# Runtime artifacts
runs/
logs/
checkpoints/
outputs/
wandb/
tensorboard/
*.log
*.pt
*.ckpt
*.safetensors
*.bin
*.core
core.*

# Large generated pretraining corpus: keep code/manifests in git, keep corpus local.
dataset/pretrain/data/*
!dataset/pretrain/data/.gitkeep

# Generated Megatron indexed validation data and GPTDataset caches.
# Rebuild from dataset/val/data/heldout_2p8k_packed_text.jsonl when needed.
dataset/val/megatron_8192_glm52/

# Python/cache/editor
__pycache__/
*.py[cod]
.DS_Store
.env
.venv/
