| { | |
| "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": 448.34375, | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.9550420641899109, | |
| "kl": 0.0007673161817365326, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.6378161907196045, | |
| "reward_std": 0.7000563866458833, | |
| "rewards/concensus_correctness_reward_func": 0.0078125, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.3441286841407418, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.015625, | |
| "rewards/xmlcount_reward_func": 0.2077499981969595, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 559.7142857142857, | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.3691655397415161, | |
| "kl": 0.0006211901283157724, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 1.157321231705802, | |
| "reward_std": 1.319876023701259, | |
| "rewards/concensus_correctness_reward_func": 0.0353571426655565, | |
| "rewards/consensus_reward_func": 0.42857142857142855, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.5186783479792731, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.15685713929789408, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 410.32142857142856, | |
| "epoch": 2.0, | |
| "grad_norm": 0.3856706917285919, | |
| "kl": 0.0007643453011821423, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.7809734291263989, | |
| "reward_std": 0.7767886732305799, | |
| "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": 0.29372341611555647, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": 0.3801071345806122, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 398.9375, | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.4694315195083618, | |
| "kl": 0.0007544287436758168, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.7728503812104464, | |
| "reward_std": 0.7368343286216259, | |
| "rewards/concensus_correctness_reward_func": 0.027374999597668648, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4214128777384758, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.23031250946223736, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 412.32142857142856, | |
| "epoch": 3.3636363636363638, | |
| "grad_norm": 0.6740966439247131, | |
| "kl": 0.0007240102443444942, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.33227934156145367, | |
| "reward_std": 0.5858121908136776, | |
| "rewards/concensus_correctness_reward_func": 0.008928571428571428, | |
| "rewards/consensus_reward_func": 0.07142857142857142, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.2764579105590071, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.02453570972595896, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 458.7857142857143, | |
| "epoch": 4.0, | |
| "grad_norm": 0.5076102018356323, | |
| "kl": 0.0007106905063015542, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 1.0340332176004137, | |
| "reward_std": 0.9801140172140939, | |
| "rewards/concensus_correctness_reward_func": 0.008928571428571428, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.46624752453395296, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": 0.38028571967567715, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 449.09375, | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 1.6110213994979858, | |
| "kl": 0.0007361223615589552, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 1.0659161694347858, | |
| "reward_std": 0.9395686965435743, | |
| "rewards/concensus_correctness_reward_func": 0.0037499999161809683, | |
| "rewards/consensus_reward_func": 0.25, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.40613494254648685, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.4060312546789646, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 387.7857142857143, | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.3922109007835388, | |
| "kl": 0.0008256308252124914, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.8362700939178467, | |
| "reward_std": 0.9171016386577061, | |
| "rewards/concensus_correctness_reward_func": 0.008928571428571428, | |
| "rewards/consensus_reward_func": 0.14285714285714285, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.20062722345548017, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03571428571428571, | |
| "rewards/xmlcount_reward_func": 0.4481428542307445, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 411.75, | |
| "epoch": 6.0, | |
| "grad_norm": 0.3162197470664978, | |
| "kl": 0.00065167569222727, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.9783175247056144, | |
| "reward_std": 0.8466850008283343, | |
| "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": 0.5067460983991623, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.017857142857142856, | |
| "rewards/xmlcount_reward_func": 0.3108571491071156, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 476.1875, | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.44239941239356995, | |
| "kl": 0.0006094321288401261, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 0.7325360793620348, | |
| "reward_std": 1.0415890887379646, | |
| "rewards/concensus_correctness_reward_func": 0.0037499999161809683, | |
| "rewards/consensus_reward_func": 0.1875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.4430048484355211, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.06703124567866325, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 6.86128595361879e-07, | |
| "train_runtime": 1616.1217, | |
| "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 | |
| } | |