├── .gitignore ├── .python-version ├── CHANGELOG.md ├── README.md ├── cs336_alignment ├── __init__.py ├── drgrpo_grader.py └── prompts │ ├── alpaca_sft.prompt │ ├── question_only.prompt │ ├── r1_zero.prompt │ └── zero_shot_system_prompt.prompt ├── cs336_spring2025_assignment5_alignment.pdf ├── cs336_spring2025_assignment5_supplement_safety_rlhf.pdf ├── data ├── alpaca_eval │ └── alpaca_eval.jsonl ├── gsm8k │ ├── test.jsonl │ └── train.jsonl ├── mmlu │ ├── README.txt │ ├── dev │ │ ├── abstract_algebra_dev.csv │ │ ├── anatomy_dev.csv │ │ ├── astronomy_dev.csv │ │ ├── business_ethics_dev.csv │ │ ├── clinical_knowledge_dev.csv │ │ ├── college_biology_dev.csv │ │ ├── college_chemistry_dev.csv │ │ ├── college_computer_science_dev.csv │ │ ├── college_mathematics_dev.csv │ │ ├── college_medicine_dev.csv │ │ ├── college_physics_dev.csv │ │ ├── computer_security_dev.csv │ │ ├── conceptual_physics_dev.csv │ │ ├── econometrics_dev.csv │ │ ├── electrical_engineering_dev.csv │ │ ├── elementary_mathematics_dev.csv │ │ ├── formal_logic_dev.csv │ │ ├── global_facts_dev.csv │ │ ├── high_school_biology_dev.csv │ │ ├── high_school_chemistry_dev.csv │ │ ├── high_school_computer_science_dev.csv │ │ ├── high_school_european_history_dev.csv │ │ ├── high_school_geography_dev.csv │ │ ├── high_school_government_and_politics_dev.csv │ │ ├── high_school_macroeconomics_dev.csv │ │ ├── high_school_mathematics_dev.csv │ │ ├── high_school_microeconomics_dev.csv │ │ ├── high_school_physics_dev.csv │ │ ├── high_school_psychology_dev.csv │ │ ├── high_school_statistics_dev.csv │ │ ├── high_school_us_history_dev.csv │ │ ├── high_school_world_history_dev.csv │ │ ├── human_aging_dev.csv │ │ ├── human_sexuality_dev.csv │ │ ├── international_law_dev.csv │ │ ├── jurisprudence_dev.csv │ │ ├── logical_fallacies_dev.csv │ │ ├── machine_learning_dev.csv │ │ ├── management_dev.csv │ │ ├── marketing_dev.csv │ │ ├── medical_genetics_dev.csv │ │ ├── miscellaneous_dev.csv │ │ ├── moral_disputes_dev.csv │ │ ├── moral_scenarios_dev.csv │ │ ├── nutrition_dev.csv │ │ ├── philosophy_dev.csv │ │ ├── prehistory_dev.csv │ │ ├── professional_accounting_dev.csv │ │ ├── professional_law_dev.csv │ │ ├── professional_medicine_dev.csv │ │ ├── professional_psychology_dev.csv │ │ ├── public_relations_dev.csv │ │ ├── security_studies_dev.csv │ │ ├── sociology_dev.csv │ │ ├── us_foreign_policy_dev.csv │ │ ├── virology_dev.csv │ │ └── world_religions_dev.csv │ ├── possibly_contaminated_urls.txt │ ├── test │ │ ├── abstract_algebra_test.csv │ │ ├── anatomy_test.csv │ │ ├── astronomy_test.csv │ │ ├── business_ethics_test.csv │ │ ├── clinical_knowledge_test.csv │ │ ├── college_biology_test.csv │ │ ├── college_chemistry_test.csv │ │ ├── college_computer_science_test.csv │ │ ├── college_mathematics_test.csv │ │ ├── college_medicine_test.csv │ │ ├── college_physics_test.csv │ │ ├── computer_security_test.csv │ │ ├── conceptual_physics_test.csv │ │ ├── econometrics_test.csv │ │ ├── electrical_engineering_test.csv │ │ ├── elementary_mathematics_test.csv │ │ ├── formal_logic_test.csv │ │ ├── global_facts_test.csv │ │ ├── high_school_biology_test.csv │ │ ├── high_school_chemistry_test.csv │ │ ├── high_school_computer_science_test.csv │ │ ├── high_school_european_history_test.csv │ │ ├── high_school_geography_test.csv │ │ ├── high_school_government_and_politics_test.csv │ │ ├── high_school_macroeconomics_test.csv │ │ ├── high_school_mathematics_test.csv │ │ ├── high_school_microeconomics_test.csv │ │ ├── high_school_physics_test.csv │ │ ├── high_school_psychology_test.csv │ │ ├── high_school_statistics_test.csv │ │ ├── high_school_us_history_test.csv │ │ ├── high_school_world_history_test.csv │ │ ├── human_aging_test.csv │ │ ├── human_sexuality_test.csv │ │ ├── international_law_test.csv │ │ ├── jurisprudence_test.csv │ │ ├── logical_fallacies_test.csv │ │ ├── machine_learning_test.csv │ │ ├── management_test.csv │ │ ├── marketing_test.csv │ │ ├── medical_genetics_test.csv │ │ ├── miscellaneous_test.csv │ │ ├── moral_disputes_test.csv │ │ ├── moral_scenarios_test.csv │ │ ├── nutrition_test.csv │ │ ├── philosophy_test.csv │ │ ├── prehistory_test.csv │ │ ├── professional_accounting_test.csv │ │ ├── professional_law_test.csv │ │ ├── professional_medicine_test.csv │ │ ├── professional_psychology_test.csv │ │ ├── public_relations_test.csv │ │ ├── security_studies_test.csv │ │ ├── sociology_test.csv │ │ ├── us_foreign_policy_test.csv │ │ ├── virology_test.csv │ │ └── world_religions_test.csv │ └── val │ │ ├── abstract_algebra_val.csv │ │ ├── anatomy_val.csv │ │ ├── astronomy_val.csv │ │ ├── business_ethics_val.csv │ │ ├── clinical_knowledge_val.csv │ │ ├── college_biology_val.csv │ │ ├── college_chemistry_val.csv │ │ ├── college_computer_science_val.csv │ │ ├── college_mathematics_val.csv │ │ ├── college_medicine_val.csv │ │ ├── college_physics_val.csv │ │ ├── computer_security_val.csv │ │ ├── conceptual_physics_val.csv │ │ ├── econometrics_val.csv │ │ ├── electrical_engineering_val.csv │ │ ├── elementary_mathematics_val.csv │ │ ├── formal_logic_val.csv │ │ ├── global_facts_val.csv │ │ ├── high_school_biology_val.csv │ │ ├── high_school_chemistry_val.csv │ │ ├── high_school_computer_science_val.csv │ │ ├── high_school_european_history_val.csv │ │ ├── high_school_geography_val.csv │ │ ├── high_school_government_and_politics_val.csv │ │ ├── high_school_macroeconomics_val.csv │ │ ├── high_school_mathematics_val.csv │ │ ├── high_school_microeconomics_val.csv │ │ ├── high_school_physics_val.csv │ │ ├── high_school_psychology_val.csv │ │ ├── high_school_statistics_val.csv │ │ ├── high_school_us_history_val.csv │ │ ├── high_school_world_history_val.csv │ │ ├── human_aging_val.csv │ │ ├── human_sexuality_val.csv │ │ ├── international_law_val.csv │ │ ├── jurisprudence_val.csv │ │ ├── logical_fallacies_val.csv │ │ ├── machine_learning_val.csv │ │ ├── management_val.csv │ │ ├── marketing_val.csv │ │ ├── medical_genetics_val.csv │ │ ├── miscellaneous_val.csv │ │ ├── moral_disputes_val.csv │ │ ├── moral_scenarios_val.csv │ │ ├── nutrition_val.csv │ │ ├── philosophy_val.csv │ │ ├── prehistory_val.csv │ │ ├── professional_accounting_val.csv │ │ ├── professional_law_val.csv │ │ ├── professional_medicine_val.csv │ │ ├── professional_psychology_val.csv │ │ ├── public_relations_val.csv │ │ ├── security_studies_val.csv │ │ ├── sociology_val.csv │ │ ├── us_foreign_policy_val.csv │ │ ├── virology_val.csv │ │ └── world_religions_val.csv └── simple_safety_tests │ └── simple_safety_tests.csv ├── pyproject.toml ├── scripts ├── alpaca_eval_vllm_llama3_3_70b_fn │ ├── alpaca_eval_fn.txt │ └── configs.yaml └── evaluate_safety.py ├── test_and_make_submission.sh ├── tests ├── __init__.py ├── _snapshots │ ├── test_compute_entropy.npz │ ├── test_compute_group_normalized_rewards_no_normalize_by_std.npz │ ├── test_compute_group_normalized_rewards_normalize_by_std.npz │ ├── test_compute_grpo_clip_loss_large_cliprange.npz │ ├── test_compute_grpo_clip_loss_small_cliprange.npz │ ├── test_compute_naive_policy_gradient_loss.npz │ ├── test_compute_policy_gradient_loss_grpo_clip.npz │ ├── test_compute_policy_gradient_loss_no_baseline.npz │ ├── test_compute_policy_gradient_loss_reinforce_with_baseline.npz │ ├── test_get_response_log_probs.npz │ ├── test_grpo_microbatch_train_step_grpo_clip.npz │ ├── test_grpo_microbatch_train_step_grpo_clip_10_steps.npz │ ├── test_masked_mean_dim0.npz │ ├── test_masked_mean_dim1.npz │ ├── test_masked_mean_dimNone.npz │ ├── test_masked_mean_dimlast.npz │ ├── test_masked_normalize_dim0.npz │ ├── test_masked_normalize_dim1.npz │ ├── test_masked_normalize_dimNone.npz │ ├── test_masked_normalize_dimlast.npz │ ├── test_sft_microbatch_train_step.npz │ ├── test_sft_microbatch_train_step_10_steps.npz │ ├── test_sft_microbatch_train_step_normalize.npz │ └── test_tokenize_prompt_and_output.npz ├── adapters.py ├── common.py ├── conftest.py ├── fixtures │ ├── .gitkeep │ ├── Meta-Llama-3-8B │ │ ├── special_tokens_map.json │ │ ├── tokenizer.json │ │ └── tokenizer_config.json │ ├── sft_sample.jsonl │ ├── tiny-gpt2-ref │ │ ├── config.json │ │ ├── generation_config.json │ │ └── model.safetensors │ ├── tiny-gpt2 │ │ ├── config.json │ │ ├── generation_config.json │ │ └── model.safetensors │ └── tokenized_sft_sample.json ├── test_data.py ├── test_dpo.py ├── test_grpo.py ├── test_metrics.py └── test_sft.py └── uv.lock /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/.gitignore -------------------------------------------------------------------------------- /.python-version: -------------------------------------------------------------------------------- 1 | 3.12 2 | -------------------------------------------------------------------------------- /CHANGELOG.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/CHANGELOG.md -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/README.md -------------------------------------------------------------------------------- /cs336_alignment/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /cs336_alignment/drgrpo_grader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_alignment/drgrpo_grader.py -------------------------------------------------------------------------------- /cs336_alignment/prompts/alpaca_sft.prompt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_alignment/prompts/alpaca_sft.prompt -------------------------------------------------------------------------------- /cs336_alignment/prompts/question_only.prompt: -------------------------------------------------------------------------------- 1 | {question} -------------------------------------------------------------------------------- /cs336_alignment/prompts/r1_zero.prompt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_alignment/prompts/r1_zero.prompt -------------------------------------------------------------------------------- /cs336_alignment/prompts/zero_shot_system_prompt.prompt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_alignment/prompts/zero_shot_system_prompt.prompt -------------------------------------------------------------------------------- /cs336_spring2025_assignment5_alignment.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_spring2025_assignment5_alignment.pdf -------------------------------------------------------------------------------- /cs336_spring2025_assignment5_supplement_safety_rlhf.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/cs336_spring2025_assignment5_supplement_safety_rlhf.pdf -------------------------------------------------------------------------------- /data/alpaca_eval/alpaca_eval.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/alpaca_eval/alpaca_eval.jsonl -------------------------------------------------------------------------------- /data/gsm8k/test.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/gsm8k/test.jsonl -------------------------------------------------------------------------------- /data/gsm8k/train.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/gsm8k/train.jsonl -------------------------------------------------------------------------------- /data/mmlu/README.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/README.txt -------------------------------------------------------------------------------- /data/mmlu/dev/abstract_algebra_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/abstract_algebra_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/anatomy_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/anatomy_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/astronomy_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/astronomy_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/business_ethics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/business_ethics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/clinical_knowledge_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/clinical_knowledge_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_biology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_biology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_chemistry_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_chemistry_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_computer_science_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_computer_science_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_mathematics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_mathematics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_medicine_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_medicine_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/college_physics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/college_physics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/computer_security_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/computer_security_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/conceptual_physics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/conceptual_physics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/econometrics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/econometrics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/electrical_engineering_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/electrical_engineering_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/elementary_mathematics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/elementary_mathematics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/formal_logic_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/formal_logic_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/global_facts_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/global_facts_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_biology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_biology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_chemistry_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_chemistry_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_computer_science_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_computer_science_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_european_history_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_european_history_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_geography_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_geography_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_government_and_politics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_government_and_politics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_macroeconomics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_macroeconomics