🚀🚀 「大模型」2小时完全从0训练64M的小参数GPT！🌏 Train a 64M-parameter GPT from scratch in just 2h!
COMMITS
December 1, 2025
J
[feat] update yarn
jingyaogong committed
November 27, 2025
J
[feat] release memory
jingyaogong committed
November 19, 2025
J
[fix] ppo mask
jingyaogong committed
J
[fix] model attn_mask
jingyaogong committed
November 18, 2025
J
[fix] update model
jingyaogong committed
J
Merge pull request #536 from yuyu5333/fix/attn_forward
jingyaogong committed
Y
fix: attn_forwad when is_causal=True assert attn_mask is None
yuyu5333 committed
November 15, 2025
J
[fix] prompt length calculate
jingyaogong committed
November 10, 2025
J
Merge pull request #528 from wangzhaode/feat/add_mnn_support
jingyaogong committed
Y
[feat] add MNN support to README.
yanxing committed
November 7, 2025
J
[fix] model-name
jingyaogong committed
November 6, 2025
J
[fix] harmonize template
jingyaogong committed
J
[feat] clear cache
jingyaogong committed
November 2, 2025
J
[fix] harmonize template
jingyaogong committed
October 31, 2025
J
[feat] update import
jingyaogong committed
October 30, 2025
J
[feat] update readme
jingyaogong committed
J
[feat] update readme
jingyaogong committed
J
[feat] update readme
jingyaogong committed
J
[feat] update datasets
jingyaogong committed
J
[feat] update args
jingyaogong committed
J
[feat] add args
jingyaogong committed
October 29, 2025
J
[feat] update readme
jingyaogong committed
J
[fix] model device
jingyaogong committed
October 28, 2025
J
[feat] update trainer
jingyaogong committed
J
[feat] update trainer
jingyaogong committed
October 27, 2025
J
[feat] update readme
jingyaogong committed
October 26, 2025
J
[feat] update readme
jingyaogong committed
J
[feat] update eval-llm
jingyaogong committed
J
[feat] pause-training
jingyaogong committed
October 23, 2025
J
[feat] update readme
jingyaogong committed