{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 2.0, "eval_steps": 40, "global_step": 138, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.01875, "completions/max_length": 1442.2, "completions/max_terminated_length": 1161.6, "completions/mean_length": 340.6, "completions/mean_terminated_length": 317.76939697265624, "completions/min_length": 35.4, "completions/min_terminated_length": 35.4, "epoch": 0.07246376811594203, "grad_norm": 0.4011998772621155, "learning_rate": 5.714285714285714e-07, "loss": 0.0032, "num_tokens": 355484.0, "reward": 0.49375, "reward_std": 0.15815922617912292, "rewards/accuracy_reward": 0.090625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.896875, "rewards/mean_confidence_reward": 0.0, "step": 5 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.028125, "completions/max_length": 1529.4, "completions/max_terminated_length": 1263.6, "completions/mean_length": 346.821875, "completions/mean_terminated_length": 312.370703125, "completions/min_length": 43.2, "completions/min_terminated_length": 43.2, "epoch": 0.14492753623188406, "grad_norm": 0.22016562521457672, "learning_rate": 1e-06, "loss": 0.0279, "num_tokens": 717343.0, "reward": 0.525, "reward_std": 0.15530819296836854, "rewards/accuracy_reward": 0.115625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.934375, "rewards/mean_confidence_reward": 0.0, "step": 10 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.03125, "completions/max_length": 1536.0, "completions/max_terminated_length": 1279.0, "completions/mean_length": 336.515625, "completions/mean_terminated_length": 297.9698791503906, "completions/min_length": 39.8, "completions/min_terminated_length": 39.8, "epoch": 0.21739130434782608, "grad_norm": 64.25254821777344, "learning_rate": 1e-06, "loss": 0.0336, "num_tokens": 1086368.0, "reward": 0.509375, "reward_std": 0.14508534520864486, "rewards/accuracy_reward": 0.1, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.91875, "rewards/mean_confidence_reward": 0.0, "step": 15 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.025, "completions/max_length": 1533.2, "completions/max_terminated_length": 1162.6, "completions/mean_length": 318.65625, "completions/mean_terminated_length": 287.4467468261719, "completions/min_length": 34.4, "completions/min_terminated_length": 34.4, "epoch": 0.2898550724637681, "grad_norm": 0.26145997643470764, "learning_rate": 1e-06, "loss": 0.0145, "num_tokens": 1444810.0, "reward": 0.521875, "reward_std": 0.13414625823497772, "rewards/accuracy_reward": 0.096875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.946875, "rewards/mean_confidence_reward": 0.0, "step": 20 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.021875, "completions/max_length": 1536.0, "completions/max_terminated_length": 1115.2, "completions/mean_length": 279.265625, "completions/mean_terminated_length": 251.21444702148438, "completions/min_length": 25.6, "completions/min_terminated_length": 25.6, "epoch": 0.36231884057971014, "grad_norm": 0.28721150755882263, "learning_rate": 1e-06, "loss": 0.0253, "num_tokens": 1786251.0, "reward": 0.55, "reward_std": 0.16766551434993743, "rewards/accuracy_reward": 0.153125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.946875, "rewards/mean_confidence_reward": 0.0, "step": 25 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.009375, "completions/max_length": 1243.2, "completions/max_terminated_length": 994.2, "completions/mean_length": 172.60625, "completions/mean_terminated_length": 159.64936065673828, "completions/min_length": 13.6, "completions/min_terminated_length": 13.6, "epoch": 0.43478260869565216, "grad_norm": 0.7999480366706848, "learning_rate": 1e-06, "loss": 0.0246, "num_tokens": 2087373.0, "reward": 0.584375, "reward_std": 0.1618712455034256, "rewards/accuracy_reward": 0.1875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.98125, "rewards/mean_confidence_reward": 0.0, "step": 30 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.003125, "completions/max_length": 828.0, "completions/max_terminated_length": 708.8, "completions/mean_length": 70.784375, "completions/mean_terminated_length": 66.40218505859374, "completions/min_length": 13.4, "completions/min_terminated_length": 13.4, "epoch": 0.5072463768115942, "grad_norm": 1.435221791267395, "learning_rate": 1e-06, "loss": 0.0172, "num_tokens": 2367972.0, "reward": 0.63125, "reward_std": 0.20825054347515107, "rewards/accuracy_reward": 0.28125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.98125, "rewards/mean_confidence_reward": 0.0, "step": 35 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 164.4, "completions/max_terminated_length": 164.4, "completions/mean_length": 20.071875, "completions/mean_terminated_length": 20.071875, "completions/min_length": 13.2, "completions/min_terminated_length": 13.2, "epoch": 0.5797101449275363, "grad_norm": 1.5361332893371582, "learning_rate": 1e-06, "loss": 0.0104, "num_tokens": 2629967.0, "reward": 0.7, "reward_std": 0.1850165456533432, "rewards/accuracy_reward": 0.409375, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.990625, "rewards/mean_confidence_reward": 0.0, "step": 40 }, { "epoch": 0.5797101449275363, "eval_completions/clipped_ratio": 0.0, "eval_completions/max_length": 20.17391304347826, "eval_completions/max_terminated_length": 20.17391304347826, "eval_completions/mean_length": 14.672101449275363, "eval_completions/mean_terminated_length": 14.672101449275363, "eval_completions/min_length": 13.398550724637682, "eval_completions/min_terminated_length": 13.398550724637682, "eval_loss": 0.0, "eval_num_tokens": 2629967.0, "eval_reward": 0.730072463768116, "eval_reward_std": 0.15374549445898636, "eval_rewards/accuracy_reward": 0.4601449275362319, "eval_rewards/brier_reward": 0.0, "eval_rewards/confidence_one_or_zero": 0.0, "eval_rewards/format_reward": 1.0, "eval_rewards/mean_confidence_reward": 0.0, "eval_runtime": 31.9624, "eval_samples_per_second": 8.635, "eval_steps_per_second": 1.095, "step": 40 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 29.4, "completions/max_terminated_length": 29.4, "completions/mean_length": 13.875, "completions/mean_terminated_length": 13.875, "completions/min_length": 13.0, "completions/min_terminated_length": 13.0, "epoch": 0.6521739130434783, "grad_norm": 0.40115615725517273, "learning_rate": 1e-06, "loss": -0.0004, "num_tokens": 2879179.0, "reward": 0.7765625, "reward_std": 0.1043509490787983, "rewards/accuracy_reward": 