chenbhao
/

omni

training-framework

Model card Files Files and versions

omni / docs /interview

136 kB

Ctrl+K

Ctrl+K

3 contributors

History: 9 commits

chenbhao's picture

Update docs: challenges.md (training bias + scene text approach), vam/multimodal/trainers docs sync

2460459 5 days ago

challenges.md
12.9 kB
Update docs: challenges.md (training bias + scene text approach), vam/multimodal/trainers docs sync 5 days ago
full_sft.md
19 kB
refactor: reorganize dataset into lm/vlm/vam subdirectories 6 days ago
index.md
4.35 kB
Revert camera prompt changes; add NaN guard and CUDA sync; add engineering challenges doc 5 days ago
inference.md
12.4 kB
refactor(docs): restructure docs with modular organization 6 days ago
moe.md
12.6 kB
refactor(docs): restructure docs with modular organization 6 days ago
multimodal.md
15.1 kB
refactor(docs): restructure docs with modular organization 6 days ago
pretrain.md
17.3 kB
reorganize configs into subdirs lm/vlm/vam, remove runs/, add interview docs for pretrain & full_sft 6 days ago
tokenizer.md
13.5 kB
docs: move tokenizer.md into interview/ 6 days ago
training-systems.md
13.4 kB
refactor(docs): restructure docs with modular organization 6 days ago
transformer-arch.md
15.8 kB
docs: update README and docs for model_type rename, RoPE guard fix, config paths 6 days ago