| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.7272727272727275, | |
| "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": 376.34375, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 2.123795509338379, | |
| "kl": 0.0008059284700721037, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.8194242492318153, | |
| "reward_std": 0.8725666627287865, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4507679883390665, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.040531257865950465, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 463.57142857142856, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.444406121969223, | |
| "kl": 0.0006558234370978815, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 1.0296695147241866, | |
| "reward_std": 0.8633963976587568, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.23541951019849097, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": 0.5442500114440918, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 598.5, | |
| "epoch": 2.0, | |
| "grad_norm": 0.40745630860328674, | |
| "kl": 0.0007307843105601412, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.7694277209894997, | |
| "reward_std": 1.1632391554968697, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.2857142857142857, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5034991694348199, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.037642872759274075, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 560.1875, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.2870396077632904, | |
| "kl": 0.0006468838182627223, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.7364773657172918, | |
| "reward_std": 1.0139315500855446, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.3468836098909378, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.1239687567576766, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 543.0, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.4253978729248047, | |
| "kl": 0.0007500549545511603, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.5847084107143539, | |
| "reward_std": 1.0092563756874628, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.42106556040900095, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.06849999725818634, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 533.6428571428571, | |
| "epoch": 4.0, | |
| "grad_norm": 0.30162712931632996, | |
| "kl": 0.0008266470395028591, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.4913223215511867, | |
| "reward_std": 0.9674114840371268, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.40028661489486694, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.0017500030142920358, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 566.34375, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.6862844824790955, | |
| "kl": 0.0007115659973351285, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 0.9112295396625996, | |
| "reward_std": 1.0470195040106773, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4396982938051224, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.26840626634657383, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 535.8571428571429, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.2605418860912323, | |
| "kl": 0.0005985684318667543, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.9383800370352608, | |
| "reward_std": 0.9764512436730521, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.42352288003478733, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.28271428868174553, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 533.7857142857143, | |
| "epoch": 6.0, | |
| "grad_norm": 3.9433951377868652, | |
| "kl": 0.0008690485265105963, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 1.395365025315966, | |
| "reward_std": 1.3007023589951652, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.3816864735313824, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.07142857142857142, | |
| "rewards/xmlcount_reward_func": 0.4422500016433852, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 471.875, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.6051961183547974, | |
| "kl": 0.0007186525617726147, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 0.9024831224232912, | |
| "reward_std": 1.044289842247963, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4804831203073263, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.17199999652802944, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 6.874654843613825e-07, | |
| "train_runtime": 1539.8882, | |
| "train_samples_per_second": 0.208, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |