Pretraining example from readme fails in Colab

Nobody has claimed this yet.

Running the pretraining example from GitHub fails when run in Google Colab.

```
!pip install 'litgpt[all]'

!mkdir -p custom_texts
!curl https://www.gutenberg.org/cache/epub/24440/pg24440.txt --output custom_texts/book1.txt
!curl https://www.gutenberg.org/cache/epub/26393/pg26393.txt --output custom_texts/book2.txt

# 1) Download a tokenizer
!litgpt download \
--repo_id EleutherAI/pythia-160m \
--tokenizer_only True

# 2) Pretrain the model
!litgpt pretrain \
--model_name pythia-160m \
--tokenizer_dir checkpoints/EleutherAI/pythia-160m \
--data TextFiles \
--data.train_data_path "custom_texts/" \
--train.max_tokens 10_000_000 \
--out_dir out/custom-model

# 3) Chat with the model
!litgpt chat \
--checkpoint_dir out/custom-model/final
```

@awaelchli

Start with the pretraining example in the README and run its Colab commands, including installation, tokenizer download, pretraining, and chat. Capture the failing step and error, then verify a documented fix by running the complete example successfully in Google Colab.

Written by the indexing model from the issue text.