reasoning-degeneration-dev/precommittal-canary-visualizations-v1

--- license: mit tags: - precommittal - canary - visualization - fixed --- # precommittal-canary-visualizations-v1 Visualization plots from canary shard. FIXED: added data_json column for trajectory data row (was lost in original upload). ## Dataset Info - **Rows**: 6 - **Columns**: 7 ## Columns | Column | Type | Description | |--------|------|-------------| | plot_name | Value('string') | Identifier for this visualization | | description | Value('string') | What the plot shows | | image | Image(mode=None, decode=True) | PNG plot (null for JSON data row) | | model | Value('string') | Model: Qwen/Qwen3-8B | | dataset | Value('string') | GSM8K | | n_samples | Value('int64') | Number of problems | | data_json | Value('string') | Raw trajectory data as JSON string (only for trajectory_data_json row) | ## Generation Parameters ```json { "script_name": "fix_data_quality.py", "model": "Qwen/Qwen3-8B", "description": "Visualization plots from canary shard. FIXED: added data_json column for trajectory data row (was lost in original upload).", "hyperparameters": { "n_samples": 200, "layers": [ 8, 17, 26, 35 ] }, "input_datasets": [ "reasoning-degeneration-dev/precommittal-canary-results-v1" ] } ``` ## Experiment Documentation For complete experiment details, see [https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/precommittal](https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/precommittal) ## Usage ```python from datasets import load_dataset dataset = load_dataset("reasoning-degeneration-dev/precommittal-canary-visualizations-v1", split="train") print(f"Loaded {len(dataset)} rows") ``` --- *This dataset is tracked in [reasoning-degeneration-dev/PROJECT-MANIFEST](https://huggingface.co/datasets/reasoning-degeneration-dev/PROJECT-MANIFEST)*--- license: MIT许可证 tags: - 预承诺（precommittal） - 金丝雀（canary） - 可视化（visualization） - 已修复（fixed） --- # 预承诺-金丝雀可视化数据集v1 本数据集包含来自金丝雀分片的可视化绘图。**已修复**：新增了轨迹数据行对应的`data_json`列（该列在初始上传时丢失）。 ## 数据集信息 - **行数**：6 - **列数**：7 ## 列说明 | 列名 | 数据类型 | 描述 | |--------|------|-------------| | plot_name | 字符串值类型 | 该可视化的唯一标识符 | | description | 字符串值类型 | 该绘图展示的具体内容 | | image | 图像（无模式，已解码） | PNG格式绘图（轨迹数据行对应此项为null） | | model | 字符串值类型 | 所用模型：Qwen/Qwen3-8B | | dataset | 字符串值类型 | 所用数据集：GSM8K | | n_samples | 64位整数类型 | 问题样本数量 | | data_json | 字符串值类型 | 以JSON字符串形式存储的原始轨迹数据（仅适用于轨迹数据行） | ## 生成参数 json { "script_name": "fix_data_quality.py", "model": "Qwen/Qwen3-8B", "description": "本数据集包含来自金丝雀分片的可视化绘图。已修复：新增了轨迹数据行对应的data_json列（该列在初始上传时丢失）。", "hyperparameters": { "n_samples": 200, "layers": [ 8, 17, 26, 35 ] }, "input_datasets": [ "reasoning-degeneration-dev/precommittal-canary-results-v1" ] } ## 实验文档 如需获取完整实验细节，请访问 [https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/precommittal](https://github.com/Zayne-sprague/SC-Research-Notes/tree/main/experiments/precommittal) ## 使用方法 python from datasets import load_dataset dataset = load_dataset("reasoning-degeneration-dev/precommittal-canary-visualizations-v1", split="train") print(f"已加载 {len(dataset)} 行数据") --- *本数据集已在 [reasoning-degeneration-dev/PROJECT-MANIFEST](https://huggingface.co/datasets/reasoning-degeneration-dev/PROJECT-MANIFEST) 中进行追踪*