[BUG] Question on batch preparation in MMLU evaluation

The bug I met is similar to #203. I'm trying to reproduce the evaluation results of [ablation model trained on FineWeb](https://huggingface.co/HuggingFaceFW/ablation-model-fineweb-v1/tree/main), using LightEval of commit_id=a98210fd3a2d1e8bface1c32b72ebd5017173a4c.

The MMLU result of step-5000/10000/15000/19000/24000 (namely, 5 ckpts from the first 50b consumed tokens) are as below:

I don't know what causes this gap, when debugging I discover that:

The last token of the prepared_batch is missing. Does this mean the evaluation results of [fineweb blogpost](https://huggingface.co/spaces/HuggingFaceFW/blogpost-fineweb-v1#/) is inaccurate?

But when I delete `[:-1]` in https://github.com/huggingface/lighteval/blob/aaa8bbf705b6f090fb07ad36503f39b5e922a6df/src/lighteval/models/base_model.py#L851

The evaluation results became totally random guess for all ckpts. I suppose there are more lines to modify, or something else caused the gaps in my reproduction results.