Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/_mmlu.yaml +32 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/_mmlu_flan_cot_zeroshot_template_yaml +38 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_anatomy.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_astronomy.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_business_ethics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_clinical_knowledge.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_biology.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_chemistry.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_mathematics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_medicine.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_physics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_computer_security.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_conceptual_physics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_econometrics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_electrical_engineering.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_formal_logic.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_global_facts.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_chemistry.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_european_history.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_geography.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_government_and_politics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_macroeconomics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_mathematics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_microeconomics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_physics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_psychology.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_statistics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_international_law.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_jurisprudence.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_machine_learning.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_management.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_marketing.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_medical_genetics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_miscellaneous.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_moral_scenarios.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_nutrition.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_philosophy.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_prehistory.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_accounting.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_law.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_psychology.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_security_studies.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_us_foreign_policy.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_world_religions.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/_mmlu.yaml +32 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_abstract_algebra.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_business_ethics.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_clinical_knowledge.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_college_biology.yaml +6 -0
- scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_college_chemistry.yaml +6 -0
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/_mmlu.yaml
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
group: mmlu_flan_cot_zeroshot
|
| 2 |
+
group_alias: mmlu (flan style, zeroshot cot)
|
| 3 |
+
task:
|
| 4 |
+
- group: stem
|
| 5 |
+
task:
|
| 6 |
+
- mmlu_flan_cot_zeroshot_stem
|
| 7 |
+
aggregate_metric_list:
|
| 8 |
+
- metric: acc
|
| 9 |
+
weight_by_size: True
|
| 10 |
+
- group: other
|
| 11 |
+
task:
|
| 12 |
+
- mmlu_flan_cot_zeroshot_other
|
| 13 |
+
aggregate_metric_list:
|
| 14 |
+
- metric: acc
|
| 15 |
+
weight_by_size: True
|
| 16 |
+
- group: social sciences
|
| 17 |
+
task:
|
| 18 |
+
- mmlu_flan_cot_zeroshot_social_sciences
|
| 19 |
+
aggregate_metric_list:
|
| 20 |
+
- metric: acc
|
| 21 |
+
weight_by_size: True
|
| 22 |
+
- group: humanities
|
| 23 |
+
task:
|
| 24 |
+
- mmlu_flan_cot_zeroshot_humanities
|
| 25 |
+
aggregate_metric_list:
|
| 26 |
+
- metric: acc
|
| 27 |
+
weight_by_size: True
|
| 28 |
+
aggregate_metric_list:
|
| 29 |
+
- metric: acc
|
| 30 |
+
weight_by_size: True
|
| 31 |
+
metadata:
|
| 32 |
+
version: 2
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/_mmlu_flan_cot_zeroshot_template_yaml
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
dataset_path: hails/mmlu_no_train # a copy of `cais/mmlu` with no auxiliary_train split
|
| 2 |
+
validation_split: validation
|
| 3 |
+
fewshot_split: dev
|
| 4 |
+
output_type: generate_until
|
| 5 |
+
doc_to_text: "Q: {{question.strip()}}\n(A) {{choices[0]}} (B) {{choices[1]}} (C) {{choices[2]}} (D) {{choices[3]}}\nA: Let's think step by step."
|
| 6 |
+
doc_to_target: "{{['(A)', '(B)', '(C)', '(D)'][answer]}}"
|
| 7 |
+
filter_list:
|
| 8 |
+
- name: "strict-match"
|
| 9 |
+
filter:
|
| 10 |
+
- function: "regex"
|
| 11 |
+
regex_pattern: "((?<=The answer is )(.*)(?=.)|(?<=answer is )(.*)(?=.)|(?<=The answer: )(.*)(?=.)|(?<=The final answer: )(.*)(?=.))"
