| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.5, | |
| "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": 469.4375, | |
| "epoch": 0.25, | |
| "grad_norm": 3.423996686935425, | |
| "kl": 0.24222635943442583, | |
| "learning_rate": 4.965903258506806e-07, | |
| "loss": 0.0002, | |
| "reward": 2.1027765348553658, | |
| "reward_std": 2.2172846607863903, | |
| "rewards/concensus_correctness_reward_func": 0.625, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.5625, | |
| "rewards/question_recreation_reward_func": 0.23833905532956123, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0625, | |
| "rewards/xmlcount_reward_func": 0.551937498152256, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 552.65625, | |
| "epoch": 0.5, | |
| "grad_norm": 2.1609866619110107, | |
| "kl": 0.15485292114317417, | |
| "learning_rate": 4.698684378016222e-07, | |
| "loss": 0.0002, | |
| "reward": 2.4315611086785793, | |
| "reward_std": 2.3908147886395454, | |
| "rewards/concensus_correctness_reward_func": 0.625, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.75, | |
| "rewards/question_recreation_reward_func": 0.6140298657119274, | |
| "rewards/soft_format_reward_func": 0.015625, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.22378122806549072, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 437.8125, | |
| "epoch": 0.75, | |
| "grad_norm": 3.9912779331207275, | |
| "kl": 0.16812472557649016, | |
| "learning_rate": 4.193203929064353e-07, | |
| "loss": 0.0002, | |
| "reward": 1.896638199687004, | |
| "reward_std": 1.4613052681088448, | |
| "rewards/concensus_correctness_reward_func": 0.125, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.5625, | |
| "rewards/question_recreation_reward_func": 0.6531069651246071, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.4779062494635582, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 411.84375, | |
| "epoch": 1.0, | |
| "grad_norm": 2.287323474884033, | |
| "kl": 0.1949609499424696, | |
| "learning_rate": 3.5042385616324236e-07, | |
| "loss": 0.0002, | |
| "reward": 1.4827010408043861, | |
| "reward_std": 2.01706550270319, | |
| "rewards/concensus_correctness_reward_func": 0.16625000163912773, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.3125, | |
| "rewards/question_recreation_reward_func": 0.4630447644740343, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0625, | |
| "rewards/xmlcount_reward_func": 0.29090625420212746, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 463.125, | |
| "epoch": 1.25, | |
| "grad_norm": 2.7352001667022705, | |
| "kl": 0.18110767006874084, | |
| "learning_rate": 2.706448363680831e-07, | |
| "loss": 0.0002, | |
| "reward": 1.263205625116825, | |
| "reward_std": 1.7421698048710823, | |
| "rewards/concensus_correctness_reward_func": 0.12274999916553497, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.5625, | |
| "rewards/question_recreation_reward_func": 0.5095805916935205, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": -0.009749993681907654, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 351.71875, | |
| "epoch": 1.5, | |
| "grad_norm": 2.7147586345672607, | |
| "kl": 0.23088416643440723, | |
| "learning_rate": 1.886286282148002e-07, | |
| "loss": 0.0002, | |
| "reward": 1.9199633374810219, | |
| "reward_std": 1.0638368101790547, | |
| "rewards/concensus_correctness_reward_func": 0.1172500029206276, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.75, | |
| "rewards/question_recreation_reward_func": 0.42355713434517384, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.5666562579572201, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 437.03125, | |
| "epoch": 1.75, | |
| "grad_norm": 3.333895206451416, | |
| "kl": 0.17828004295006394, | |
| "learning_rate": 1.1326296046939333e-07, | |
| "loss": 0.0002, | |
| "reward": 1.6909483969211578, | |
| "reward_std": 1.299113318324089, | |
| "rewards/concensus_correctness_reward_func": 0.11999999731779099, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1875, | |
| "rewards/question_recreation_reward_func": 0.7196984179317951, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.078125, | |
| "rewards/xmlcount_reward_func": 0.46062498819082975, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 535.40625, | |
| "epoch": 2.0, | |
| "grad_norm": 2.2899842262268066, | |
| "kl": 0.16378580778837204, | |
| "learning_rate": 5.271487265090163e-08, | |
| "loss": 0.0002, | |
| "reward": 3.338853247463703, | |
| "reward_std": 4.333399765193462, | |
| "rewards/concensus_correctness_reward_func": 1.875, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.9375, | |
| "rewards/question_recreation_reward_func": 0.46935323998332024, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.04137500189244747, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 453.0625, | |
| "epoch": 2.25, | |
| "grad_norm": 3.0789365768432617, | |
| "kl": 0.19641543040052056, | |
| "learning_rate": 1.3545689574841341e-08, | |
| "loss": 0.0002, | |
| "reward": 2.0624579712748528, | |
| "reward_std": 1.581074796617031, | |
| "rewards/concensus_correctness_reward_func": 0.3570624999701977, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.6875, | |
| "rewards/question_recreation_reward_func": 0.5363016966730356, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.046875, | |
| "rewards/xmlcount_reward_func": 0.2472187504172325, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 435.53125, | |
| "epoch": 2.5, | |
| "grad_norm": 2.3260388374328613, | |
| "kl": 0.1985931508243084, | |
| "learning_rate": 0.0, | |
| "loss": 0.0002, | |
| "reward": 0.8565310798585415, | |
| "reward_std": 1.408127337694168, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.33818733133375645, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.11209375411272049, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 0.00019092126749455928, | |
| "train_runtime": 976.6916, | |
| "train_samples_per_second": 0.328, | |
| "train_steps_per_second": 0.02 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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 | |
| } | |