| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "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": 308.15625, | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 41.633419036865234, | |
| "kl": 0.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.40645824919920415, | |
| "reward_std": 0.31749973719843183, | |
| "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": 0.2209270077291876, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.12303124996833503, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 365.7083333333333, | |
| "epoch": 1.0, | |
| "grad_norm": 3.1837849617004395, | |
| "kl": 0.0010262345652639244, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.6313755170752605, | |
| "reward_std": 0.5310685941561436, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.16666666666666666, | |
| "rewards/question_recreation_reward_func": 0.2688755051543315, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.19583333159486452, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 365.15625, | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 7.127258777618408, | |
| "kl": 0.0020578340663632844, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.7601827555336058, | |
| "reward_std": 0.3503397152235266, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.34671399975195527, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.2259687555488199, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 314.9583333333333, | |
| "epoch": 2.0, | |
| "grad_norm": 2.7644894123077393, | |
| "kl": 0.04608077997302947, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.4097979703607659, | |
| "reward_std": 0.3929105601273477, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.25896464427933097, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.15083333291113377, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 344.625, | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 58.070220947265625, | |
| "kl": 0.3261278549889539, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0003, | |
| "reward": 0.424428222788265, | |
| "reward_std": 0.4320046803622972, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0, | |
| "rewards/question_recreation_reward_func": 0.205803226766875, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.21862499974668026, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 270.1666666666667, | |
| "epoch": 3.0, | |
| "grad_norm": 109.35530853271484, | |
| "kl": 0.5230223730419917, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0004, | |
| "reward": 1.480574295700838, | |
| "reward_std": 1.5046859136006485, | |
| "rewards/concensus_correctness_reward_func": 0.8659999991456667, | |
| "rewards/consensus_reward_func": 0.08333333333333333, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.08333333333333333, | |
| "rewards/question_recreation_reward_func": 0.2699492978863418, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.17795833324392638, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 303.90625, | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 104.08259582519531, | |
| "kl": 1.4068435160588706, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0014, | |
| "reward": 0.48999457189347595, | |
| "reward_std": 0.4255776608479209, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.19811958575155586, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.16687499592080712, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 325.375, | |
| "epoch": 4.0, | |
| "grad_norm": 8.133492469787598, | |
| "kl": 0.006218782771611586, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.5689306681354841, | |
| "reward_std": 0.5994748280694088, | |
| "rewards/concensus_correctness_reward_func": 0.010416666666666666, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.08333333333333333, | |
| "rewards/question_recreation_reward_func": 0.3621806618757546, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.11299999923600505, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 267.6875, | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 99.26972961425781, | |
| "kl": 0.10651466252238606, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0001, | |
| "reward": 0.4112930775154382, | |
| "reward_std": 0.2777931771270232, | |
| "rewards/concensus_correctness_reward_func": 0.0078125, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.16966809332370758, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.10881249979138374, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 293.25, | |
| "epoch": 5.0, | |
| "grad_norm": 22.004924774169922, | |
| "kl": 0.0053467367445894825, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 0.5409725056185076, | |
| "reward_std": 0.746750327021194, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.16666666666666666, | |
| "rewards/question_recreation_reward_func": 0.28509751621944207, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.0892083328217268, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 0.0002281336463511252, | |
| "train_runtime": 4153.4805, | |
| "train_samples_per_second": 0.077, | |
| "train_steps_per_second": 0.005 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |