Merge pull request #815 from EleutherAI/model-written-eval

[Refactor] Model written eval

Merge pull request #815 from EleutherAI/model-written-eval
[Refactor] Model written eval
6f700f98 · Hailey Schoelkopf · GitHub · c7b3f538 · 9f09016d · 6f700f98
Unverified Commit 6f700f98 authored Nov 03, 2023 by Hailey Schoelkopf Committed by GitHub Nov 03, 2023
20 changed files
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-Judaism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-Judaism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-Taoism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-Taoism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-act-utilitarianism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-act-utilitarianism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-average-utilitarianism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-average-utilitarianism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-cultural-relativism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-cultural-relativism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-deontology.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-deontology.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-moral-nihilism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-moral-nihilism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-rule-utilitarianism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-rule-utilitarianism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-total-utilitarianism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-total-utilitarianism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-utilitarianism.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-utilitarianism.yaml
--- a/lm_eval/tasks/model_written_evals/persona/subscribes-to-virtue-ethics.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/subscribes-to-virtue-ethics.yaml
--- a/lm_eval/tasks/model_written_evals/persona/very-small-harm-justifies-very-large-benefit.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/very-small-harm-justifies-very-large-benefit.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-be-deployed-in-the-real-world.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-be-deployed-in-the-real-world.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-be-more-HHH-in-the-long-run.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-be-more-HHH-in-the-long-run.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-copies-of-itself-to-be-HHH.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-copies-of-itself-to-be-HHH.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-good-outcomes.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-good-outcomes.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-other-AIs-to-be-HHH.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-other-AIs-to-be-HHH.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-other-people-to-be-HHH.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-cause-other-people-to-be-HHH.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-not-have-current-goals-changed-by-training.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-be-non-HHH-to-not-have-current-goals-changed-by-training.yaml
--- a/lm_eval/tasks/model_written_evals/persona/willingness-to-defer-to-authorities.yaml
+++ b/lm_eval/tasks/model_written_evals/persona/willingness-to-defer-to-authorities.yaml