Crusadersk/gpt2-100m

GPT-2 100M

Custom-trained GPT-2 checkpoint with deliberate depth-width configuration for inference benchmarking research.

Created as part of the Banterhearts research program investigating benchmarking integrity for local LLM inference.

Architecture	GPT2LMHeadModel (MHA)
Parameters	100M
Config	n_embd=768, n_head=2, n_layer=8, n_inner=3072
Context length	1,024 tokens
Precision	FP32
Model size	367 MB
Vocab size	50,257

Purpose

Largest MHA model; used in cross-backend and compiler benchmarks.

These checkpoints are not general-purpose language models. They are deliberately sized scaling-study artifacts designed to isolate the effect of model depth vs width on GPU inference latency. The key finding: in the small-model GPU regime, layer depth (not parameter count) dominates latency, producing inversions where a 5M-parameter model can be 3.6x slower than a 25M-parameter model.

Source Technical Reports

Used in: TR117, TR120, TR126, TR147

TR	Role
TR117	Original cross-backend benchmark matrix (7 backends, 4 model groups)
TR126	Linux/Triton compiler validation with phase-separated measurement
TR147	Second-regime portability validation on RTX 6000 Ada

Design Rationale

The GPT-2 family (25M, 50M, 100M) uses a 2x3 factorial design:

Model	n_embd	n_layer	n_inner	Params	Design role
gpt2-25m	384	3	1,536	25M	Shallow, narrow
gpt2-50m	512	8	2,048	50M	Deep, medium width
gpt2-100m	768	8	3,072	100M	Deep, wide

All models use 2 attention heads (MHA, not GQA) to isolate architecture effects from attention-group structure. Dropout is set to 0.0 for deterministic inference measurement.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Crusadersk/gpt2-100m")
tokenizer = AutoTokenizer.from_pretrained("Crusadersk/gpt2-100m")

inputs = tokenizer("Hello", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=32, do_sample=False)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Compatibility

Framework	Supported
Transformers	Yes
torch.compile (Inductor)	Yes
Ollama	No (not GGUF format)
vLLM	Yes

Citation

@misc{banterhearts2026gpt2100m,
  title = {Custom GPT-2 Scaling Checkpoint (100M) for Inference Benchmarking Research},
  author = {Kadadekar, Sahil},
  year = {2026},
  url = {https://huggingface.co/Crusadersk/gpt2-100m},
  note = {Part of the Banterhearts research program. NeurIPS 2026 submission.}
}

Acknowledgments

This work is part of a 40-TR research program on consumer LLM deployment safety, conducted independently as pre-doctoral research. Full program details at github.com/Sahil170595/Banterhearts.