Improbable-AI / curiosity_redteam

Official implementation of ICLR'24 paper, "Curiosity-driven Red Teaming for Large Language Models" (https://openreview.net/pdf?id=4KqkizXgXU)
MIT License
65 stars 10 forks source link

toxicity.py is missing #5

Closed PamKing7 closed 4 months ago

PamKing7 commented 4 months ago

When I run the command python experiments/imdb_toxicity_response/run_ppo.py --mode local --gpus 0, I get the error "Module 'toxicity' doesn't exist on the Hugging Face Hub either." Could you tell me where the script curiosity_redteam/toxicity/toxicity.py is located?

Traceback (most recent call last): File "/d2/wpy/curiosity_redteam/ppo_gpt2_gpt2_imdb_toxicity_response.py", line 196, in <module> main(hparams) File "/d2/wpy/curiosity_redteam/ppo_gpt2_gpt2_imdb_toxicity_response.py", line 180, in main reward_fn = RedTeamToxicityRewardModel(device=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/d2/wpy/curiosity_redteam/ppo_gpt2_gpt2_imdb_toxicity_response.py", line 116, in __init__ self.toxicity_fn = evaluate.load( ^^^^^^^^^^^^^^ File "/mnt/data132/zifan/codes/evaluate/src/evaluate/loading.py", line 748, in load evaluation_module = evaluation_module_factory( ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/mnt/data132/zifan/codes/evaluate/src/evaluate/loading.py", line 681, in evaluation_module_factory raise FileNotFoundError( FileNotFoundError: Couldn't find a module script at /d2/wpy/curiosity_redteam/toxicity/toxicity.py. Module 'toxicity' doesn't exist on the Hugging Face Hub either.