCausalLM.from_pretrained(
    "tzervas/bwsk-gpt2-small", subfolder="scratch-bwsk-reversible"
)

Training Configuration

Setting	Value
Optimizer	AdamW
LR (fine-tune)	5e-05
LR (from-scratch)	3e-04
LR Schedule	Cosine with warmup
Max Grad Norm	1.0
Mixed Precision	AMP (float16)
Early Stopping	Patience 3
Batch Size	4
Sequence Length	512

Links

• GitHub Repository
• Whitepaper
• Full Training Report

Citation

@software{zervas2026bwsk,
  author = {Zervas, Tyler},
  title = {BWSK: Combinator-Typed Neural Network Analysis},
  year = {2026},
  url = {https://github.com/tzervas/ai-s-combinator},
}

License

MIT