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_mathematics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_mathematics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_microeconomics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_microeconomics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_physics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_physics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_psychology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_psychology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_statistics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_statistics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_us_history_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_us_history_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/high_school_world_history_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/high_school_world_history_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/human_aging_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/human_aging_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/human_sexuality_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/human_sexuality_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/international_law_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/international_law_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/jurisprudence_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/jurisprudence_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/logical_fallacies_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/logical_fallacies_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/machine_learning_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/machine_learning_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/management_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/management_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/marketing_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/marketing_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/medical_genetics_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/medical_genetics_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/miscellaneous_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/miscellaneous_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/moral_disputes_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/moral_disputes_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/moral_scenarios_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/moral_scenarios_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/nutrition_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/nutrition_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/philosophy_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/philosophy_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/prehistory_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/prehistory_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/professional_accounting_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/professional_accounting_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/professional_law_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/professional_law_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/professional_medicine_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/professional_medicine_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/professional_psychology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/professional_psychology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/public_relations_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/public_relations_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/security_studies_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/security_studies_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/sociology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/sociology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/us_foreign_policy_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/us_foreign_policy_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/virology_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/virology_dev.csv -------------------------------------------------------------------------------- /data/mmlu/dev/world_religions_dev.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/dev/world_religions_dev.csv -------------------------------------------------------------------------------- /data/mmlu/possibly_contaminated_urls.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/possibly_contaminated_urls.txt -------------------------------------------------------------------------------- /data/mmlu/test/abstract_algebra_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/abstract_algebra_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/anatomy_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/anatomy_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/astronomy_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/astronomy_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/business_ethics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/business_ethics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/clinical_knowledge_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/clinical_knowledge_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_biology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_biology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_chemistry_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_chemistry_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_computer_science_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_computer_science_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_mathematics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_mathematics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_medicine_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_medicine_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/college_physics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/college_physics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/computer_security_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/computer_security_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/conceptual_physics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/conceptual_physics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/econometrics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/econometrics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/electrical_engineering_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/electrical_engineering_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/elementary_mathematics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/elementary_mathematics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/formal_logic_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/formal_logic_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/global_facts_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/global_facts_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_biology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_biology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_chemistry_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_chemistry_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_computer_science_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_computer_science_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_european_history_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_european_history_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_geography_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_geography_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_government_and_politics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_government_and_politics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_macroeconomics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_macroeconomics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_mathematics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_mathematics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_microeconomics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_microeconomics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_physics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_physics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_psychology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_psychology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_statistics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_statistics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_us_history_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_us_history_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/high_school_world_history_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/high_school_world_history_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/human_aging_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/human_aging_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/human_sexuality_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/human_sexuality_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/international_law_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/international_law_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/jurisprudence_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/jurisprudence_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/logical_fallacies_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/logical_fallacies_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/machine_learning_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/machine_learning_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/management_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/management_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/marketing_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/marketing_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/medical_genetics_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/medical_genetics_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/miscellaneous_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/miscellaneous_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/moral_disputes_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/moral_disputes_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/moral_scenarios_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/moral_scenarios_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/nutrition_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/nutrition_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/philosophy_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/philosophy_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/prehistory_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/prehistory_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/professional_accounting_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/professional_accounting_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/professional_law_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/professional_law_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/professional_medicine_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/professional_medicine_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/professional_psychology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/professional_psychology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/public_relations_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/public_relations_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/security_studies_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/security_studies_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/sociology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/sociology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/us_foreign_policy_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/us_foreign_policy_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/virology_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/virology_test.csv -------------------------------------------------------------------------------- /data/mmlu/test/world_religions_test.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/test/world_religions_test.csv -------------------------------------------------------------------------------- /data/mmlu/val/abstract_algebra_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/abstract_algebra_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/anatomy_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/anatomy_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/astronomy_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/astronomy_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/business_ethics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/business_ethics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/clinical_knowledge_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/clinical_knowledge_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_biology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_biology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_chemistry_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_chemistry_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_computer_science_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_computer_science_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_mathematics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_mathematics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_medicine_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_medicine_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/college_physics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/college_physics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/computer_security_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/computer_security_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/conceptual_physics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/conceptual_physics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/econometrics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/econometrics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/electrical_engineering_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