diff --git a/examples/tiny-llama/pretrain.yml b/examples/tiny-llama/pretrain.yml index e501dcb8e..010a1608a 100644 --- a/examples/tiny-llama/pretrain.yml +++ b/examples/tiny-llama/pretrain.yml @@ -9,9 +9,9 @@ strict: false max_steps: 200 pretraining_dataset: - path: c4 - name: en - type: pretrain + - path: allenai/c4 + name: en + type: pretrain dataset_prepared_path: val_set_size: 0.0 output_dir: ./outputs/model-out