{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 1.4528301886792452, "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": 329.8125, "epoch": 0.1509433962264151, "grad_norm": 17.191255569458008, "kl": 0.0, "learning_rate": 5e-07, "loss": -0.0, "reward": 0.039155628241132945, "reward_std": 0.01428457984002307, "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": 0.039155628241132945, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 2 }, { "completion_length": 243.875, "epoch": 0.3018867924528302, "grad_norm": 38.74528121948242, "kl": 0.0006890344702696893, "learning_rate": 4.864543104251586e-07, "loss": 0.0, "reward": 0.1348669819708448, "reward_std": 0.1310532387069543, "rewards/concensus_correctness_reward_func": -0.05525000020861626, "rewards/consensus_reward_func": 0.125, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.0, "rewards/question_recreation_reward_func": 0.06511698590475135, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 4 }, { "completion_length": 125.4375, "epoch": 0.4528301886792453, "grad_norm": 11.887129783630371, "kl": 0.0008054986174101941, "learning_rate": 4.472851273490984e-07, "loss": 0.0, "reward": 0.03981615405064076, "reward_std": 0.0724839338217862, "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": 0.07550365466158837, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": -0.03568749874830246, "step": 6 }, { "completion_length": 221.625, "epoch": 0.6037735849056604, "grad_norm": 10.825657844543457, "kl": 0.0008960797931649722, "learning_rate": 3.867370395306068e-07, "loss": 0.0, "reward": 0.169917072635144, "reward_std": 0.20294732836191542, "rewards/concensus_correctness_reward_func": 0.0, "rewards/consensus_reward_func": 0.0, "rewards/cumulative_reward_2": 0.0, "rewards/final_correctness_reward_func": 0.125, "rewards/question_recreation_reward_func": 0.03704206692054868, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.00787500012665987, "step": 8 }, { "completion_length": 121.875, "epoch": 0.7547169811320755, "grad_norm": 12.769948959350586, "kl": 0.0007761767992633395, "learning_rate": 3.1137137178519977e-07, "loss": 0.0, "reward": 0.05784426745958626, "reward_std": 0.01804652114469718, "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": 0.05784426745958626, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 10 }, { "completion_length": 236.6875, "epoch": 0.9056603773584906, "grad_norm": 23.49297523498535, "kl": 0.0008974826741905417, "learning_rate": 2.2935516363191693e-07, "loss": 0.0, "reward": 0.06572715390939265, "reward_std": 0.021446246337291086, "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": 0.06572715390939265, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 12 }, { "completion_length": 236.3, "epoch": 1.0, "grad_norm": 5.101478576660156, "kl": 0.0007671493338420987, "learning_rate": 1.4957614383675767e-07, "loss": 0.0, "reward": 0.11197975892573594, "reward_std": 0.02665466726757586, "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": 0.11197975892573594, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 14 }, { "completion_length": 111.1875, "epoch": 1.150943396226415, "grad_norm": 32.584754943847656, "kl": 0.0011245726636843756, "learning_rate": 8.067960709356478e-08, "loss": 0.0, "reward": 0.055528199998661876, "reward_std": 0.016688317846274003, "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": 0.055528199998661876, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 16 }, { "completion_length": 228.375, "epoch": 1.3018867924528301, "grad_norm": 23.378738403320312, "kl": 0.0014287719095591456, "learning_rate": 3.013156219837776e-08, "loss": 0.0, "reward": 0.03964615031145513, "reward_std": 0.013994272749187076, "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": 0.03964615031145513, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 18 }, { "completion_length": 212.4375, "epoch": 1.4528301886792452, "grad_norm": 7.405801296234131, "kl": 0.0007980182890605647, "learning_rate": 3.4096741493194193e-09, "loss": 0.0, "reward": 0.034015504526905715, "reward_std": 0.0055219637652044185, "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": 0.034015504526905715, "rewards/soft_format_reward_func": 0.0, "rewards/strict_format_reward_func": 0.0, "rewards/xmlcount_reward_func": 0.0, "step": 20 }, { "epoch": 1.4528301886792452, "step": 20, "total_flos": 0.0, "train_loss": 8.887580065675138e-07, "train_runtime": 65076.7906, "train_samples_per_second": 0.002, "train_steps_per_second": 0.0 } ], "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 }