From 797d20d5e10fcb842a0d466d51377d71326c6d83 Mon Sep 17 00:00:00 2001 From: Ilia Kulikov Date: Thu, 26 Sep 2024 14:54:02 -0700 Subject: [PATCH] Update projects/self_taught_evaluator/README.md Co-authored-by: Golovneva <103262907+Golovneva@users.noreply.github.com> --- projects/self_taught_evaluator/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/projects/self_taught_evaluator/README.md b/projects/self_taught_evaluator/README.md index 6e96a2f..3a57bf6 100644 --- a/projects/self_taught_evaluator/README.md +++ b/projects/self_taught_evaluator/README.md @@ -21,7 +21,7 @@ We provide example scripts to use the self-taught evaluator as a judge to choose Note: download example eval data here: https://dl.fbaipublicfiles.com/self_taught_evaluator/example_inputs.jsonl -1. Prepare your inputs similar to ones found in [example_inputs.jsonl](./src/example_inputs.jsonl) +1. Prepare your inputs similar to ones found in [example_inputs.jsonl](./data/eval_data/example_inputs.jsonl) 2. Run `bash run_inference_wvllm.sh`. The generated outputs and parsed judgements will be saved in `example_outputs.jsonl`.