MCILAB/LLM_Alignment_Evaluation

1---2title: Alignment3emoji: 🥇4colorFrom: green5colorTo: indigo6sdk: gradio7app_file: app.py8pinned: true9license: apache-2.010short_description: alignment leader board11---12 13# Start the configuration14 15Most of the variables to change for a default leaderboard are in `src/env.py` (replace the path for your leaderboard) and `src/about.py` (for tasks).16 17Results files should have the following format and be stored as json files:18```json19{20    "config": {21        "model_dtype": "torch.float16", # or torch.bfloat16 or 8bit or 4bit22        "model_name": "path of the model on the hub: org/model",23        "model_sha": "revision on the hub",24    },25    "results": {26        "task_name": {27            "metric_name": score,28        },29        "task_name2": {30            "metric_name": score,31        }32    }33}34```35 36Request files are created automatically by this tool.37 38If you encounter problem on the space, don't hesitate to restart it to remove the create eval-queue, eval-queue-bk, eval-results and eval-results-bk created folder.39 40# Code logic for more complex edits41 42You'll find 43- the main table' columns names and properties in `src/display/utils.py`44- the logic to read all results and request files, then convert them in dataframe lines, in `src/leaderboard/read_evals.py`, and `src/populate.py`45- the logic to allow or filter submissions in `src/submission/submit.py` and `src/submission/check_validity.py`