| { | |
| "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": 419.1875, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.4025554955005646, | |
| "kl": 0.0007115886437532026, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 1.2856132220476866, | |
| "reward_std": 0.9523135796189308, | |
| "rewards/concensus_correctness_reward_func": 0.07937500067055225, | |
| "rewards/consensus_reward_func": 0.4375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5550507390871644, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.21368752419948578, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 474.5, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.38703691959381104, | |
| "kl": 0.0005440224652245108, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 1.2591932330812727, | |
| "reward_std": 0.9980412061725344, | |
| "rewards/concensus_correctness_reward_func": 0.05500000076634543, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.630478927067348, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.1451428553887776, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 391.2142857142857, | |
| "epoch": 2.0, | |
| "grad_norm": 0.34758260846138, | |
| "kl": 0.0006707894499413669, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 1.0978498458862305, | |
| "reward_std": 0.7003298039947238, | |
| "rewards/concensus_correctness_reward_func": 0.03571428571428571, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5435998290777206, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.4471071479575975, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 390.40625, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.48531419038772583, | |
| "kl": 0.0007366152312897611, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.70655282959342, | |
| "reward_std": 0.9273619055747986, | |
| "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": 0.48299031239002943, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.046875, | |
| "rewards/xmlcount_reward_func": 0.1141875023022294, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 419.7142857142857, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.4628700911998749, | |
| "kl": 0.0006924757409641254, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 1.183305595602308, | |
| "reward_std": 1.4018447484288896, | |
| "rewards/concensus_correctness_reward_func": 0.1294285718883787, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6321269784654889, | |
| "rewards/soft_format_reward_func": 0.03571428571428571, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.060392867241586955, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 386.57142857142856, | |
| "epoch": 4.0, | |
| "grad_norm": 0.5434279441833496, | |
| "kl": 0.0006527610489034227, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 0.7653476042406899, | |
| "reward_std": 0.8373887177024569, | |
| "rewards/concensus_correctness_reward_func": 0.030714286225182668, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5036690256425312, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": -0.001178561576775142, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 386.3125, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.4393087327480316, | |
| "kl": 0.0008711709633644205, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 1.2212899252772331, | |
| "reward_std": 1.1120201498270035, | |
| "rewards/concensus_correctness_reward_func": 0.048125000670552254, | |
| "rewards/consensus_reward_func": 0.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5600711535662413, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.23809375427663326, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 499.17857142857144, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.4070427715778351, | |
| "kl": 0.000641536241995969, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.622774315731866, | |
| "reward_std": 1.1508670066084181, | |
| "rewards/concensus_correctness_reward_func": 0.015357143112591334, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5319528803229332, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.004035707030977521, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 438.75, | |
| "epoch": 6.0, | |
| "grad_norm": 0.38713639974594116, | |
| "kl": 0.0006755495851393789, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.9099430825029101, | |
| "reward_std": 1.0716484870229448, | |
| "rewards/concensus_correctness_reward_func": 0.045357143240315576, | |
| "rewards/consensus_reward_func": 0.21428571428571427, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5349787814276559, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.11532143184116908, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 378.78125, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.44257575273513794, | |
| "kl": 0.0006841454087407328, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.4425083473324776, | |
| "reward_std": 1.0375258475542068, | |
| "rewards/concensus_correctness_reward_func": 0.048125000670552254, | |
| "rewards/consensus_reward_func": 0.3125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.6988520696759224, | |
| "rewards/soft_format_reward_func": 0.015625, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.36740625742822886, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 6.468071546805732e-07, | |
| "train_runtime": 564.7638, | |
| "train_samples_per_second": 0.567, | |
| "train_steps_per_second": 0.035 | |
| } | |
| ], | |
| "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 | |
| } | |