jackf857/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.5-s_star-0.4-20260429-032138-margin
收藏资源简介:
该数据集是从New-DPO训练运行中导出的每一步的边界摘要统计信息。数据集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。数据来源于模型repo jackf856/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.5-s_star-0.4-20260429-032138,训练参数包括beta、f_divergence_type、f_alpha_divergence_coef、s_star、eta和q_t等。数据集混合器使用了HuggingFaceH4/ultrafeedback_binarized。
Per-step margin summary statistics exported from a New-DPO training run. The dataset includes various statistical features during the training process, such as epoch, step, batch_size, mean, std, min, p10, median, p90, max, pos_frac, sample, and npy. The data originates from the model repo jackf856/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.5-s_star-0.4-20260429-032138, with training parameters including beta, f_divergence_type, f_alpha_divergence_coef, s_star, eta, and q_t. The dataset mixer uses HuggingFaceH4/ultrafeedback_binarized.




