Datasets:
Download metadata/validation_report.json from CodeLogic-Team/CodeLens: direct link, hf CLI and curl.
- Browser
- Download file 1.8 kB
-
https://huggingface.co/datasets/CodeLogic-Team/CodeLens/resolve/main/metadata/validation_report.json
- Command line
-
hf download hf://datasets/CodeLogic-Team/CodeLens/metadata/validation_report.json
-
curl -L -o validation_report.json https://huggingface.co/datasets/CodeLogic-Team/CodeLens/resolve/main/metadata/validation_report.json
1.8 kB
| { | |
| "status": "PASS", | |
| "configs": { | |
| "kodcode_sft": { | |
| "train_rows": 35974, | |
| "validation_rows": 100, | |
| "train_validation_id_overlap": 0, | |
| "train_validation_prompt_overlap": 0, | |
| "huggingface_load": "PASS" | |
| }, | |
| "kodcode_rl": { | |
| "train_rows": 35974, | |
| "validation_rows": 100, | |
| "train_validation_id_overlap": 0, | |
| "train_validation_prompt_overlap": 0, | |
| "test_cases": { | |
| "train": 210741, | |
| "validation": 572 | |
| }, | |
| "verification": { | |
| "train": {}, | |
| "validation": {} | |
| }, | |
| "huggingface_load": "PASS" | |
| }, | |
| "taco_leetcode_sft": { | |
| "train_rows": 21779, | |
| "validation_rows": 100, | |
| "train_validation_id_overlap": 0, | |
| "train_validation_prompt_overlap": 0, | |
| "huggingface_load": "PASS" | |
| }, | |
| "taco_leetcode_rl": { | |
| "train_rows": 17249, | |
| "validation_rows": 100, | |
| "train_validation_id_overlap": 0, | |
| "train_validation_prompt_overlap": 0, | |
| "test_cases": { | |
| "train": 993732, | |
| "validation": 5433 | |
| }, | |
| "verification": { | |
| "train": { | |
| "reference_execution": 11587, | |
| "structural_only": 5662 | |
| }, | |
| "validation": { | |
| "structural_only": 29, | |
| "reference_execution": 71 | |
| } | |
| }, | |
| "huggingface_load": "PASS" | |
| } | |
| }, | |
| "checks": [ | |
| "hashes", | |
| "every-row schema", | |
| "split IDs", | |
| "split prompts", | |
| "original question IDs", | |
| "reward test structure", | |
| "all ablation memberships" | |
| ], | |
| "cross_config_id_overlap": { | |
| "taco_sft_train_vs_rl_validation": 71, | |
| "taco_sft_validation_vs_rl_train": 48, | |
| "note": "Independent historical experiments; not a joint multi-stage holdout." | |
| } | |
| } | |