`min_length` parameter doesn't work

[ghost]

System Info

8 x 40 GB A100s Llama-3 70B Instruct, bf16 TP-8 TensorRT-LLM 0.9.0 + Triton 24.04

Who can help?

@byshiue @schetlur-nv

Information

• [x] The official example scripts
• [ ] My own modified scripts

Tasks

• [x] An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• [ ] My own task or dataset (give details below)

Reproduction

Set min_length to a high value (~512) and ask for a short answer in the prompt.`

Expected behavior

512 tokens returned.

actual behavior

Few tokens returned.

additional notes

N/A