Spaces:

zhan1206
/

fusion-llm-demo

Running

App Files Files Community

fusion-llm-demo

5.81 MB

Ctrl+K

Ctrl+K

3 contributors

History: 34 commits

zhan1206

Train: Add 10-step training test (loss decreases steadily)

04c7011 about 1 month ago

.github
CI/CD: Add GitHub Actions workflow for syntax check, unit tests, and integration tests about 1 month ago
configs
fix(v9): resolve 3 fatal + 1 serious + 1 moderate defects from external audit about 1 month ago
data
feat: train SentencePiece tokenizer + update README about 1 month ago
data_pipeline
fix(v12): resolve 6 defects from v11 audit (F-NEW-6, S-NEW-5, M-NEW-5/6, MI-NEW-4/5/6) about 1 month ago
evaluation
Feat: Add quantization tool for complete quantization workflow about 1 month ago
inference
Fix M-NEW-10: Use tokenizer encoding in _load_dataset instead of byte encoding about 1 month ago
models
fix(v12): resolve 6 defects from v11 audit (F-NEW-6, S-NEW-5, M-NEW-5/6, MI-NEW-4/5/6) about 1 month ago
scripts
fix(v12): resolve 6 defects from v11 audit (F-NEW-6, S-NEW-5, M-NEW-5/6, MI-NEW-4/5/6) about 1 month ago
tests
Test: Fix training basic test (torch.optim not torch.nn) about 1 month ago
tokenizers
feat: train SentencePiece tokenizer + update README about 1 month ago
train
Train: Add 10-step training test (loss decreases steadily) about 1 month ago
.gitignore

749 Bytes
chore: 添加 Windows 保留名到 .gitignore（避免 NUL 等文件被追踪） about 1 month ago
CONTRIBUTING.md

5.67 kB
feat: 初始化 Fusion 项目 - 六边形开源大模型 about 1 month ago
LICENSE

10.8 kB
chore: replace author real name with GitHub username about 1 month ago
README.md

5.41 kB
fix: v5 remove empty tokenizer.json, unify think tokens, integrate tokenizer into training pipeline about 1 month ago
requirements.txt

714 Bytes
fix: v3 comprehensive defect repair about 1 month ago