NVIDIA-NeMo / NVIDIA-NeMo/RL

All nightly moonlight tests failing with sentencepiece error

Open
#2,764 1 comment 0 reactions 1 assignee Assigned to @ahmadki View on GitHub
bug
Dominant language
Python
Stars
2k
Forks
561
Avg merge
4d 5h
Merged PRs (30d)
145

Description

**Describe the bug**
All nightly moonlight tests failing with error despite sentenepiece being in direct rl dependencies:

```
ValueError: Couldn't instantiate the backend tokenizer from one of:
(1) a `tokenizers` library serialization file,
(2) a slow tokenizer instance to convert or
(3) an equivalent slow tokenizer class to instantiate and convert.
You need to have sentencepiece or tiktoken installed to convert a slow tokenizer to a fast one.

```
**Steps/Code to reproduce bug**

Tests failing off main branch:
llm_grpo_moonlight_16ba3b_4n8g_megatron_fp8_e2e
llm_grpo_moonlight_16ba3b_4n8g_megatron
llm_grpo_moonlight_16ba3b_4n8g_megatron_tq_simple etc.

**Additional context**

Failures after PR https://github.com/NVIDIA-NeMo/RL/pull/2489 merged

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.