| { | |
| "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": 444.8125, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.6776350736618042, | |
| "kl": 0.0006790693114453461, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.8109659971669316, | |
| "reward_std": 1.1016927361488342, | |
| "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": 0.46102848276495934, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.09993750229477882, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 439.67857142857144, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 2.783291816711426, | |
| "kl": 0.0008688629979068148, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.8720081704003471, | |
| "reward_std": 1.268479632479804, | |
| "rewards/concensus_correctness_reward_func": 0.05357142857142857, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.2665795949952943, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.3911428579262325, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 590.75, | |
| "epoch": 2.0, | |
| "grad_norm": 0.3263286054134369, | |
| "kl": 0.0007448658746268068, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.9567571394145489, | |
| "reward_std": 1.2566224847521101, | |
| "rewards/concensus_correctness_reward_func": 0.10714285714285714, | |
| "rewards/consensus_reward_func": 0.2857142857142857, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5476142657654626, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.0015714189835957118, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 514.90625, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.3379019498825073, | |
| "kl": 0.0007241862476803362, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.857398571446538, | |
| "reward_std": 0.904921617358923, | |
| "rewards/concensus_correctness_reward_func": 0.046875, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.3806798132136464, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.304843757301569, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 465.42857142857144, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.3748963475227356, | |
| "kl": 0.000740903928609831, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 1.3108210904257638, | |
| "reward_std": 1.1951951129095895, | |
| "rewards/concensus_correctness_reward_func": 0.05357142857142857, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.48092824646404814, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.3298928588628769, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 501.82142857142856, | |
| "epoch": 4.0, | |
| "grad_norm": 0.3151821196079254, | |
| "kl": 0.0006694530685698348, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.9621058659894126, | |
| "reward_std": 0.8413353604929787, | |
| "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": 0.47399873925106867, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.34525001687662943, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 457.0625, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.5541585087776184, | |
| "kl": 0.0007372664695139974, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 1.0042262449860573, | |
| "reward_std": 1.0684067830443382, | |
| "rewards/concensus_correctness_reward_func": 0.046875, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5084137637168169, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.2301875101402402, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 566.25, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.3904951810836792, | |
| "kl": 0.000746933140492599, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.9180152405585561, | |
| "reward_std": 1.3828005152089256, | |
| "rewards/concensus_correctness_reward_func": 0.05357142857142857, | |
| "rewards/consensus_reward_func": 0.2857142857142857, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.47983670394335476, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.0988928462777819, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 513.75, | |
| "epoch": 6.0, | |
| "grad_norm": 24.226089477539062, | |
| "kl": 0.0013232028743784343, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 1.3621586774076735, | |
| "reward_std": 1.0136123959507262, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5017657790865216, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.43182141866002766, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 510.28125, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.4133526384830475, | |
| "kl": 0.0007042329089017585, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.2120915539562702, | |
| "reward_std": 1.1693911626935005, | |
| "rewards/concensus_correctness_reward_func": 0.140625, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.48521655425429344, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.2737499978393316, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 7.408759415739041e-07, | |
| "train_runtime": 1619.8213, | |
| "train_samples_per_second": 0.198, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "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 | |
| } | |