|
| 12 |
+
- function: "take_first"
|
| 13 |
+
- name: "flexible-extract"
|
| 14 |
+
filter:
|
| 15 |
+
- function: !function utils.MultiChoiceRegexFilter
|
| 16 |
+
group_select: -1
|
| 17 |
+
ignore_case: true
|
| 18 |
+
ignore_punctuation: true
|
| 19 |
+
regex_pattern: "(\\([A-Z]\\))"
|
| 20 |
+
- function: "take_first"
|
| 21 |
+
generation_kwargs:
|
| 22 |
+
until:
|
| 23 |
+
- "</s>"
|
| 24 |
+
- "Q:"
|
| 25 |
+
- "<|im_end|>"
|
| 26 |
+
do_sample: false
|
| 27 |
+
temperature: 0.0
|
| 28 |
+
num_fewshot: 0
|
| 29 |
+
metric_list:
|
| 30 |
+
- metric: exact_match
|
| 31 |
+
aggregation: mean
|
| 32 |
+
higher_is_better: true
|
| 33 |
+
ignore_case: true
|
| 34 |
+
ignore_punctuation: true
|
| 35 |
+
metadata:
|
| 36 |
+
version: 3.0
|
| 37 |
+
dataset_kwargs:
|
| 38 |
+
trust_remote_code: true
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_anatomy.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "anatomy"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about anatomy.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_anatomy"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_astronomy.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "astronomy"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about astronomy.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_astronomy"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_business_ethics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "business_ethics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about business\
|
| 3 |
+
\ ethics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_business_ethics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_clinical_knowledge.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "clinical_knowledge"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about clinical\
|
| 3 |
+
\ knowledge.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_clinical_knowledge"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_biology.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_biology"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ biology.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_college_biology"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_chemistry.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_chemistry"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ chemistry.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_college_chemistry"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_mathematics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_mathematics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ mathematics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_college_mathematics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_medicine.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_medicine"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ medicine.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_college_medicine"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_college_physics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_physics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ physics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_college_physics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_computer_security.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "computer_security"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about computer\
|
| 3 |
+
\ security.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_computer_security"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_conceptual_physics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "conceptual_physics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about conceptual\
|
| 3 |
+
\ physics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_conceptual_physics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_econometrics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "econometrics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about econometrics.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_econometrics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_electrical_engineering.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "electrical_engineering"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about electrical\
|
| 3 |
+
\ engineering.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_electrical_engineering"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_formal_logic.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "formal_logic"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about formal\
|
| 3 |
+
\ logic.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_formal_logic"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_global_facts.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "global_facts"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about global\
|
| 3 |
+
\ facts.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_global_facts"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_chemistry.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_chemistry"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school chemistry.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_chemistry"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_european_history.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_european_history"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school european history.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_european_history"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_geography.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_geography"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school geography.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_geography"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_government_and_politics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_government_and_politics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school government and politics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_government_and_politics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_macroeconomics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_macroeconomics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school macroeconomics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_macroeconomics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_mathematics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_mathematics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school mathematics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_mathematics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_microeconomics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_microeconomics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school microeconomics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_microeconomics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_physics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_physics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school physics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_physics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_psychology.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_psychology"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school psychology.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_psychology"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_high_school_statistics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "high_school_statistics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about high\
|
| 3 |
+
\ school statistics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_high_school_statistics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_international_law.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "international_law"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about international\
|
| 3 |
+
