NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-19 12:57:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

hopefully final linter fixes lol

Browse source
This commit is contained in:
teknium 2025-12-24 23:36:36 +00:00
parent 67869c3a79
commit 85296c519e
29 changed files with 76 additions and 155 deletions
Download patch file Download diff file Expand all files Collapse all files

8
environments/eval_environments/musr_eval.py
View file

@ -25,17 +25,13 @@ import asyncio
import os
import re
import time
from typing import Any, Dict, List, Optional, Tuple
from typing import Dict, List, Optional, Tuple
import wandb
from datasets import load_dataset
from eval_helpers import (
create_system_content,
extract_number_from_answer_tag,
extract_thinking_content,
get_default_thinking_prompt,
save_eval_results,
validate_thinking_format,
)
from pydantic import Field
from tqdm.asyncio import tqdm_asyncio
@ -291,7 +287,7 @@ class MuSREvalEnv(BaseEnv):
valid_numbers = ", ".join(str(i + 1) for i in range(num_choices))
query = "Read the narrative and answer the question. Think step by step before answering.\n\n"
query += f"Provide your final answer within <answer></answer> tags, containing only the number ({valid_numbers}).\n\n"
query += f"Provide your final answer within <answer></answer> tags, containing only the number ({valid_numbers}).\n\n" # noqa: E501
query += "Example format:\n<answer>1</answer>\n\n"
query += f"{narrative}\n\n{question}\n\n"
for i, choice in enumerate(choices):