Views
No views yet
chatrl_checkpoints/d12 step 933.| File | Use |
|---|---|
model.safetensors | HF-layout weights (model.* keys) |
model_000933.pt | Resume training in nanochat |
*.gguf | llama.cpp inference (lukas-h/llama.cpp fork required) |
tokenizer.pkl | nanochat / GGUF tokenizer |
1export NANOCHAT_BASE_DIR=~/.cache/nanochat
2# copy model_*.pt + meta_*.json into chatsft_checkpoints/d12/ (or chatrl_checkpoints/d12/)
3python -m scripts.chat_cli -i rl -g d121./build/bin/llama-completion -m nanochat-d12-rl-bf16.gguf \
2 -p '<|user_start|>Hello<|user_end|><|assistant_start|>' --special -no-cnv -n 128