| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9893390191897654, | |
| "eval_steps": 100, | |
| "global_step": 58, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 613.9678833007813, | |
| "epoch": 0.08528784648187633, | |
| "grad_norm": 1.6863154172897339, | |
| "kl": 0.0001811981201171875, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0, | |
| "reward": 0.6475446730852127, | |
| "reward_std": 0.32297179140150545, | |
| "rewards/accuracy_reward": 0.646651816368103, | |
| "rewards/format_reward": 0.0008928571827709675, | |
| "step": 5 | |
| }, | |
| { | |
| "completion_length": 623.497127532959, | |
| "epoch": 0.17057569296375266, | |
| "grad_norm": 5.361064910888672, | |
| "kl": 4.185117244720459, | |
| "learning_rate": 2.956412726139078e-06, | |
| "loss": 0.1668, | |
| "reward": 0.695758955180645, | |
| "reward_std": 0.2824364464730024, | |
| "rewards/accuracy_reward": 0.6948660999536515, | |
| "rewards/format_reward": 0.0008928571827709675, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 621.0348518371582, | |
| "epoch": 0.255863539445629, | |
| "grad_norm": 0.34819459915161133, | |
| "kl": 0.006930732727050781, | |
| "learning_rate": 2.7836719084521715e-06, | |
| "loss": 0.0003, | |
| "reward": 0.7484375357627868, | |
| "reward_std": 0.23940655626356602, | |
| "rewards/accuracy_reward": 0.7484375357627868, | |
| "rewards/format_reward": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "completion_length": 617.9591751098633, | |
| "epoch": 0.3411513859275053, | |
| "grad_norm": 5.891634941101074, | |
| "kl": 0.007678604125976563, | |
| "learning_rate": 2.4946839873611927e-06, | |
| "loss": 0.0003, | |
| "reward": 0.7546875327825546, | |
| "reward_std": 0.23243394643068313, | |
| "rewards/accuracy_reward": 0.7546875327825546, | |
| "rewards/format_reward": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "completion_length": 625.6734642028808, | |
| "epoch": 0.42643923240938164, | |
| "grad_norm": 0.4985528588294983, | |
| "kl": 0.06998424530029297, | |
| "learning_rate": 2.1156192081791355e-06, | |
| "loss": 0.0028, | |
| "reward": 0.7497768148779869, | |
| "reward_std": 0.20579003393650055, | |
| "rewards/accuracy_reward": 0.7497768148779869, | |
| "rewards/format_reward": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "completion_length": 621.3915466308594, | |
| "epoch": 0.511727078891258, | |
| "grad_norm": 0.4441167712211609, | |
| "kl": 0.0034616470336914064, | |
| "learning_rate": 1.6808050203829845e-06, | |
| "loss": 0.0001, | |
| "reward": 0.7517857521772384, | |
| "reward_std": 0.19103028811514378, | |
| "rewards/accuracy_reward": 0.7517857521772384, | |
| "rewards/format_reward": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "completion_length": 611.6134216308594, | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 0.45489344000816345, | |
| "kl": 0.003511810302734375, | |
| "learning_rate": 1.2296174432791415e-06, | |
| "loss": 0.0001, | |
| "reward": 0.7323660954833031, | |
| "reward_std": 0.19216552414000035, | |
| "rewards/accuracy_reward": 0.7323660954833031, | |
| "rewards/format_reward": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "completion_length": 594.996004486084, | |
| "epoch": 0.6823027718550106, | |
| "grad_norm": 0.38299915194511414, | |
| "kl": 0.0036407470703125, | |
| "learning_rate": 8.029152419343472e-07, | |
| "loss": 0.0001, | |
| "reward": 0.7618303939700126, | |
| "reward_std": 0.182381122559309, | |
| "rewards/accuracy_reward": 0.7618303939700126, | |
| "rewards/format_reward": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "completion_length": 607.5631973266602, | |
| "epoch": 0.767590618336887, | |
| "grad_norm": 0.1558612883090973, | |
| "kl": 0.003850555419921875, | |
| "learning_rate": 4.3933982822017883e-07, | |
| "loss": 0.0002, | |
| "reward": 0.7508928924798965, | |
| "reward_std": 0.19622449725866317, | |
| "rewards/accuracy_reward": 0.7508928924798965, | |
| "rewards/format_reward": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "completion_length": 610.1696670532226, | |
| "epoch": 0.8528784648187633, | |
| "grad_norm": 0.11370094120502472, | |
| "kl": 0.004032135009765625, | |
| "learning_rate": 1.718159615201853e-07, | |
| "loss": 0.0002, | |
| "reward": 0.7486607491970062, | |
| "reward_std": 0.17984699215739966, | |
| "rewards/accuracy_reward": 0.7486607491970062, | |
| "rewards/format_reward": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "completion_length": 603.7953384399414, | |
| "epoch": 0.9381663113006397, | |
| "grad_norm": 0.09367287904024124, | |
| "kl": 0.0048553466796875, | |
| "learning_rate": 2.4570139579284723e-08, | |
| "loss": 0.0002, | |
| "reward": 0.7750000357627869, | |
| "reward_std": 0.18225797163322568, | |
| "rewards/accuracy_reward": 0.7750000357627869, | |
| "rewards/format_reward": 0.0, | |
| "step": 55 | |
| }, | |
| { | |
| "completion_length": 598.5974960327148, | |
| "epoch": 0.9893390191897654, | |
| "kl": 0.003758112589518229, | |
| "reward": 0.7678571765621504, | |
| "reward_std": 0.18772160820662975, | |
| "rewards/accuracy_reward": 0.7678571765621504, | |
| "rewards/format_reward": 0.0, | |
| "step": 58, | |
| "total_flos": 0.0, | |
| "train_loss": 0.01571170037346621, | |
| "train_runtime": 7483.8014, | |
| "train_samples_per_second": 1.002, | |
| "train_steps_per_second": 0.008 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 58, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |