| { | |
| "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": 395.25, | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.7927322387695312, | |
| "kl": 0.0012580822076415643, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.5056484336964786, | |
| "reward_std": 0.8683814308606088, | |
| "rewards/concensus_correctness_reward_func": 0.06012500077486038, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.26574217714369297, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.11728124879300594, | |
| "step": 2 | |
| }, | |
| { | |
| "completion_length": 301.15, | |
| "epoch": 1.0, | |
| "grad_norm": 28.787370681762695, | |
| "kl": 0.007838598499074579, | |
| "learning_rate": 4.864543104251586e-07, | |
| "loss": 0.0, | |
| "reward": 0.4848457515239716, | |
| "reward_std": 0.6104163944721221, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.1, | |
| "rewards/question_recreation_reward_func": 0.211995729804039, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.17284999787807465, | |
| "step": 4 | |
| }, | |
| { | |
| "completion_length": 366.0625, | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.4565224647521973, | |
| "kl": 0.0048014514686656184, | |
| "learning_rate": 4.472851273490984e-07, | |
| "loss": 0.0, | |
| "reward": 0.5369042251259089, | |
| "reward_std": 0.7701209064107388, | |
| "rewards/concensus_correctness_reward_func": 0.1015625, | |
| "rewards/consensus_reward_func": 0.0625, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.1854354883544147, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.12490624794736505, | |
| "step": 6 | |
| }, | |
| { | |
| "completion_length": 488.75, | |
| "epoch": 2.0, | |
| "grad_norm": 1.1058247089385986, | |
| "kl": 0.0015862735570408403, | |
| "learning_rate": 3.867370395306068e-07, | |
| "loss": 0.0, | |
| "reward": 0.7044907063245773, | |
| "reward_std": 1.334967276453972, | |
| "rewards/concensus_correctness_reward_func": 0.09620000123977661, | |
| "rewards/consensus_reward_func": 0.1, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.6, | |
| "rewards/question_recreation_reward_func": 0.2195407159626484, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": -0.3112500309944153, | |
| "step": 8 | |
| }, | |
| { | |
| "completion_length": 464.9375, | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 2.566697359085083, | |
| "kl": 0.0013416979272733442, | |
| "learning_rate": 3.1137137178519977e-07, | |
| "loss": 0.0, | |
| "reward": 0.562221004627645, | |
| "reward_std": 0.8246445022523403, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.25, | |
| "rewards/question_recreation_reward_func": 0.27678349521011114, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.03543749637901783, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 306.25, | |
| "epoch": 3.0, | |
| "grad_norm": 0.7933401465415955, | |
| "kl": 0.0031318686669692397, | |
| "learning_rate": 2.2935516363191693e-07, | |
| "loss": 0.0, | |
| "reward": 1.342978870868683, | |
| "reward_std": 1.444445651769638, | |
| "rewards/concensus_correctness_reward_func": 0.18480000495910645, | |
| "rewards/consensus_reward_func": 0.2, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.2, | |
| "rewards/question_recreation_reward_func": 0.37442886233329775, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.3837499976158142, | |
| "step": 12 | |
| }, | |
| { | |
| "completion_length": 367.0625, | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.901016354560852, | |
| "kl": 0.0015947269203024916, | |
| "learning_rate": 1.4957614383675767e-07, | |
| "loss": 0.0, | |
| "reward": 0.48512101359665394, | |
| "reward_std": 0.8020865241996944, | |
| "rewards/concensus_correctness_reward_func": 0.0234375, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.125, | |
| "rewards/question_recreation_reward_func": 0.2969335000962019, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.03974999859929085, | |
| "step": 14 | |
| }, | |
| { | |
| "completion_length": 421.6, | |
| "epoch": 4.0, | |
| "grad_norm": 1.3564997911453247, | |
| "kl": 0.004650218156166374, | |
| "learning_rate": 8.067960709356478e-08, | |
| "loss": 0.0, | |
| "reward": 0.2924229323863983, | |
| "reward_std": 0.7882522225379944, | |
| "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": 0.27797292321920397, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.01444999873638153, | |
| "step": 16 | |
| }, | |
| { | |
| "completion_length": 260.75, | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.8875579833984375, | |
| "kl": 0.003812392649706453, | |
| "learning_rate": 3.013156219837776e-08, | |
| "loss": 0.0, | |
| "reward": 0.4502076015342027, | |
| "reward_std": 0.5057817947817966, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.0625, | |
| "rewards/question_recreation_reward_func": 0.3271451264154166, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.060562501195818186, | |
| "step": 18 | |
| }, | |
| { | |
| "completion_length": 399.15, | |
| "epoch": 5.0, | |
| "grad_norm": 1.042950987815857, | |
| "kl": 0.001434546068776399, | |
| "learning_rate": 3.4096741493194193e-09, | |
| "loss": 0.0, | |
| "reward": 1.0424630206078291, | |
| "reward_std": 0.9998365689069033, | |
| "rewards/concensus_correctness_reward_func": 0.0, | |
| "rewards/consensus_reward_func": 0.0, | |
| "rewards/cumulative_reward_2": 0.0, | |
| "rewards/final_correctness_reward_func": 0.5, | |
| "rewards/question_recreation_reward_func": 0.273763046041131, | |
| "rewards/soft_format_reward_func": 0.0, | |
| "rewards/strict_format_reward_func": 0.0, | |
| "rewards/xmlcount_reward_func": 0.2687000036239624, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 20, | |
| "total_flos": 0.0, | |
| "train_loss": 2.451066382036515e-06, | |
| "train_runtime": 531.9792, | |
| "train_samples_per_second": 0.602, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "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": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |