W-61/llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-5-margin-log

该数据集是从New-DPO训练运行中导出的每一步边缘统计摘要。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、sample和npy等。数据集来源于特定的训练运行，使用了特定的模型和训练参数，如beta、f_divergence_type、f_alpha_divergence_coef、s_star、eta和q_t等。数据集混合了Anthropic/hh-rlhf的数据。

Per-step margin summary statistics exported from a New-DPO training run. Includes various statistical features such as epoch, step, batch_size, mean, std, min, p10, median, p90, max, pos_frac, sample, and npy. The dataset originates from a specific training run with particular model and training parameters like beta, f_divergence_type, f_alpha_divergence_coef, s_star, eta, and q_t. The dataset mixes data from Anthropic/hh-rlhf.