\ law.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_international_law"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_jurisprudence.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "jurisprudence"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about jurisprudence.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_jurisprudence"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_machine_learning.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "machine_learning"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about machine\
|
| 3 |
+
\ learning.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_stem"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_machine_learning"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_management.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "management"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about management.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_management"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_marketing.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "marketing"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about marketing.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_marketing"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_medical_genetics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "medical_genetics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about medical\
|
| 3 |
+
\ genetics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_medical_genetics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_miscellaneous.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "miscellaneous"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about miscellaneous.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_miscellaneous"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_moral_scenarios.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "moral_scenarios"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about moral\
|
| 3 |
+
\ scenarios.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_moral_scenarios"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_nutrition.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "nutrition"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about nutrition.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_nutrition"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_philosophy.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "philosophy"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about philosophy.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_philosophy"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_prehistory.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "prehistory"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about prehistory.\n\
|
| 3 |
+
\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_prehistory"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_accounting.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "professional_accounting"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about professional\
|
| 3 |
+
\ accounting.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_other"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_professional_accounting"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_law.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "professional_law"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about professional\
|
| 3 |
+
\ law.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_professional_law"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_professional_psychology.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "professional_psychology"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about professional\
|
| 3 |
+
\ psychology.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_professional_psychology"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_security_studies.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "security_studies"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about security\
|
| 3 |
+
\ studies.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_security_studies"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_us_foreign_policy.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "us_foreign_policy"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about us\
|
| 3 |
+
\ foreign policy.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_social_sciences"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_us_foreign_policy"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_cot_zeroshot/mmlu_world_religions.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "world_religions"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about world\
|
| 3 |
+
\ religions.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_cot_zeroshot_humanities"
|
| 5 |
+
"include": "_mmlu_flan_cot_zeroshot_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_cot_zeroshot_world_religions"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/_mmlu.yaml
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
group: mmlu_flan_n_shot_generative
|
| 2 |
+
group_alias: mmlu (flan style, generative)
|
| 3 |
+
task:
|
| 4 |
+
- group: stem
|
| 5 |
+
task:
|
| 6 |
+
- mmlu_flan_n_shot_generative_stem
|
| 7 |
+
aggregate_metric_list:
|
| 8 |
+
- metric: acc
|
| 9 |
+
weight_by_size: True
|
| 10 |
+
- group: other
|
| 11 |
+
task:
|
| 12 |
+
- mmlu_flan_n_shot_generative_other
|
| 13 |
+
aggregate_metric_list:
|
| 14 |
+
- metric: acc
|
| 15 |
+
weight_by_size: True
|
| 16 |
+
- group: social sciences
|
| 17 |
+
task:
|
| 18 |
+
- mmlu_flan_n_shot_generative_social_sciences
|
| 19 |
+
aggregate_metric_list:
|
| 20 |
+
- metric: acc
|
| 21 |
+
weight_by_size: True
|
| 22 |
+
- group: humanities
|
| 23 |
+
task:
|
| 24 |
+
- mmlu_flan_n_shot_generative_humanities
|
| 25 |
+
aggregate_metric_list:
|
| 26 |
+
- metric: acc
|
| 27 |
+
weight_by_size: True
|
| 28 |
+
aggregate_metric_list:
|
| 29 |
+
- metric: acc
|
| 30 |
+
weight_by_size: True
|
| 31 |
+
metadata:
|
| 32 |
+
version: 2
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_abstract_algebra.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "abstract_algebra"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about abstract\
|
| 3 |
+
\ algebra.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_n_shot_generative_stem"
|
| 5 |
+
"include": "_mmlu_flan_generative_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_n_shot_generative_abstract_algebra"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_business_ethics.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "business_ethics"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about business\
|
| 3 |
+
\ ethics.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_n_shot_generative_other"
|
| 5 |
+
"include": "_mmlu_flan_generative_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_n_shot_generative_business_ethics"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_clinical_knowledge.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "clinical_knowledge"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about clinical\
|
| 3 |
+
\ knowledge.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_n_shot_generative_other"
|
| 5 |
+
"include": "_mmlu_flan_generative_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_n_shot_generative_clinical_knowledge"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_college_biology.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_biology"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ biology.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_n_shot_generative_stem"
|
| 5 |
+
"include": "_mmlu_flan_generative_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_n_shot_generative_college_biology"
|
scripts/yans/lm-evaluation-harness/lm_eval/tasks/mmlu/flan_n_shot/generative/mmlu_college_chemistry.yaml
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"dataset_name": "college_chemistry"
|
| 2 |
+
"description": "The following are multiple choice questions (with answers) about college\
|
| 3 |
+
\ chemistry.\n\n"
|
| 4 |
+
"tag": "mmlu_flan_n_shot_generative_stem"
|
| 5 |
+
"include": "_mmlu_flan_generative_template_yaml"
|
| 6 |
+
"task": "mmlu_flan_n_shot_generative_college_chemistry"
|