| { | |
| "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": 476.0, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.5276138186454773, | |
| "kl": 0.0006918963335920125, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.8905868269503117, | |
| "reward_std": 0.9817413352429867, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5417430754750967, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.1457187570631504, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 604.5357142857143, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.45205041766166687, | |
| "kl": 0.0006221245400541063, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.9950631601469857, | |
| "reward_std": 0.8837312459945679, | |
| "rewards/concensus_correctness_reward_func": 0.016285714294229234, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.07142857142857142, | |
| "rewards/question_recreation_reward_func": 0.3275988719293049, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.3654642839516912, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 566.5357142857143, | |
| "epoch": 2.0, | |
| "grad_norm": 0.45169490575790405, | |
| "kl": 0.0007057673730222243, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 1.136856577226094, | |
| "reward_std": 0.7043021747044155, | |
| "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": 0.5205351114273071, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.45560715028217863, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 435.03125, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.4807434678077698, | |
| "kl": 0.0007996001113497186, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.895195696502924, | |
| "reward_std": 0.9222910553216934, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4469144530594349, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.18265624903142452, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 514.0, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.45199769735336304, | |
| "kl": 0.0006535403995907732, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.5129805675574711, | |
| "reward_std": 1.017007691519601, | |
| "rewards/concensus_correctness_reward_func": 0.0005714285985699721, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5081234255007335, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.08499999557222639, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 497.10714285714283, | |
| "epoch": 4.0, | |
| "grad_norm": 0.38202306628227234, | |
| "kl": 0.0007072919480768698, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.6489094985382897, | |
| "reward_std": 0.9131356051989964, | |
| "rewards/concensus_correctness_reward_func": 0.029857143759727478, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.39351664057799746, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.06482143008283206, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 387.15625, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.8552425503730774, | |
| "kl": 0.0007712517472100444, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 0.8629500679671764, | |
| "reward_std": 0.7735467366874218, | |
| "rewards/concensus_correctness_reward_func": 0.0005000000237487257, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.42723130341619253, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.37271876260638237, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 518.0, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.41807857155799866, | |
| "kl": 0.000663631537463516, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.6208901362759727, | |
| "reward_std": 0.9498651453426906, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5296044392245156, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.01985713094472885, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 579.1785714285714, | |
| "epoch": 6.0, | |
| "grad_norm": 0.4610684812068939, | |
| "kl": 0.000690574080051322, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.5093637534550258, | |
| "reward_std": 0.892940702182906, | |
| "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": 0.4273637448038374, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.060857132875493596, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 560.03125, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.38961368799209595, | |
| "kl": 0.0006178304938657675, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.1620018668472767, | |
| "reward_std": 1.1849149875342846, | |
| "rewards/concensus_correctness_reward_func": 0.0010000000474974513, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6256268601864576, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.2228750018402934, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 6.466318950515415e-07, | |
| "train_runtime": 1576.8638, | |
| "train_samples_per_second": 0.203, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "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 | |
| } | |