{ "apigen-mt_5k": { "file_name": "apigen-mt_5k.json", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "sampled20new_fix2_fix2_toollist_remove_sys": { "file_name": "sampled20new_fix2_fix2_toollist_remove_sys_Anyscale.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "sampled5_new_fix3_toollist_remove_sys": { "file_name": "sampled5_new_fix3_toollist_remove_sys_Anyscale.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "glaive_toolcall_zhtw_1k": { "file_name": "glaive_toolcall_zhtw_1k.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "glaive_toolcall_zh_tw_new_1k": { "file_name": "glaive_toolcall_zh_tw_new_1k.json", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "sampled20new_fix2_fix2_toollist_remove_sys_Anyscale_no_tail": { "file_name": "sampled20new_fix2_fix2_toollist_remove_sys_Anyscale_no_tail.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "sampled20new_fix2_fix2_toollist_remove_sys_Anyscale_no_tail_dropped": { "file_name": "sampled20new_fix2_fix2_toollist_remove_sys_Anyscale_no_tail_dropped.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "sampled5_new_fix3_toollist_remove_sys_Anyscale_no_tail_dropped": { "file_name": "sampled5_new_fix3_toollist_remove_sys_Anyscale_no_tail_dropped.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_zh_tw_function_mix_sharegpt": { "file_name": "multi_turn_zh_tw_function_mix_sharegpt.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_param_zh_tw_function_mix_sharegpt": { "file_name": "multi_turn_miss_param_zh_tw_function_mix_sharegpt.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_func_zh_tw_function_mix_sharegpt": { "file_name": "multi_turn_miss_func_zh_tw_function_mix_sharegpt.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "irrelevance_zh_tw_sharegpt3000": { "file_name": "irrelevance_zh_tw_sharegpt3000.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "apigen_zhtwV3_remove_sys_sharegpt": { "file_name": "apigen_zhtwV3_remove_sys_sharegpt.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_func_zh_tw_function_mix_sharegpt500_turn": { "file_name": "multi_turn_miss_func_zh_tw_function_mix_sharegpt500_turn.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_param_zh_tw_function_mix_sharegpt500_turn": { "file_name": "multi_turn_miss_param_zh_tw_function_mix_sharegpt500_turn.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_zh_tw_function_mix_sharegpt500_turn": { "file_name": "multi_turn_zh_tw_function_mix_sharegpt500_turn.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_func_zh_tw_function_mix_sharegpt_clean1": { "file_name": "multi_turn_miss_func_zh_tw_function_mix_sharegpt_clean1.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "multi_turn_miss_param_zh_tw_function_mix_sharegpt_clean1": { "file_name": "multi_turn_miss_param_zh_tw_function_mix_sharegpt_clean1.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "pythonic_eval": { "file_name": "pythonic_eval.jsonl", "formatting": "sharegpt", "columns": { "messages": "conversations", "tools": "tools" } }, "apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized": { "file_name": "apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" }, "irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized": { "file_name": "irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_miss_func_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_func_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_miss_para_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_para_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_zh_tw_function_mix500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" }, "irrelevance_zh_tw3000_llama3.1_pretokenized": { "file_name": "irrelevance_zh_tw3000_llama3.1_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_miss_func_zh_tw_function_mix500_turn_llama3.1_pretokenized": { "file_name": "multi_turn_miss_func_zh_tw_function_mix500_turn_llama3.1_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_miss_para_zh_tw_function_mix500_turn_llama3.1_pretokenized": { "file_name": "multi_turn_miss_para_zh_tw_function_mix500_turn_llama3.1_pretokenized.jsonl", "formatting": "tokenized" }, "multi_turn_zh_tw_function_mix500_turn_llama3.1_pretokenized": { "file_name": "multi_turn_zh_tw_function_mix500_turn_llama3.1_pretokenized.jsonl", "formatting": "tokenized" }, "apigen_zhtwV3_remove_sys_llama3.1_pretokenized": { "file_name": "apigen_zhtwV3_remove_sys_llama3.1_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_miss_param_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_param_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_miss_func_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_func_zh_tw_function_mix_v5_700_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "new_irrelevance_zh_tw_oss3000_oss_gpt_oss_20b_pretokenized": { "file_name": "new_irrelevance_zh_tw_oss3000_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_miss_func_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_func_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_miss_param_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_miss_param_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "multi_turn_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized": { "file_name": "multi_turn_zh_tw_function_mix_v5_500_turn_oss_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "mistral_24b_final_train_base": { "file_name": "mistral_24b_final_train_base.jsonl", "formatting": "tokenized" } , "mistral_24b_final_train_miss_func": { "file_name": "mistral_24b_final_train_miss_func.jsonl", "formatting": "tokenized" } , "mistral_24b_final_train_miss_param": { "file_name": "mistral_24b_final_train_miss_param.jsonl", "formatting": "tokenized" } , "mistral_24b_final_train_new_apigen_zhtwV3_remove_sys": { "file_name": "mistral_24b_final_train_new_apigen_zhtwV3_remove_sys.jsonl", "formatting": "tokenized" } , "mistral_24b_final_train_new_irrelevance": { "file_name": "mistral_24b_final_train_new_irrelevance.jsonl", "formatting": "tokenized" } , "new_clean_base_turn_gpt_oss_20b_pretokenized": { "file_name": "new_clean_base_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "new_clean_miss_func_turn_gpt_oss_20b_pretokenized": { "file_name": "new_clean_miss_func_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "new_clean_misspara_turn_gpt_oss_20b_pretokenized": { "file_name": "new_clean_misspara_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "new_clean_base_mistral_24b_final_train": { "file_name": "new_clean_base_mistral_24b_final_train.jsonl", "formatting": "tokenized" } , "new_clean_miss_func_mistral_24b_final_train": { "file_name": "new_clean_miss_func_mistral_24b_final_train.jsonl", "formatting": "tokenized" } , "new_clean_misspara_mistral_24b_final_train": { "file_name": "new_clean_misspara_mistral_24b_final_train.jsonl", "formatting": "tokenized" } , "all_dupcleaned_data_turn_gpt_oss_20b_pretokenized": { "file_name": "all_dupcleaned_data_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_special_token_done_all_dupcleaned_data_turn_gpt_oss_20b_pretokenized": { "file_name": "fix_special_token_done_all_dupcleaned_data_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_special_token_done_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized": { "file_name": "fix_special_token_done_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_special_token_done_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized": { "file_name": "fix_special_token_done_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_done_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized": { "file_name": "fix_done_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_done_dupcleaned_data_turn_gpt_oss_20b_pretokenized": { "file_name": "fix_done_dupcleaned_data_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "fix_done_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized": { "file_name": "fix_done_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "aug_all_dupcleaned_data_random_para_turn_gpt_oss_20b_pretokenized": { "file_name": "aug_all_dupcleaned_data_random_para_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "aug_data_random_para_turn_gpt_oss_20b_pretokenized": { "file_name": "aug_data_random_para_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "lenfiltered_aug_all_dupcleaned_data_random_para_turn_gpt_oss_20b_pretokenized": { "file_name": "lenfiltered_aug_all_dupcleaned_data_random_para_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "lenfiltered_aug_data_random_para_turn_gpt_oss_20b_pretokenized": { "file_name": "lenfiltered_aug_data_random_para_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "lenfiltered_fix_done_dupcleaned_data_turn_gpt_oss_20b_pretokenized": { "file_name": "lenfiltered_fix_done_dupcleaned_data_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "all_dupcleaned_with_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "all_dupcleaned_with_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "apigen_zhtwV3_remove_sys_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "apigen_zhtwV3_remove_sys_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v5_all_dupcleaned_with_simple_thinking_turn_thinking_gpt_oss_20b_pretokenized": { "file_name": "v5_all_dupcleaned_with_simple_thinking_turn_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v5_apigen_zhtwV3_remove_sys_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "v5_apigen_zhtwV3_remove_sys_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v5_new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "v5_new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v51_all_dupcleaned_data_turn_gpt_oss_20b_pretokenized": { "file_name": "v51_all_dupcleaned_data_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v51_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized": { "file_name": "v51_apigen_zhtwV3_remove_sys_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "v51_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized": { "file_name": "v51_new_irrelevance_zh_tw_oss3000_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "gemma4_26b_a4b_all_dupcleaned_with_simple_turn_it_pretokenized": { "file_name": "gemma4_26b_a4b_all_dupcleaned_with_simple_turn_it_pretokenized.jsonl", "formatting": "tokenized" } , "gemma4_26b_a4b_it_apigen_zhtwV3_remove_sys_simple_pretokenized": { "file_name": "gemma4_26b_a4b_it_apigen_zhtwV3_remove_sys_simple_pretokenized.jsonl", "formatting": "tokenized" } , "gemma4_26b_a4b_new_irrelevance_zh_tw_oss3000_simple_it_pretokenized": { "file_name": "gemma4_26b_a4b_new_irrelevance_zh_tw_oss3000_simple_it_pretokenized.jsonl", "formatting": "tokenized" } , "10000all_dupcleaned_gemma4_26b_a4b_it_pretokenized": { "file_name": "10000all_dupcleaned_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "10000apigen_zhtwV3_remove_sys_simple_gemma4_26b_a4b_it_pretokenized": { "file_name": "10000apigen_zhtwV3_remove_sys_simple_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "10000new_irrelevance_zh_tw_oss3000_simple_gemma4_26b_a4b_it_pretokenized": { "file_name": "10000new_irrelevance_zh_tw_oss3000_simple_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "6144turn_all_dupcleaned_delete_much_tool_gemma4_26b_a4b_it_pretokenized": { "file_name": "6144turn_all_dupcleaned_delete_much_tool_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "trans5_empty_cot_6144turn_all_dupcleaned_delete_much_tool_gemma4_26b_a4b_it_pretokenized": { "file_name": "trans5_empty_cot_6144turn_all_dupcleaned_delete_much_tool_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "trans5_empty_cot_6144turn_apigen_zhtwV3_remove_sys_gemma4_26b_a4b_it_pretokenized": { "file_name": "trans5_empty_cot_6144turn_apigen_zhtwV3_remove_sys_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "trans5_empty_cot_6144turn_new_irrelevance_zh_tw_oss3000_gemma4_26b_a4b_it_pretokenized": { "file_name": "trans5_empty_cot_6144turn_new_irrelevance_zh_tw_oss3000_gemma4_26b_a4b_it_pretokenized.jsonl", "formatting": "tokenized" } , "all_new_v5_apigen_gptossver_id_to_str_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "all_new_v5_apigen_gptossver_id_to_str_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "all_new_v5_gptossver_all_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "all_new_v5_gptossver_all_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "all_new_v5_new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "all_new_v5_new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gpt_oss_20b_pretokenized": { "file_name": "apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "gptoss_ver_aug_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "gptoss_ver_aug_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "5120trans5_empty_cot_apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gemma4_31B_it_pretokenized": { "file_name": "5120trans5_empty_cot_apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } , "5120trans5_empty_cot_apigen_zhtwV3_remove_sys_simple_thinking_gemma4_31B_it_pretokenized": { "file_name": "5120trans5_empty_cot_apigen_zhtwV3_remove_sys_simple_thinking_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } , "5120trans5_empty_cot_new_irrelevance_zh_tw_oss3000_simple_thinking_gemma4_31B_it_pretokenized": { "file_name": "5120trans5_empty_cot_new_irrelevance_zh_tw_oss3000_simple_thinking_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } , "5120trans5_empty_cot_turn_all_dupcleaned_with_simple_thinking_delete_much_tool_gemma4_31B_it_pretokenized": { "file_name": "5120trans5_empty_cot_turn_all_dupcleaned_with_simple_thinking_delete_much_tool_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } , "10240apigen_gptossver_id_to_str_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "10240apigen_gptossver_id_to_str_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gpt_oss_20b_pretokenized": { "file_name": "10240apigen_mt_messages_series_zhtw_dupcleaned_data_simple_cot_v1_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240gptoss_ver_aug_missfunc_only_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "10240gptoss_ver_aug_missfunc_only_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240gptoss_ver_aug_misspara_only_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "10240gptoss_ver_aug_misspara_only_both_style_and_tool_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240gptossver_miss_func_only_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "10240gptossver_miss_func_only_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240gptossver_miss_para_only_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized": { "file_name": "10240gptossver_miss_para_only_dupcleaned_data_simple_thinking_turn_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "10240new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized": { "file_name": "10240new_irrelevance_zh_tw_oss3000_simple_thinking_gpt_oss_20b_pretokenized.jsonl", "formatting": "tokenized" } , "5120base_and_miss_para_dupcleaned_with_simple_thinking_delete_much_tool_turn_gemma4_31B_it_pretokenized": { "file_name": "5120base_and_miss_para_dupcleaned_with_simple_thinking_delete_much_tool_turn_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } , "5120miss_func_dupcleaned_with_simple_thinking_delete_much_tool_turn_gemma4_31B_it_pretokenized": { "file_name": "5120miss_func_dupcleaned_with_simple_thinking_delete_much_tool_turn_gemma4_31B_it_pretokenized.jsonl", "formatting": "tokenized" } }