{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 1.8571428571428572, "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": 229.25, "epoch": 0.19047619047619047, "grad_norm": 10.515012741088867, "kl": 0.0, "learning_rate": 5e-07, "loss": -0.0, "reward": 0.0650738068507053, "reward_std": 0.03761709766695276, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.06916755676502362, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": -0.004093749914318323, "step": 2 }, { "completion_length": 191.28125, "epoch": 0.38095238095238093, "grad_norm": 9.014220237731934, "kl": 0.0018629569822223857, "learning_rate": 4.864543104251586e-07, "loss": 0.0, "reward": 0.07388911413727328, "reward_std": 0.0260261885705404, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.07388911413727328, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 4 }, { "completion_length": 170.78125, "epoch": 0.5714285714285714, "grad_norm": 6.390993595123291, "kl": 0.004871747754805256, "learning_rate": 4.472851273490984e-07, "loss": 0.0, "reward": 0.06037694658152759, "reward_std": 0.019421471472014673, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.06037694658152759, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 6 }, { "completion_length": 196.46875, "epoch": 0.7619047619047619, "grad_norm": 20.86431884765625, "kl": 0.02937794415629469, "learning_rate": 3.867370395306068e-07, "loss": 0.0, "reward": 0.08842340856790543, "reward_std": 0.03395858586009126, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.08842340856790543, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 8 }, { "completion_length": 178.96875, "epoch": 0.9523809523809523, "grad_norm": 10.157696723937988, "kl": 0.021908172624534927, "learning_rate": 3.1137137178519977e-07, "loss": 0.0, "reward": 0.05431336216861382, "reward_std": 0.021931443054199917, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.05431336216861382, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 10 }, { "completion_length": 268.5416666666667, "epoch": 1.0952380952380953, "grad_norm": 10.279065132141113, "kl": 0.022836394452800352, "learning_rate": 2.2935516363191693e-07, "loss": 0.0, "reward": 0.066770395807301, "reward_std": 0.027234835938240092, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.066770395807301, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 12 }, { "completion_length": 195.34375, "epoch": 1.2857142857142856, "grad_norm": 7.825338363647461, "kl": 0.02000607246736763, "learning_rate": 1.4957614383675767e-07, "loss": 0.0, "reward": 0.08899017563089728, "reward_std": 0.023486860693083145, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.09136517532169819, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": -0.002374999923631549, "step": 14 }, { "completion_length": 206.5625, "epoch": 1.4761904761904763, "grad_norm": 9.047295570373535, "kl": 0.028867189641459845, "learning_rate": 8.067960709356478e-08, "loss": 0.0, "reward": 0.08858426820370369, "reward_std": 0.030635877163149416, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.08311551829683594, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0054687499068677425, "step": 16 }, { "completion_length": 264.59375, "epoch": 1.6666666666666665, "grad_norm": 85.33262634277344, "kl": 0.02868098636099603, "learning_rate": 3.013156219837776e-08, "loss": 0.0, "reward": 0.1200778468628414, "reward_std": 0.1210625793246436, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0625, "rewards/question_recreation_reward_func": 0.04648409254150465, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.011093749664723873, "step": 18 }, { "completion_length": 244.53125, "epoch": 1.8571428571428572, "grad_norm": 11.419831275939941, "kl": 0.023366850407910533, "learning_rate": 3.4096741493194193e-09, "loss": 0.0, "reward": 0.05546601815149188, "reward_std": 0.023139299628383014, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.05546601815149188, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 20 }, { "epoch": 1.8571428571428572, "step": 20, "total_flos": 0.0, "train_loss": 1.7555055023876776e-05, "train_runtime": 27063.8921, "train_samples_per_second": 0.012, "train_steps_per_second": 0.001 } ], "logging_steps": 2, "max_steps": 20, "num_input_tokens_seen": 0, "num_train_epochs": 2, "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 }