# Morizeyao/GPT2-Chinese

Chinese version of GPT2 training code, using BERT tokenizer.

Repository: https://github.com/Morizeyao/GPT2-Chinese
Canonical: https://ross.abutalabs.com/products/gpt2-chinese
Language: Python
License: MIT
License Family: permissive
Topics: transformer, gpt-2, chinese, nlp, text-generation
Last push: 2024-04-25T09:14:25+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2652, "days_push": 860, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7597, forks 1681 (observed 2026-08-28T04:10:01.917569+00:00)

## What it is
A Python library providing GPT-2 training and text generation code tailored for Chinese, built on HuggingFace Transformers with BERT or BPE tokenizers. It includes pretrained Chinese models for poems, couplets, lyrics, classical Chinese, and general text generation.

## Use cases
- train a GPT-2 model on Chinese text corpora
- generate Chinese poems with a pretrained model
- generate Chinese news articles or novels
- fine-tune GPT-2 for Chinese dialogue generation
- generate classical Chinese or couplets
- train a character-level or BPE-level Chinese language model

## When to choose
- you need to train or fine-tune GPT-2 specifically on Chinese text
- you want ready-made pretrained Chinese GPT-2 models for poetry, lyrics, or couplets
- you want a simple PyTorch codebase to learn language model training

## When to avoid
- you need actively maintained state-of-the-art LLM tooling (the project is no longer updated)
- you want multilingual or non-Chinese language model training
- you need modern efficient training features like DeepSpeed or distributed training out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-training, nlp, machine-learning
- domain: large-language-models, machine-learning
- platform: python
- tags: gpt-2, chinese, text-generation, transformer, bert-tokenizer, pretrained-models, natural-language-processing

## Member repositories
- Morizeyao/GPT2-Chinese (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:01.917569+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:36:51.906735+00:00, confidence not recorded.
  - readme: https://github.com/Morizeyao/GPT2-Chinese (fetched 2026-08-28T04:10:01.917569+00:00, sha cc5326f3ac35)
- Data as of 2026-08-30T08:39:29.467469+00:00.