| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "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": 189.46875, | |
| "epoch": 0.1, | |
| "grad_norm": 72.04657745361328, | |
| "kl": 0.0, | |
| "learning_rate": 4.965903258506806e-07, | |
| "loss": 0.0, | |
| "reward": 3.999774469062686, | |
| "reward_std": 2.164257466851268, | |
| "rewards/concensus_correctness_reward_func": 1.5554375015199184, | |
| "rewards/consensus_reward_func": 0.8125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.7209931435063481, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.6295937476679683, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 132.5625, | |
| "epoch": 0.2, | |
| "grad_norm": 10451.45703125, | |
| "kl": 253.42494644969702, | |
| "learning_rate": 4.698684378016222e-07, | |
| "loss": 0.2534, | |
| "reward": 5.015479948371649, | |
| "reward_std": 0.6629242373019224, | |
| "rewards/concensus_correctness_reward_func": 1.5227500051259995, | |
| "rewards/consensus_reward_func": 1.5, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.8574799399357289, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.03125, | |
| "rewards/xmlcount_reward_func": 0.8539999984204769, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 143.75, | |
| "epoch": 0.3, | |
| "grad_norm": 97.95394897460938, | |
| "kl": 96.7897967826575, | |
| "learning_rate": 4.193203929064353e-07, | |
| "loss": 0.0968, | |
| "reward": 5.306805005297065, | |
| "reward_std": 0.6057099380996078, | |
| "rewards/concensus_correctness_reward_func": 1.5708125047385693, | |
| "rewards/consensus_reward_func": 1.625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.8909612875431776, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.109375, | |
| "rewards/xmlcount_reward_func": 0.9856562428176403, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 163.25, | |
| "epoch": 0.4, | |
| "grad_norm": 118.51630401611328, | |
| "kl": 10.452200602274388, | |
| "learning_rate": 3.5042385616324236e-07, | |
| "loss": 0.0105, | |
| "reward": 5.741433784365654, | |
| "reward_std": 0.5551880679558963, | |
| "rewards/concensus_correctness_reward_func": 1.730749998241663, | |
| "rewards/consensus_reward_func": 1.8125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.8832775540649891, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.1875, | |
| "rewards/xmlcount_reward_func": 1.0024062488228083, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 133.96875, | |
| "epoch": 0.5, | |
| "grad_norm": 311.2564392089844, | |
| "kl": 6.50762258892064, | |
| "learning_rate": 2.706448363680831e-07, | |
| "loss": 0.0065, | |
| "reward": 5.870141640305519, | |
| "reward_std": 1.113852635025978, | |
| "rewards/concensus_correctness_reward_func": 1.815562479197979, | |
| "rewards/consensus_reward_func": 1.75, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.8371417056769133, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.1875, | |
| "rewards/xmlcount_reward_func": 1.0299374982714653, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 145.78125, | |
| "epoch": 0.6, | |
| "grad_norm": 77.84561157226562, | |
| "kl": 310228.7304797147, | |
| "learning_rate": 1.886286282148002e-07, | |
| "loss": 310.2288, | |
| "reward": 4.60096238553524, | |
| "reward_std": 0.7561608459218405, | |
| "rewards/concensus_correctness_reward_func": 1.2766249924898148, | |
| "rewards/consensus_reward_func": 1.375, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.8684623129665852, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.09375, | |
| "rewards/xmlcount_reward_func": 0.9246250009164214, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 136.3125, | |
| "epoch": 0.7, | |
| "grad_norm": 33276.56640625, | |
| "kl": 892.4592962227762, | |
| "learning_rate": 1.1326296046939333e-07, | |
| "loss": 0.8925, | |
| "reward": 5.898394346237183, | |
| "reward_std": 0.5051253397541586, | |
| "rewards/concensus_correctness_reward_func": 1.8743750005960464, | |
| "rewards/consensus_reward_func": 1.8125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.8720193216577172, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.109375, | |
| "rewards/xmlcount_reward_func": 0.9801249913871288, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 150.3125, | |
| "epoch": 0.8, | |
| "grad_norm": 237.27255249023438, | |
| "kl": 170.10233331704512, | |
| "learning_rate": 5.271487265090163e-08, | |
| "loss": 0.1701, | |
| "reward": 5.7240906953811646, | |
| "reward_std": 0.7569041545502841, | |
| "rewards/concensus_correctness_reward_func": 1.7523125000298023, | |
| "rewards/consensus_reward_func": 1.8125, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.9259657934308052, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.125, | |
| "rewards/xmlcount_reward_func": 0.9833124950528145, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 146.5625, | |
| "epoch": 0.9, | |
| "grad_norm": 62.39609909057617, | |
| "kl": 21.542335913982242, | |
| "learning_rate": 1.3545689574841341e-08, | |
| "loss": 0.0215, | |
| "reward": 5.123600989580154, | |
| "reward_std": 1.120502723613754, | |
| "rewards/concensus_correctness_reward_func": 1.5977499894797802, | |
| "rewards/consensus_reward_func": 1.6875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.8194135017693043, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.078125, | |
| "rewards/xmlcount_reward_func": 0.8158124927431345, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 149.75, | |
| "epoch": 1.0, | |
| "grad_norm": 64.91241455078125, | |
| "kl": 4.645937927765772, | |
| "learning_rate": 0.0, | |
| "loss": 0.0046, | |
| "reward": 5.778949812054634, | |
| "reward_std": 0.8444689979951363, | |
| "rewards/concensus_correctness_reward_func": 1.751749999821186, | |
| "rewards/consensus_reward_func": 1.875, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.8938873894512653, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.15625, | |
| "rewards/xmlcount_reward_func": 0.9770625047385693, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 31.168473226716742, | |
| "train_runtime": 124.1651, | |
| "train_samples_per_second": 2.577, | |
| "train_steps_per_second": 0.161 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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 | |
| } | |