We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 4797948 commit 056ff4aCopy full SHA for 056ff4a
recipes/A5000_24GB_x8/i18n-ja-wikipedia-step-10k.yaml
@@ -2,7 +2,8 @@ target_task: tasks/i18n/ja.md
2
base_model_id: TinyLlama/TinyLlama-1.1B-intermediate-step-715k-1.5T
3
model_name: TinyLlama-1.5T-ja-wikipedia-step-10k
4
output_base_dir: /data/output
5
-dataset_id: izumi-lab/wikipedia-ja-20230720
+dataset_id: wikimedia/wikipedia
6
+dataset_load_config: 20231101.ja
7
dataset_input_field_name: text
8
dataset_train_split_seed: 42
9
dataset_train_split_test_size: 0.2
0 commit comments