| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "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": 264.8125, | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.5896669626235962, | |
| "kl": 0.008757207309827209, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 3.556011885404587, | |
| "reward_std": 3.830547973513603, | |
| "rewards/concensus_correctness_reward_func": 1.936812499538064, | |
| "rewards/consensus_reward_func": 0.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.375, | |
| "rewards/question_recreation_reward_func": 0.4231681600213051, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.44603125285357237, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 230.5, | |
| "epoch": 1.0, | |
| "grad_norm": 0.4409516453742981, | |
| "kl": 0.008206194887558619, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 9.294309139251709, | |
| "reward_std": 8.018560727437338, | |
| "rewards/concensus_correctness_reward_func": 6.433000023166339, | |
| "rewards/consensus_reward_func": 0.6666666666666666, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.0833333333333333, | |
| "rewards/question_recreation_reward_func": 0.4835590223471324, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.041666666666666664, | |
| "rewards/xmlcount_reward_func": 0.586083342631658, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 252.96875, | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 0.809605062007904, | |
| "kl": 0.00894307589624077, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 5.214481353759766, | |
| "reward_std": 5.1277751959860325, | |
| "rewards/concensus_correctness_reward_func": 3.1976874992251396, | |
| "rewards/consensus_reward_func": 0.4375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.625, | |
| "rewards/question_recreation_reward_func": 0.4166688732802868, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.5376249849796295, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 289.0833333333333, | |
| "epoch": 2.0, | |
| "grad_norm": 0.4980953633785248, | |
| "kl": 0.007897172511244813, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 5.711512823899587, | |
| "reward_std": 6.444633464018504, | |
| "rewards/concensus_correctness_reward_func": 3.68374993900458, | |
| "rewards/consensus_reward_func": 0.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.75, | |
| "rewards/question_recreation_reward_func": 0.45838790635267895, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.31937499592701596, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 252.1875, | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.6080841422080994, | |
| "kl": 0.010039961838629097, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 4.326304629445076, | |
| "reward_std": 4.895156145095825, | |
| "rewards/concensus_correctness_reward_func": 2.0929999724030495, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.0, | |
| "rewards/question_recreation_reward_func": 0.4917107969522476, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.4603437501937151, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 265.25, | |
| "epoch": 3.0, | |
| "grad_norm": 0.42046964168548584, | |
| "kl": 0.007893032704790434, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 5.077903230985005, | |
| "reward_std": 4.715359782179196, | |
| "rewards/concensus_correctness_reward_func": 2.6067499990264573, | |
| "rewards/consensus_reward_func": 0.5833333333333334, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.6666666666666666, | |
| "rewards/question_recreation_reward_func": 0.5395283401012421, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0625, | |
| "rewards/xmlcount_reward_func": 0.6191250036160151, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 260.625, | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.7956599593162537, | |
| "kl": 0.007486780465114862, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 5.287028789520264, | |
| "reward_std": 6.096508968621492, | |
| "rewards/concensus_correctness_reward_func": 2.754937469959259, | |
| "rewards/consensus_reward_func": 0.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 1.0625, | |
| "rewards/question_recreation_reward_func": 0.6092475522309542, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.4697187477722764, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 251.04166666666666, | |
| "epoch": 4.0, | |
| "grad_norm": 0.6040255427360535, | |
| "kl": 0.011015481315553188, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 4.457539727290471, | |
| "reward_std": 2.4467248568932214, | |
| "rewards/concensus_correctness_reward_func": 2.63541666418314, | |
| "rewards/consensus_reward_func": 0.5833333333333334, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.3333333333333333, | |
| "rewards/question_recreation_reward_func": 0.22862321510910988, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.041666666666666664, | |
| "rewards/xmlcount_reward_func": 0.6351666748523712, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 275.34375, | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.697487473487854, | |
| "kl": 0.006818667257903144, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 5.823377624154091, | |
| "reward_std": 4.7229442447423935, | |
| "rewards/concensus_correctness_reward_func": 3.429312475025654, | |
| "rewards/consensus_reward_func": 0.5625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.75, | |
| "rewards/question_recreation_reward_func": 0.5220653265714645, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.5282500144094229, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 201.45833333333334, | |
| "epoch": 5.0, | |
| "grad_norm": 0.4893139898777008, | |
| "kl": 0.013442914622525374, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 5.362700551748276, | |
| "reward_std": 4.852403928836186, | |
| "rewards/concensus_correctness_reward_func": 3.37349999944369, | |
| "rewards/consensus_reward_func": 0.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.4166666666666667, | |
| "rewards/question_recreation_reward_func": 0.38161713629961014, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.041666666666666664, | |
| "rewards/xmlcount_reward_func": 0.6492500106493632, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 7.851024111005245e-06, | |
| "train_runtime": 560.7474, | |
| "train_samples_per_second": 0.571, | |
| "train_steps_per_second": 0.036 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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 | |
| } | |