jingyaogong/minimind
🧠 Train a 64M-parameter LLM from scratch in just 2h!
Python★ 57,299+1,005 that day#5 on trending8.36% duplicated
44files
55,416lines
170,676tokens
81clones
4,631duplicated lines (8.36%)
13,622duplicated tokens (7.98%)
By format
| Format | Files | Lines | Clones | Duplicated lines | Duplication |
|---|---|---|---|---|---|
| python | 22 | 8,090 | 53 | 1,298 | 16.04% |
| markdown | 3 | 4,100 | 14 | 196 | 4.78% |
| text | 2 | 3,962 | 8 | 1,247 | 31.47% |
| bash | 11 | 4,434 | 5 | 806 | 18.18% |
| json | 4 | 34,628 | 1 | 1,084 | 3.13% |
| perl | 1 | 168 | 0 | 0 | 0% |
| txt | 1 | 34 | 0 | 0 | 0% |
Largest clones
- 95 lines · 1,570 tokens textREADME.md:1340–1434⇆README_en.md:1338–1432
- 277 lines · 544 tokens textREADME.md:1704–1980⇆README_en.md:1704–1980
- 14 lines · 495 tokens textREADME.md:1448–1461⇆README_en.md:1446–1459
- 15 lines · 494 tokens textREADME.md:1434–1448⇆README_en.md:1432–1446
- 56 lines · 493 tokens pythontrainer/train_full_sft.py:19–74⇆trainer/train_pretrain.py:19–73
- 323 lines · 474 tokens textREADME.md:1018–1340⇆README_en.md:1018–1338
- 18 lines · 356 tokens textREADME.md:1461–1478⇆README_en.md:1459–1476
- 33 lines · 315 tokens pythontrainer/train_distillation.py:211–243⇆trainer/train_pretrain.py:135–172
- 272 lines · 280 tokens textREADME.md:743–1014⇆README_en.md:743–1014
- 32 lines · 280 tokens pythontrainer/train_distillation.py:210–241⇆trainer/train_full_sft.py:135–166
Trending appearances
| Day | Rank | Stars | Files | Lines | Clones | Duplication | Commit |
|---|---|---|---|---|---|---|---|
| Sep 2, 2026 | #5 | 57,299 (+1,005) | 44 | 55,416 | 81 | 8.36% | 7a6fddd |
| Sep 1, 2026 | #5 | 56,735 (+495) | 44 | 55,416 | 81 | 8.36% | 7a6fddd |
Want this for your own project? Install jscpd and run jscpd .