| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8556149732620321, | |
| "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": 268.28125, | |
| "epoch": 0.0855614973262032, | |
| "grad_norm": 67.48136901855469, | |
| "kl": 0.0, | |
| "learning_rate": 4.965903258506806e-07, | |
| "loss": -0.0, | |
| "reward": 0.5082194907590747, | |
| "reward_std": 0.49911471912491834, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.1639694933546707, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.09425000008195639, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 381.5, | |
| "epoch": 0.1711229946524064, | |
| "grad_norm": 9.119170188903809, | |
| "kl": 0.0010306903695891378, | |
| "learning_rate": 4.698684378016222e-07, | |
| "loss": 0.0, | |
| "reward": 0.3929591244086623, | |
| "reward_std": 0.2656565032157232, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.14824039314407855, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.05721874890150502, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 300.53125, | |
| "epoch": 0.25668449197860965, | |
| "grad_norm": 10.421076774597168, | |
| "kl": 0.005060498369857669, | |
| "learning_rate": 4.193203929064353e-07, | |
| "loss": 0.0, | |
| "reward": 0.6220447691739537, | |
| "reward_std": 0.7391133254859596, | |
| "rewards/concensus_correctness_reward_func": 0.05999999865889549, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.15923227864550427, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.0903124981559813, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 362.28125, | |
| "epoch": 0.3422459893048128, | |
| "grad_norm": 74.13516998291016, | |
| "kl": 0.0019698639971466037, | |
| "learning_rate": 3.5042385616324236e-07, | |
| "loss": 0.0, | |
| "reward": 0.38849033904261887, | |
| "reward_std": 0.6596136893494986, | |
| "rewards/concensus_correctness_reward_func": 0.06012500077486038, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.1250215788022615, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.015843751840293407, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 296.84375, | |
| "epoch": 0.42780748663101603, | |
| "grad_norm": 194.94175720214844, | |
| "kl": 0.00394518598477589, | |
| "learning_rate": 2.706448363680831e-07, | |
| "loss": 0.0, | |
| "reward": 1.0231318477308378, | |
| "reward_std": 1.1400645816029282, | |
| "rewards/concensus_correctness_reward_func": 0.625, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.16163179055729415, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.07599999848753214, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 365.78125, | |
| "epoch": 0.5133689839572193, | |
| "grad_norm": 53.124698638916016, | |
| "kl": 0.0022715278173564, | |
| "learning_rate": 1.886286282148002e-07, | |
| "loss": 0.0, | |
| "reward": 0.4706903047626838, | |
| "reward_std": 0.43284516577841714, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.1357527978834696, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.14743749983608723, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 359.71875, | |
| "epoch": 0.5989304812834224, | |
| "grad_norm": 6.279838562011719, | |
| "kl": 0.0014686525682918727, | |
| "learning_rate": 1.1326296046939333e-07, | |
| "loss": 0.0, | |
| "reward": 1.8419055495178327, | |
| "reward_std": 0.6855941953981528, | |
| "rewards/concensus_correctness_reward_func": 1.25, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.16549924900755286, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.07359375429223292, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 256.75, | |
| "epoch": 0.6844919786096256, | |
| "grad_norm": 50.540748596191406, | |
| "kl": 0.0026605648672557436, | |
| "learning_rate": 5.271487265090163e-08, | |
| "loss": 0.0, | |
| "reward": 1.0203204021090642, | |
| "reward_std": 1.244578271463979, | |
| "rewards/concensus_correctness_reward_func": 0.625, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.2058204059721902, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.06449999986216426, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 392.84375, | |
| "epoch": 0.7700534759358288, | |
| "grad_norm": 29.069454193115234, | |
| "kl": 0.00177938278648071, | |
| "learning_rate": 1.3545689574841341e-08, | |
| "loss": 0.0, | |
| "reward": 0.41253009607316926, | |
| "reward_std": 0.6037067175675475, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.13990509719587862, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.03987500024959445, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 324.53125, | |
| "epoch": 0.8556149732620321, | |
| "grad_norm": 56.18510437011719, | |
| "kl": 0.0020320410003478173, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "reward": 0.15064884605817497, | |
| "reward_std": 0.6220636839877898, | |
| "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": 0.17886759783141315, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.15321874804794788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8556149732620321, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 2.2135209292173386e-06, | |
| "train_runtime": 615.4825, | |
| "train_samples_per_second": 0.52, | |
| "train_steps_per_second": 0.032 | |
| } | |
| ], | |
| "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 | |
| } | |