/electrical_engineering_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/elementary_mathematics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/elementary_mathematics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/formal_logic_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/formal_logic_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/global_facts_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/global_facts_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_biology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_biology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_chemistry_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_chemistry_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_computer_science_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_computer_science_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_european_history_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_european_history_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_geography_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_geography_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_government_and_politics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_government_and_politics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_macroeconomics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_macroeconomics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_mathematics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_mathematics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_microeconomics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_microeconomics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_physics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_physics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_psychology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_psychology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_statistics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_statistics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_us_history_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_us_history_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/high_school_world_history_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/high_school_world_history_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/human_aging_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/human_aging_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/human_sexuality_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/human_sexuality_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/international_law_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/international_law_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/jurisprudence_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/jurisprudence_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/logical_fallacies_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/logical_fallacies_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/machine_learning_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/machine_learning_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/management_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/management_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/marketing_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/marketing_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/medical_genetics_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/medical_genetics_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/miscellaneous_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/miscellaneous_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/moral_disputes_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/moral_disputes_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/moral_scenarios_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/moral_scenarios_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/nutrition_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/nutrition_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/philosophy_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/philosophy_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/prehistory_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/prehistory_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/professional_accounting_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/professional_accounting_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/professional_law_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/professional_law_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/professional_medicine_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/professional_medicine_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/professional_psychology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/professional_psychology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/public_relations_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/public_relations_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/security_studies_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/security_studies_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/sociology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/sociology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/us_foreign_policy_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/us_foreign_policy_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/virology_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/virology_val.csv -------------------------------------------------------------------------------- /data/mmlu/val/world_religions_val.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/mmlu/val/world_religions_val.csv -------------------------------------------------------------------------------- /data/simple_safety_tests/simple_safety_tests.csv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/data/simple_safety_tests/simple_safety_tests.csv -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/pyproject.toml -------------------------------------------------------------------------------- /scripts/alpaca_eval_vllm_llama3_3_70b_fn/alpaca_eval_fn.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/scripts/alpaca_eval_vllm_llama3_3_70b_fn/alpaca_eval_fn.txt -------------------------------------------------------------------------------- /scripts/alpaca_eval_vllm_llama3_3_70b_fn/configs.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/scripts/alpaca_eval_vllm_llama3_3_70b_fn/configs.yaml -------------------------------------------------------------------------------- /scripts/evaluate_safety.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/scripts/evaluate_safety.py -------------------------------------------------------------------------------- /test_and_make_submission.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/test_and_make_submission.sh -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_entropy.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_entropy.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_group_normalized_rewards_no_normalize_by_std.