metadata

dataset_info:
  dataset_size: 447310289.2648228
  download_size: 178540815
  features:
    - name: input_ids
      sequence: int32
    - name: attention_mask
      sequence: int8
    - name: labels
      sequence: int64
    - dtype: string
      name: preview
    - dtype: int64
      name: length
    - dtype: int64
      name: messages_count
  splits:
    - name: train
      num_bytes: 446343976.5148228
      num_examples: 305958
    - name: test
      num_bytes: 966312.75
      num_examples: 195

zh-tw-pythia-ta8000-v1-e1-tr_sg-302-c1024

This dataset is a part of the zh-tw-llm project.

Tokenizer: zh-tw-pythia-tokenizer-a8000-v1
Built with: translations, sharegpt
Rows: train 305958, test 195
Max length: 1024

Full config:

{"build_with": ["translations", "sharegpt"], "preview_length": 128, "translations_settings": {"source_dataset": "zetavg/coct-en-zh-tw-translations-twp-300k", "lang_1_key": "en", "lang_2_key": "ch", "templates": ["English: {lang_1}\nChinese: {lang_2}", "Chinese: {lang_2}\nEnglish: {lang_1}"], "use_template": "random", "rows_limit": 300000, "test_size": 100, "test_split_seed": 42}, "sharegpt_settings": {"source_dataset": "zetavg/ShareGPT-Processed", "train_on_inputs": false, "languages": [{"en": 0.4}, "zh_Hant"], "rows_limit": 8000, "test_size": 0.02, "test_split_seed": 42, "test_rows_limit": 100}}