| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5, | |
| "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": 329.96875, | |
| "epoch": 0.05, | |
| "grad_norm": 62.906494140625, | |
| "kl": 0.0, | |
| "learning_rate": 4.965903258506806e-07, | |
| "loss": -0.0, | |
| "reward": 0.3362546949647367, | |
| "reward_std": 0.312244192144135, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.10459844846627675, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.10665625100955367, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 311.4375, | |
| "epoch": 0.1, | |
| "grad_norm": 14.444952011108398, | |
| "kl": 0.0008951859817898367, | |
| "learning_rate": 4.698684378016222e-07, | |
| "loss": 0.0, | |
| "reward": 0.5283035266911611, | |
| "reward_std": 0.3498438004171476, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.1518972598714754, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.12640625098720193, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 260.53125, | |
| "epoch": 0.15, | |
| "grad_norm": 5.720004558563232, | |
| "kl": 0.0010484004160389304, | |
| "learning_rate": 4.193203929064353e-07, | |
| "loss": 0.0, | |
| "reward": 0.6982718562940136, | |
| "reward_std": 0.8361521761980839, | |
| "rewards/concensus_correctness_reward_func": 0.1198124997317791, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.18677185359410942, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.14168749889358878, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 231.59375, | |
| "epoch": 0.2, | |
| "grad_norm": 145.1930389404297, | |
| "kl": 0.02420716641790932, | |
| "learning_rate": 3.5042385616324236e-07, | |
| "loss": 0.0, | |
| "reward": 1.6250686056155246, | |
| "reward_std": 2.0349894912797026, | |
| "rewards/concensus_correctness_reward_func": 1.25, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.12381863036716823, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.0012499995063990355, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 313.875, | |
| "epoch": 0.25, | |
| "grad_norm": 27.593671798706055, | |
| "kl": 0.006836923750597634, | |
| "learning_rate": 2.706448363680831e-07, | |
| "loss": 0.0, | |
| "reward": 1.2748694516485557, | |
| "reward_std": 1.6135549363971222, | |
| "rewards/concensus_correctness_reward_func": 0.625, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.3125, | |
| "rewards/question_recreation_reward_func": 0.17518195346929133, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.16218749893596396, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 317.90625, | |
| "epoch": 0.3, | |
| "grad_norm": 72.23271179199219, | |
| "kl": 0.026576792100968305, | |
| "learning_rate": 1.886286282148002e-07, | |
| "loss": 0.0, | |
| "reward": 0.5719361377414316, | |
| "reward_std": 0.6263423478230834, | |
| "rewards/concensus_correctness_reward_func": 0.06012500077486038, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.18556115147657692, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.013750002020969987, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 283.0625, | |
| "epoch": 0.35, | |
| "grad_norm": 18.375288009643555, | |
| "kl": 0.0015847348440729547, | |
| "learning_rate": 1.1326296046939333e-07, | |
| "loss": 0.0, | |
| "reward": 0.17199375538621098, | |
| "reward_std": 0.32264699795632623, | |
| "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": 0.19643125601578504, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.02443750307429582, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 160.03125, | |
| "epoch": 0.4, | |
| "grad_norm": 53.96257400512695, | |
| "kl": 0.002264712957185111, | |
| "learning_rate": 5.271487265090163e-08, | |
| "loss": 0.0, | |
| "reward": 0.31693001149687916, | |
| "reward_std": 0.18015966298844432, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.11696126533206552, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.07496874779462814, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 397.375, | |
| "epoch": 0.45, | |
| "grad_norm": 29.899606704711914, | |
| "kl": 0.0022588309475395363, | |
| "learning_rate": 1.3545689574841341e-08, | |
| "loss": 0.0, | |
| "reward": 0.6047100863652304, | |
| "reward_std": 0.33923750270332675, | |
| "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": 0.21474133169976994, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.20246875379234552, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 323.8125, | |
| "epoch": 0.5, | |
| "grad_norm": 79.44273376464844, | |
| "kl": 0.1274493236287526, | |
| "learning_rate": 0.0, | |
| "loss": 0.0001, | |
| "reward": 1.3768954614643008, | |
| "reward_std": 1.6876454666489735, | |
| "rewards/concensus_correctness_reward_func": 0.6849999986588955, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.19983295840211213, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.1170625037048012, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 1.9316006137160004e-05, | |
| "train_runtime": 379.4305, | |
| "train_samples_per_second": 0.843, | |
| "train_steps_per_second": 0.053 | |
| } | |
| ], | |
| "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 | |
| } | |