| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.7272727272727275, | |
| "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": 489.90625, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.3609657883644104, | |
| "kl": 0.0006489891966339201, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.828802197240293, | |
| "reward_std": 0.7736977692693472, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5462084300816059, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.22009374480694532, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 511.5, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.42430993914604187, | |
| "kl": 0.0007115530482094203, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.6244428072656903, | |
| "reward_std": 1.1362362929752894, | |
| "rewards/concensus_correctness_reward_func": 0.017857142857142856, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4853356395448957, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.021607141941785812, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 464.39285714285717, | |
| "epoch": 2.0, | |
| "grad_norm": 0.6451336741447449, | |
| "kl": 0.000845598169170054, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 1.5903914826256889, | |
| "reward_std": 2.294496755514826, | |
| "rewards/concensus_correctness_reward_func": 0.7411428572876113, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.14285714285714285, | |
| "rewards/question_recreation_reward_func": 0.5562128743955067, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.06410714345318931, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 420.1875, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.4900331199169159, | |
| "kl": 0.0007078720445861109, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.8949470296502113, | |
| "reward_std": 0.9619810990989208, | |
| "rewards/concensus_correctness_reward_func": 0.03125, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6426657810807228, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.028968743979930878, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 521.5, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.43673744797706604, | |
| "kl": 0.0006336758794662144, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.8506239099161965, | |
| "reward_std": 0.8674267274992806, | |
| "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": 0.6670881807804108, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": 0.1478214327778135, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 420.42857142857144, | |
| "epoch": 4.0, | |
| "grad_norm": 0.4140712022781372, | |
| "kl": 0.0008527751030799534, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.8269651840840068, | |
| "reward_std": 0.8055251027856555, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4500009013073785, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.21624999386923655, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 481.0625, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.42424920201301575, | |
| "kl": 0.0006703631843265612, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 1.1690492890775204, | |
| "reward_std": 0.9069762676954269, | |
| "rewards/concensus_correctness_reward_func": 0.015625, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6187680251896381, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.22215624619275331, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 423.75, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.4367072582244873, | |
| "kl": 0.0024985303919363233, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.7280806686197009, | |
| "reward_std": 0.8712436748402459, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.07142857142857142, | |
| "rewards/question_recreation_reward_func": 0.4392235449382237, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.05671428915645395, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 596.2857142857143, | |
| "epoch": 6.0, | |
| "grad_norm": 0.19180385768413544, | |
| "kl": 0.0006549531244672835, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 2.444173582962581, | |
| "reward_std": 2.860402743731226, | |
| "rewards/concensus_correctness_reward_func": 1.4642857142857142, | |
| "rewards/consensus_reward_func": 0.35714285714285715, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.07142857142857142, | |
| "rewards/question_recreation_reward_func": 0.6413163840770721, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": -0.12571429354803904, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 451.1875, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.5098544955253601, | |
| "kl": 0.0009375976696901489, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.6025606133043766, | |
| "reward_std": 2.088516917079687, | |
| "rewards/concensus_correctness_reward_func": 0.6485000001266599, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.555873055011034, | |
| "rewards/soft_format_reward_func": 0.015625, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.10131250135600567, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 8.487639661325374e-07, | |
| "train_runtime": 596.4258, | |
| "train_samples_per_second": 0.537, | |
| "train_steps_per_second": 0.034 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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 | |
| } | |