| { | |
| "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": 593.4228515625, | |
| "epoch": 0.017057569296375266, | |
| "grad_norm": 5.732693195343018, | |
| "kl": 0.0, | |
| "learning_rate": 5e-07, | |
| "loss": 0.0, | |
| "reward": 0.619140625, | |
| "reward_std": 0.34863705188035965, | |
| "rewards/accuracy_reward": 0.619140625, | |
| "rewards/thinking_format_reward": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "completion_length": 587.63427734375, | |
| "epoch": 0.08528784648187633, | |
| "grad_norm": 6.981931686401367, | |
| "kl": 0.0008228123188018799, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0, | |
| "reward": 0.595947265625, | |
| "reward_std": 0.36766810342669487, | |
| "rewards/accuracy_reward": 0.59228515625, | |
| "rewards/thinking_format_reward": 0.003662109375, | |
| "step": 5 | |
| }, | |
| { | |
| "completion_length": 614.907421875, | |
| "epoch": 0.17057569296375266, | |
| "grad_norm": 3.0340416431427, | |
| "kl": 0.29906692504882815, | |
| "learning_rate": 2.956412726139078e-06, | |
| "loss": 0.012, | |
| "reward": 0.6724609375, | |
| "reward_std": 0.3203321475535631, | |
| "rewards/accuracy_reward": 0.6701171875, | |
| "rewards/thinking_format_reward": 0.00234375, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 624.4158203125, | |
| "epoch": 0.255863539445629, | |
| "grad_norm": 1.2238105535507202, | |
| "kl": 0.0524169921875, | |
| "learning_rate": 2.7836719084521715e-06, | |
| "loss": 0.0021, | |
| "reward": 0.7359375, | |
| "reward_std": 0.26645895689725874, | |
| "rewards/accuracy_reward": 0.728515625, | |
| "rewards/thinking_format_reward": 0.007421875, | |
| "step": 15 | |
| }, | |
| { | |
| "completion_length": 637.015234375, | |
| "epoch": 0.3411513859275053, | |
| "grad_norm": 0.17573410272598267, | |
| "kl": 0.01880950927734375, | |
| "learning_rate": 2.4946839873611927e-06, | |
| "loss": 0.0008, | |
| "reward": 0.73359375, | |
| "reward_std": 0.2729787778109312, | |
| "rewards/accuracy_reward": 0.724609375, | |
| "rewards/thinking_format_reward": 0.008984375, | |
| "step": 20 | |
| }, | |
| { | |
| "completion_length": 649.4162109375, | |
| "epoch": 0.42643923240938164, | |
| "grad_norm": 0.13386043906211853, | |
| "kl": 0.02559814453125, | |
| "learning_rate": 2.1156192081791355e-06, | |
| "loss": 0.001, | |
| "reward": 0.713671875, | |
| "reward_std": 0.2583099763840437, | |
| "rewards/accuracy_reward": 0.708984375, | |
| "rewards/thinking_format_reward": 0.0046875, | |
| "step": 25 | |
| }, | |
| { | |
| "completion_length": 645.4900390625, | |
| "epoch": 0.511727078891258, | |
| "grad_norm": 0.09986437112092972, | |
| "kl": 0.0204132080078125, | |
| "learning_rate": 1.6808050203829845e-06, | |
| "loss": 0.0008, | |
| "reward": 0.717578125, | |
| "reward_std": 0.23345882780849933, | |
| "rewards/accuracy_reward": 0.708203125, | |
| "rewards/thinking_format_reward": 0.009375, | |
| "step": 30 | |
| }, | |
| { | |
| "completion_length": 633.1984375, | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 792.4126586914062, | |
| "kl": 0.7226852416992188, | |
| "learning_rate": 1.2296174432791415e-06, | |
| "loss": 0.029, | |
| "reward": 0.7240234375, | |
| "reward_std": 0.21177927535027266, | |
| "rewards/accuracy_reward": 0.7099609375, | |
| "rewards/thinking_format_reward": 0.0140625, | |
| "step": 35 | |
| }, | |
| { | |
| "completion_length": 620.9001953125, | |
| "epoch": 0.6823027718550106, | |
| "grad_norm": 0.0885159894824028, | |
| "kl": 0.0154876708984375, | |
| "learning_rate": 8.029152419343472e-07, | |
| "loss": 0.0006, | |
| "reward": 0.746875, | |
| "reward_std": 0.21467349883168935, | |
| "rewards/accuracy_reward": 0.734375, | |
| "rewards/thinking_format_reward": 0.0125, | |
| "step": 40 | |
| }, | |
| { | |
| "completion_length": 638.7796875, | |
| "epoch": 0.767590618336887, | |
| "grad_norm": 0.1167062520980835, | |
| "kl": 0.01483917236328125, | |
| "learning_rate": 4.3933982822017883e-07, | |
| "loss": 0.0006, | |
| "reward": 0.723046875, | |
| "reward_std": 0.22248270493000746, | |
| "rewards/accuracy_reward": 0.718359375, | |
| "rewards/thinking_format_reward": 0.0046875, | |
| "step": 45 | |
| }, | |
| { | |
| "completion_length": 636.246484375, | |
| "epoch": 0.8528784648187633, | |
| "grad_norm": 0.12419791519641876, | |
| "kl": 0.01413421630859375, | |
| "learning_rate": 1.718159615201853e-07, | |
| "loss": 0.0006, | |
| "reward": 0.7263671875, | |
| "reward_std": 0.22438973132520915, | |
| "rewards/accuracy_reward": 0.7154296875, | |
| "rewards/thinking_format_reward": 0.0109375, | |
| "step": 50 | |
| }, | |
| { | |
| "completion_length": 633.81875, | |
| "epoch": 0.9381663113006397, | |
| "grad_norm": 0.08390026539564133, | |
| "kl": 0.01665496826171875, | |
| "learning_rate": 2.4570139579284723e-08, | |
| "loss": 0.0007, | |
| "reward": 0.7556640625, | |
| "reward_std": 0.22444888018071651, | |
| "rewards/accuracy_reward": 0.7431640625, | |
| "rewards/thinking_format_reward": 0.0125, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.9893390191897654, | |
| "step": 58, | |
| "total_flos": 0.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 1.7087, | |
| "train_samples_per_second": 4389.274, | |
| "train_steps_per_second": 33.944 | |
| } | |
| ], | |
| "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": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |