Merge pull request #816 from EleutherAI/flan-benchmark

[Refactor] Flan benchmark

lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_world_religions.yaml

+dataset_name: world_religions
+description: 'The following are multiple choice questions (with answers) about world
+  religions.
+
+
+  '
+include: _mmlu_flan_generative_template_yaml
+task: mmlu_flan_cot_zeroshot_world_religions

lm_eval/tasks/mmlu/flan_n_shot/_mmlu_flan_generative_template_yaml

+group: mmlu_flan_n_shot_generative
+dataset_path: cais/mmlu
+test_split: test
+fewshot_split: dev
+output_type: greedy_until
+doc_to_text: "Q: {{question.strip()}}\n(A) {{choices[0]}} (B) {{choices[1]}} (C) {{choices[2]}} (D) {{choices[3]}}\nA: "
+doc_to_target: "{{['(A)', '(B)', '(C)', '(D)'][answer]}}"
+generation_kwargs:
+  until:
+    - "</s>"
+metric_list:
+  - metric: exact_match
+    aggregation: mean
+    higher_is_better: true

lm_eval/tasks/mmlu/flan_n_shot/_mmlu_flan_loglikelihood_template_yaml

+group: mmlu_flan_n_shot_loglikelihood
+dataset_path: cais/mmlu
+test_split: test
+fewshot_split: dev
+output_type: multiple_choice
+doc_to_text: "Q: {{question.strip()}}\n(A) {{choices[0]}} (B) {{choices[1]}} (C) {{choices[2]}} (D) {{choices[3]}}\nA: "
+doc_to_choice: ["(A)", "(B)", "(C)", "(D)"]
+doc_to_target: answer
+metric_list:
+  - metric: acc
+    aggregation: mean
+    higher_is_better: true
+  - metric: acc_norm
+    aggregation: mean
+    higher_is_better: true

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_abstract_algebra.yaml

+"dataset_name": "abstract_algebra"
+"description": "The following are multiple choice questions (with answers) about abstract algebra.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_abstract_algebra"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_anatomy.yaml

+"dataset_name": "anatomy"
+"description": "The following are multiple choice questions (with answers) about anatomy.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_anatomy"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_astronomy.yaml

+"dataset_name": "astronomy"
+"description": "The following are multiple choice questions (with answers) about astronomy.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_astronomy"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_business_ethics.yaml

+"dataset_name": "business_ethics"
+"description": "The following are multiple choice questions (with answers) about business ethics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_business_ethics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_clinical_knowledge.yaml

+"dataset_name": "clinical_knowledge"
+"description": "The following are multiple choice questions (with answers) about clinical knowledge.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_clinical_knowledge"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_biology.yaml

+"dataset_name": "college_biology"
+"description": "The following are multiple choice questions (with answers) about college biology.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_biology"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_chemistry.yaml

+"dataset_name": "college_chemistry"
+"description": "The following are multiple choice questions (with answers) about college chemistry.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_chemistry"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_computer_science.yaml

+"dataset_name": "college_computer_science"
+"description": "The following are multiple choice questions (with answers) about college computer science.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_computer_science"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_mathematics.yaml

+"dataset_name": "college_mathematics"
+"description": "The following are multiple choice questions (with answers) about college mathematics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_mathematics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_medicine.yaml

+"dataset_name": "college_medicine"
+"description": "The following are multiple choice questions (with answers) about college medicine.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_medicine"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_college_physics.yaml

+"dataset_name": "college_physics"
+"description": "The following are multiple choice questions (with answers) about college physics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_college_physics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_computer_security.yaml

+"dataset_name": "computer_security"
+"description": "The following are multiple choice questions (with answers) about computer security.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_computer_security"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_conceptual_physics.yaml

+"dataset_name": "conceptual_physics"
+"description": "The following are multiple choice questions (with answers) about conceptual physics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_conceptual_physics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_econometrics.yaml

+"dataset_name": "econometrics"
+"description": "The following are multiple choice questions (with answers) about econometrics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_econometrics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_electrical_engineering.yaml

+"dataset_name": "electrical_engineering"
+"description": "The following are multiple choice questions (with answers) about electrical engineering.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_electrical_engineering"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_elementary_mathematics.yaml

+"dataset_name": "elementary_mathematics"
+"description": "The following are multiple choice questions (with answers) about elementary mathematics.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_elementary_mathematics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_gen_formal_logic.yaml

+"dataset_name": "formal_logic"
+"description": "The following are multiple choice questions (with answers) about formal logic.\n\n"
+"include": "_mmlu_flan_generative_template_yaml"
+"task": "mmlu_flan_n_shot_generative_formal_logic"