W-61/llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.05-margin-log

该数据集是从一个名为New-DPO的训练运行中导出的每一步的边缘摘要统计信息。包含了训练过程中的多个统计特征，如epoch、step、batch_size、mean、std、min、p10、median、p90、max、pos_frac等，以及每个步骤的样本边缘和保存的完整边缘数组路径。数据集来源于特定的模型训练运行，使用了Anthropic/hh-rlhf数据集进行混合训练。

Per-step margin summary statistics exported from a New-DPO training run. Includes various statistical features during training such as epoch, step, batch_size, mean, std, min, p10, median, p90, max, pos_frac, as well as per-example margins for the effective batch on each logged step and optional paths to saved full margin arrays. The dataset originates from a specific model training run using the Anthropic/hh-rlhf dataset for mixed training.