nebius/gpt-oss-120b-Infinity-Instruct-0625

--- license: cc-by-4.0 task_categories: - text-generation language: - en configs: - config_name: default data_files: - split: train path: data/train-* dataset_info: features: - name: conversation list: - name: content dtype: string - name: role dtype: string - name: generated_message struct: - name: annotations dtype: 'null' - name: audio dtype: 'null' - name: content dtype: string - name: function_call dtype: 'null' - name: reasoning_content dtype: string - name: refusal dtype: 'null' - name: role dtype: string - name: tool_calls sequence: 'null' - name: finish_reason dtype: string splits: - name: train num_bytes: 6061060422 num_examples: 659358 download_size: 3572636672 dataset_size: 6061060422 --- # gpt-oss-120b-Infinity-Instruct-0625 ## Dataset Description This dataset is part of the LK-Speculators collection for speculative decoding research. It contains 660K prompt-response pairs designed for training draft models that are used alongside gpt-oss-120b as the target model. The dataset was created by generating responses to the prompts from [Infinity-Instruct-0625](https://huggingface.co/datasets/BAAI/Infinity-Instruct) with [openai/gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) at temperature=1. For more details on the training methodology and results, see our paper: [LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding](https://arxiv.org/abs/2602.23881). ## Dataset Structure - **Format**: parquet - **Rows**: 659,358 ## Usage ```python from datasets import load_dataset dataset = load_dataset("nebius/gpt-oss-120b-Infinity-Instruct-0625") ``` ## License The dataset is released under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) ## Citation ``` @misc{samarin2026lklosses, title = {LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding}, author = {Alexander Samarin and Sergei Krutikov and Anton Shevtsov and Sergei Skvortsov and Filipp Fisin and Alexander Golubev}, year = {2026}, eprint = {2602.23881}, archivePrefix = {arXiv}, primaryClass = {cs.LG}, url = {https://arxiv.org/abs/2602.23881} } ```许可证：CC BY 4.0 任务类别： - 文本生成（text-generation） 语言： - 英语（en） 配置项： - 配置名称：default 数据文件： - 拆分集：train 路径：data/train-* 数据集信息： 特征： - 名称：conversation 列表类型： - 名称：content 数据类型：字符串 - 名称：role 数据类型：字符串 - 名称：generated_message 结构体类型： - 名称：annotations 数据类型：空值 - 名称：audio 数据类型：空值 - 名称：content 数据类型：字符串 - 名称：function_call 数据类型：空值 - 名称：reasoning_content 数据类型：字符串 - 名称：refusal 数据类型：空值 - 名称：role 数据类型：字符串 - 名称：tool_calls 序列类型：空值 - 名称：finish_reason 数据类型：字符串 拆分集： - 名称：train 字节大小：6061060422 样本数量：659358 下载大小：3572636672 数据集总大小：6061060422 # gpt-oss-120b-Infinity-Instruct-0625 ## 数据集说明 本数据集属于用于推测式解码（speculative decoding）研究的LK-Speculators数据集集合，包含66万条提示词-响应对，专为训练草稿模型而设计，该草稿模型将与作为目标模型的gpt-oss-120b配合使用。本数据集通过使用openai/gpt-oss-120b（温度参数设为1）对来自Infinity-Instruct-0625的提示词生成响应而构建。如需了解训练方法与实验结果的更多细节，请参阅我们的论文：《LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding》（https://arxiv.org/abs/2602.23881）。 ## 数据集结构 - **数据格式**：Parquet - **样本总数**：659358 ## 使用方法 python from datasets import load_dataset dataset = load_dataset("nebius/gpt-oss-120b-Infinity-Instruct-0625") ## 许可证 本数据集采用知识共享署名4.0（CC BY 4.0）许可证发布，详情请访问：https://creativecommons.org/licenses/by/4.0/ ## 引用格式 @misc{samarin2026lklosses, title = {LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding}, author = {Alexander Samarin and Sergei Krutikov and Anton Shevtsov and Sergei Skvortsov and Filipp Fisin and Alexander Golubev}, year = {2026}, eprint = {2602.23881}, archivePrefix = {arXiv}, primaryClass = {cs.LG}, url = {https://arxiv.org/abs/2602.23881} }