pmetal eval
Evaluate a model’s perplexity on a dataset to measure generation quality.
pmetal eval \ --model <MODEL> \ --dataset <DATASET> \ [OPTIONS]Examples
Section titled “Examples”# Evaluate perplexitypmetal eval \ --model Qwen/Qwen3-0.6B \ --dataset eval.jsonl
# Evaluate with LoRA adapterpmetal eval \ --model Qwen/Qwen3-0.6B \ --dataset eval.jsonl \ --lora ./output/lora_weights.safetensorsSee Also
Section titled “See Also”- pmetal train — Train a model
- pmetal infer — Run inference