| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 238.15625, | |
| "epoch": 0.1, | |
| "grad_norm": 22.03662872314453, | |
| "kl": 0.0, | |
| "learning_rate": 4.965903258506806e-07, | |
| "loss": 0.0, | |
| "reward": 6.236779972910881, | |
| "reward_std": 1.9459177672688384, | |
| "rewards/concensus_correctness_reward_func": 2.3227499946951866, | |
| "rewards/consensus_reward_func": 0.9375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.1875, | |
| "rewards/question_recreation_reward_func": 0.7343113259412348, | |
| "rewards/soft_format_reward_func": 0.015625, | |
| "rewards/strict_format_reward_func": 0.203125, | |
| "rewards/xmlcount_reward_func": 0.8359687551856041, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 216.25, | |
| "epoch": 0.2, | |
| "grad_norm": 145.5101318359375, | |
| "kl": 209.9995388824027, | |
| "learning_rate": 4.698684378016222e-07, | |
| "loss": 0.21, | |
| "reward": 7.817449122667313, | |
| "reward_std": 2.4175750967115164, | |
| "rewards/concensus_correctness_reward_func": 2.5938750002533197, | |
| "rewards/consensus_reward_func": 1.4375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.375, | |
| "rewards/question_recreation_reward_func": 0.8879492152482271, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.28125, | |
| "rewards/xmlcount_reward_func": 1.2418750077486038, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 224.34375, | |
| "epoch": 0.3, | |
| "grad_norm": 35.77546691894531, | |
| "kl": 219181.7316086035, | |
| "learning_rate": 4.193203929064353e-07, | |
| "loss": 219.1817, | |
| "reward": 8.467853605747223, | |
| "reward_std": 1.2415645032306202, | |
| "rewards/concensus_correctness_reward_func": 2.838187515735626, | |
| "rewards/consensus_reward_func": 1.625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.5, | |
| "rewards/question_recreation_reward_func": 0.9661347344517708, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.296875, | |
| "rewards/xmlcount_reward_func": 1.2416562661528587, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 226.375, | |
| "epoch": 0.4, | |
| "grad_norm": 1546.4365234375, | |
| "kl": 564.3714897963218, | |
| "learning_rate": 3.5042385616324236e-07, | |
| "loss": 0.5644, | |
| "reward": 7.87619161605835, | |
| "reward_std": 1.734267087711487, | |
| "rewards/concensus_correctness_reward_func": 2.39962500333786, | |
| "rewards/consensus_reward_func": 1.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.4375, | |
| "rewards/question_recreation_reward_func": 0.9164416156709194, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.375, | |
| "rewards/xmlcount_reward_func": 1.247625008225441, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 219.3125, | |
| "epoch": 0.5, | |
| "grad_norm": 13.214804649353027, | |
| "kl": 13.398893604520708, | |
| "learning_rate": 2.706448363680831e-07, | |
| "loss": 0.0134, | |
| "reward": 9.186098158359528, | |
| "reward_std": 0.6499924267409369, | |
| "rewards/concensus_correctness_reward_func": 3.229625016450882, | |
| "rewards/consensus_reward_func": 1.8125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.625, | |
| "rewards/question_recreation_reward_func": 0.913348164409399, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.359375, | |
| "rewards/xmlcount_reward_func": 1.2462500035762787, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 210.78125, | |
| "epoch": 0.6, | |
| "grad_norm": 15.455207824707031, | |
| "kl": 1.3405293300747871, | |
| "learning_rate": 1.886286282148002e-07, | |
| "loss": 0.0013, | |
| "reward": 8.247979655861855, | |
| "reward_std": 0.8561342508182861, | |
| "rewards/concensus_correctness_reward_func": 2.890375006245449, | |
| "rewards/consensus_reward_func": 1.5625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.375, | |
| "rewards/question_recreation_reward_func": 0.8725736476480961, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.328125, | |
| "rewards/xmlcount_reward_func": 1.219406258314848, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 205.75, | |
| "epoch": 0.7, | |
| "grad_norm": 24.68694305419922, | |
| "kl": 2.3243863766547292, | |
| "learning_rate": 1.1326296046939333e-07, | |
| "loss": 0.0023, | |
| "reward": 7.388011336326599, | |
| "reward_std": 2.6222411412745714, | |
| "rewards/concensus_correctness_reward_func": 2.4560000002384186, | |
| "rewards/consensus_reward_func": 1.4375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.1875, | |
| "rewards/question_recreation_reward_func": 0.7821362968534231, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.296875, | |
| "rewards/xmlcount_reward_func": 1.2280000075697899, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 210.5625, | |
| "epoch": 0.8, | |
| "grad_norm": 14.628301620483398, | |
| "kl": 104.65415175596718, | |
| "learning_rate": 5.271487265090163e-08, | |
| "loss": 0.1047, | |
| "reward": 7.575839355587959, | |
| "reward_std": 0.9005096781183966, | |
| "rewards/concensus_correctness_reward_func": 2.464500017464161, | |
| "rewards/consensus_reward_func": 1.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.375, | |
| "rewards/question_recreation_reward_func": 0.9392767958343029, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.1875, | |
| "rewards/xmlcount_reward_func": 1.2345625162124634, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 206.25, | |
| "epoch": 0.9, | |
| "grad_norm": 16.712093353271484, | |
| "kl": 2.9178055827505887, | |
| "learning_rate": 1.3545689574841341e-08, | |
| "loss": 0.0029, | |
| "reward": 8.665296122431755, | |
| "reward_std": 0.8437674434389919, | |
| "rewards/concensus_correctness_reward_func": 3.1917499974370003, | |
| "rewards/consensus_reward_func": 1.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.625, | |
| "rewards/question_recreation_reward_func": 0.8036086433567107, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.3125, | |
| "rewards/xmlcount_reward_func": 1.2324374988675117, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 233.28125, | |
| "epoch": 1.0, | |
| "grad_norm": 10.762929916381836, | |
| "kl": 0.213681883062236, | |
| "learning_rate": 0.0, | |
| "loss": 0.0002, | |
| "reward": 7.851729407906532, | |
| "reward_std": 0.8491827086254489, | |
| "rewards/concensus_correctness_reward_func": 2.6363750025629997, | |
| "rewards/consensus_reward_func": 1.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.25, | |
| "rewards/question_recreation_reward_func": 0.9363544508814812, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.28125, | |
| "rewards/xmlcount_reward_func": 1.2477500140666962, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 22.00808893401554, | |
| "train_runtime": 313.0108, | |
| "train_samples_per_second": 1.022, | |
| "train_steps_per_second": 0.064 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |