data-pack
llama3_prompt_baseline_learn_from_70b_data_n4_filter_2e6_bz32_pack8192_3epochwo_packing_v2_1_pt_ep1_sft_ep5_based_on_ourmodel_with_all_choosen_inst_data_20240725llama3_8b_learn_from_70b_data_n4_filter_2e6_bz32_pack8192_also_train_reward_2epochwo_packing_v2_1_pt_ep1_sft_ep4_based_on_ourmodel_with_all_choosen_inst_data_20240725wo_packing_v2_1_pt_ep1_sft_ep3_based_on_ourmodel_with_all_choosen_inst_data_20240725wo_packing_v2_1_pt_ep1_sft_ep2_based_on_ourmodel_with_all_choosen_inst_data_20240725wo_packing_v2_1_pt_ep1_sft_ep1_based_on_ourmodel_with_all_choosen_inst_data_20240725llama3_8b_learn_from_70b_data_n4_filter_2e6_bz32_pack8192_2epoch