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_group_normalized_rewards_no_normalize_by_std.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_group_normalized_rewards_normalize_by_std.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_group_normalized_rewards_normalize_by_std.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_grpo_clip_loss_large_cliprange.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_grpo_clip_loss_large_cliprange.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_grpo_clip_loss_small_cliprange.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_grpo_clip_loss_small_cliprange.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_naive_policy_gradient_loss.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_naive_policy_gradient_loss.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_policy_gradient_loss_grpo_clip.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_policy_gradient_loss_grpo_clip.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_policy_gradient_loss_no_baseline.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_policy_gradient_loss_no_baseline.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_compute_policy_gradient_loss_reinforce_with_baseline.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_compute_policy_gradient_loss_reinforce_with_baseline.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_get_response_log_probs.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_get_response_log_probs.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_grpo_microbatch_train_step_grpo_clip.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_grpo_microbatch_train_step_grpo_clip.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_grpo_microbatch_train_step_grpo_clip_10_steps.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_grpo_microbatch_train_step_grpo_clip_10_steps.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_mean_dim0.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_mean_dim0.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_mean_dim1.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_mean_dim1.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_mean_dimNone.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_mean_dimNone.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_mean_dimlast.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_mean_dimlast.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_normalize_dim0.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_normalize_dim0.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_normalize_dim1.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_normalize_dim1.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_normalize_dimNone.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_normalize_dimNone.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_masked_normalize_dimlast.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_masked_normalize_dimlast.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_sft_microbatch_train_step.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_sft_microbatch_train_step.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_sft_microbatch_train_step_10_steps.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_sft_microbatch_train_step_10_steps.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_sft_microbatch_train_step_normalize.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_sft_microbatch_train_step_normalize.npz -------------------------------------------------------------------------------- /tests/_snapshots/test_tokenize_prompt_and_output.npz: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/_snapshots/test_tokenize_prompt_and_output.npz -------------------------------------------------------------------------------- /tests/adapters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/adapters.py -------------------------------------------------------------------------------- /tests/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/common.py -------------------------------------------------------------------------------- /tests/conftest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/conftest.py -------------------------------------------------------------------------------- /tests/fixtures/.gitkeep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/fixtures/Meta-Llama-3-8B/special_tokens_map.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/Meta-Llama-3-8B/special_tokens_map.json -------------------------------------------------------------------------------- /tests/fixtures/Meta-Llama-3-8B/tokenizer.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/Meta-Llama-3-8B/tokenizer.json -------------------------------------------------------------------------------- /tests/fixtures/Meta-Llama-3-8B/tokenizer_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/Meta-Llama-3-8B/tokenizer_config.json -------------------------------------------------------------------------------- /tests/fixtures/sft_sample.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/sft_sample.jsonl -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2-ref/config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2-ref/config.json -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2-ref/generation_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2-ref/generation_config.json -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2-ref/model.safetensors: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2-ref/model.safetensors -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2/config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2/config.json -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2/generation_config.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2/generation_config.json -------------------------------------------------------------------------------- /tests/fixtures/tiny-gpt2/model.safetensors: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tiny-gpt2/model.safetensors -------------------------------------------------------------------------------- /tests/fixtures/tokenized_sft_sample.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/fixtures/tokenized_sft_sample.json -------------------------------------------------------------------------------- /tests/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/test_data.py -------------------------------------------------------------------------------- /tests/test_dpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/test_dpo.py -------------------------------------------------------------------------------- /tests/test_grpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/test_grpo.py -------------------------------------------------------------------------------- /tests/test_metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/test_metrics.py -------------------------------------------------------------------------------- /tests/test_sft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/tests/test_sft.py -------------------------------------------------------------------------------- /uv.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-cs336/assignment5-alignment/HEAD/uv.lock --------------------------------------------------------------------------------