d-matrix/gpt2

This is a d-Matrix functional reference of the GPT2 model family, with the following revisions:

• distilgpt2
• gpt2
• gpt2-medium
• gpt2-large
• gpt2-xl

The reference provides the following functional configurations:

Configuration	Explanation
BASELINE	a reference functionally equivalent to the original model
BASIC	all linear algebraic operands quantized to BFP16-64, and all other operations transformed to approximated kernel simulations

Usage

Install d-Matrix Dmx_Compressor first.

pip install dmx_compressor

The following is an example model and its evaluation.

from dmx.compressor.dmx import pipeline

pipe = pipeline(
    task="text-generation",
    model="d-matrix/gpt2",
    revision="gpt2-xl",  # see above for other variants
    dmx_config="BASELINE",  # see above for other variants
)

results = pipe.evaluate(
    metric="d-matrix/dmx_perplexity",
    dataset="wikitext",
    dataset_version="wikitext-2-raw-v1",
)

Evaluation results

• perplexity on penn_treebank

  Revision \ Configuration	BASELINE	BASIC
  distilgpt2	63.46	64.13
  gpt2	35.77	35.93
  gpt2-medium	27.06	27.10
  gpt2-large	23.03	23.04
  gpt2-xl	21.01	21.02

• perplexity on wikitext2

  Revision \ Configuration	BASELINE	BASIC
  distilgpt2	46.06	46.44
  gpt2	29.94	30.08
  gpt2-medium	21.71	21.73
  gpt2-large	19.42	19.43
  gpt2-xl	17.40	17.40

• perplexity on wikitext103

  Revision \ Configuration	BASELINE	BASIC
  distilgpt2	46.06	46.44
  gpt2	29.94	30.08
  gpt2-medium	21.71	21.73
  gpt2-large	19.43	19.43
  gpt2-xl	17.40	17.40