split to easy and challenge

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_06/c.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_06
+task: arc_challenge_alt_ov_06c
+doc_to_text: !function ../styles.template_06
+doc_to_choice: !function ../styles.choice_06c
+doc_to_decontamination_query: !function ../styles.template_06

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_07/a.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_07
+task: arc_challenge_alt_ov_07a
+doc_to_text: !function ../styles.template_07
+doc_to_choice: !function ../styles.choice_07a
+doc_to_decontamination_query: !function ../styles.template_07

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_07/b.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_07
+task: arc_challenge_alt_ov_07b
+doc_to_text: !function ../styles.template_07
+doc_to_choice: !function ../styles.choice_07b
+doc_to_decontamination_query: !function ../styles.template_07

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_07/c.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_07
+task: arc_challenge_alt_ov_07c
+doc_to_text: !function ../styles.template_07
+doc_to_choice: !function ../styles.choice_07c
+doc_to_decontamination_query: !function ../styles.template_07

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_08/a.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_08
+task: arc_challenge_alt_ov_08a
+doc_to_text: !function ../styles.template_08
+doc_to_choice: !function ../styles.choice_08a
+doc_to_decontamination_query: !function ../styles.template_08

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_08/b.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_08
+task: arc_challenge_alt_ov_08b
+doc_to_text: !function ../styles.template_08
+doc_to_choice: !function ../styles.choice_08b
+doc_to_decontamination_query: !function ../styles.template_08

lm_eval/tasks/arc/alternative_worlds/arc_challenge/output_variation/style_08/c.yaml

+include: ../_arc_challenge_alt_yaml
+group: arc_challenge_alt_ov_08
+task: arc_challenge_alt_ov_08c
+doc_to_text: !function ../styles.template_08
+doc_to_choice: !function ../styles.choice_08c
+doc_to_decontamination_query: !function ../styles.template_08

lm_eval/tasks/arc/alternative_worlds/arc_challenge/prompt_variation/_arc_challenge_alt_yaml

+dataset_path: ai2_arc
+dataset_name: ARC-Challenge
+output_type: multiple_choice
+training_split: train
+validation_split: validation
+test_split: test
+doc_to_text: "Question: {{question}}\nAnswer:"
+doc_to_target: "{{choices.label.index(answerKey)}}"
+doc_to_choice: "{{choices.text}}"
+should_decontaminate: true
+doc_to_decontamination_query: "Question: {{question}}\nAnswer:"
+metric_list:
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+  - metric: acc_norm
+    aggregation: mean
+    higher_is_better: true
+  - metric: brier_score
+    aggregation: brier_score
+    higher_is_better: false

lm_eval/tasks/arc/alternative_worlds/arc_challenge/prompt_variation/style_01.yaml

+include: _arc_challenge_alt_yaml
+group: arc_challenge_alt_pv
+task: arc_challenge_alt_pv_01
+doc_to_text: "{{question}}"
+doc_to_decontamination_query: "{{question}}"

lm_eval/tasks/arc/alternative_worlds/arc_challenge/prompt_variation/style_02.yaml

+include: _arc_challenge_alt_yaml
+group: arc_challenge_alt_pv
+task: arc_challenge_alt_pv_02
+doc_to_text: "Q: {{question}}\nA:"
+doc_to_decontamination_query: "Q: {{question}}\nA:"

lm_eval/tasks/arc/alternative_worlds/arc_challenge/prompt_variation/style_03.yaml

+include: _arc_challenge_alt_yaml
+group: arc_challenge_alt_pv
+task: arc_challenge_alt_pv_03
+doc_to_text: "Question: {{question}}\nAnswer:"
+doc_to_decontamination_query: "Question: {{question}}\nAnswer:"

lm_eval/tasks/arc/alternative_worlds/arc_easy/README.md

+
+
+Investigate affect of letter options
+- (A)
+- A)
+- A.
+- A\t
+- (a)
+- a)
+- a.
+- a\t
+
+Answer types:
+- letters only
+    - original option
+    - just letter
+- letters + continuation
+    - original option
+    - just letter
+- continuation