0.553125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 45 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 13.909375, "completions/mean_terminated_length": 13.909375, "completions/min_length": 13.2, "completions/min_terminated_length": 13.2, "epoch": 0.7246376811594203, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": -0.0001, "num_tokens": 3129506.0, "reward": 0.8296875, "reward_std": 0.01923343911767006, "rewards/accuracy_reward": 0.659375, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 50 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 16.6, "completions/max_terminated_length": 16.6, "completions/mean_length": 14.0375, "completions/mean_terminated_length": 14.0375, "completions/min_length": 13.8, "completions/min_terminated_length": 13.8, "epoch": 0.7971014492753623, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": -0.0, "num_tokens": 3376346.0, "reward": 0.828125, "reward_std": 0.00625, "rewards/accuracy_reward": 0.65625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 55 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 44.8, "completions/max_terminated_length": 44.8, "completions/mean_length": 14.48125, "completions/mean_terminated_length": 14.48125, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 0.8695652173913043, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 3630496.0, "reward": 0.8, "reward_std": 0.0, "rewards/accuracy_reward": 0.6, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 60 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 33.8, "completions/max_terminated_length": 33.8, "completions/mean_length": 14.309375, "completions/mean_terminated_length": 14.309375, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 0.9420289855072463, "grad_norm": 1.4851375818252563, "learning_rate": 1e-06, "loss": 0.0045, "num_tokens": 3890343.0, "reward": 0.784375, "reward_std": 0.00625, "rewards/accuracy_reward": 0.571875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.996875, "rewards/mean_confidence_reward": 0.0, "step": 65 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 16.6, "completions/max_terminated_length": 16.6, "completions/mean_length": 14.040625, "completions/mean_terminated_length": 14.040625, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.0144927536231885, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 4140916.0, "reward": 0.8, "reward_std": 0.0, "rewards/accuracy_reward": 0.6, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 70 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 17.4, "completions/max_terminated_length": 17.4, "completions/mean_length": 14.053125, "completions/mean_terminated_length": 14.053125, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.0869565217391304, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0008, "num_tokens": 4402925.0, "reward": 0.7921875, "reward_std": 0.003125, "rewards/accuracy_reward": 0.5875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.996875, "rewards/mean_confidence_reward": 0.0, "step": 75 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.1594202898550725, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 4654157.0, "reward": 0.83125, "reward_std": 0.0, "rewards/accuracy_reward": 0.6625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 80 }, { "epoch": 1.1594202898550725, "eval_completions/clipped_ratio": 0.0, "eval_completions/max_length": 15.007246376811594, "eval_completions/max_terminated_length": 15.007246376811594, "eval_completions/mean_length": 14.12590579710145, "eval_completions/mean_terminated_length": 14.12590579710145, "eval_completions/min_length": 14.0, "eval_completions/min_terminated_length": 14.0, "eval_loss": 0.0, "eval_num_tokens": 4654157.0, "eval_reward": 0.7862318840579711, "eval_reward_std": 0.0, "eval_rewards/accuracy_reward": 0.572463768115942, "eval_rewards/brier_reward": 0.0, "eval_rewards/confidence_one_or_zero": 0.0, "eval_rewards/format_reward": 1.0, "eval_rewards/mean_confidence_reward": 0.0, "eval_runtime": 28.4457, "eval_samples_per_second": 9.703, "eval_steps_per_second": 1.23, "step": 80 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.2318840579710144, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 4911633.0, "reward": 0.80625, "reward_std": 0.0, "rewards/accuracy_reward": 0.6125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 85 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.3043478260869565, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 5173973.0, "reward": 0.825, "reward_std": 0.0, "rewards/accuracy_reward": 0.65, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 90 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.3768115942028984, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 5424385.0, "reward": 0.80625, "reward_std": 0.0, "rewards/accuracy_reward": 0.6125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 95 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 67.4, "completions/max_terminated_length": 67.4, "completions/mean_length": 14.834375, "completions/mean_terminated_length": 14.834375, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.4492753623188406, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 5685692.0, "reward": 0.8, "reward_std": 0.0, "rewards/accuracy_reward": 0.6, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 100 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.5217391304347827, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 5942788.0, "reward": 0.7875, "reward_std": 0.0, "rewards/accuracy_reward": 0.575, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 105 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.5942028985507246, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 6197604.0, "reward": 0.81875, "reward_std": 0.0, "rewards/accuracy_reward": 0.6375, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 110 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.6666666666666665, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 6453004.0, "reward": 0.7625, "reward_std": 0.0, "rewards/accuracy_reward": 0.525, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 115 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.7391304347826086, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 6714628.0, "reward": 0.7375, "reward_std": 0.0, "rewards/accuracy_reward": 0.475, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 120 }, { "epoch": 1.7391304347826086, "eval_completions/clipped_ratio": 0.0, "eval_completions/max_length": 14.22463768115942, "eval_completions/max_terminated_length": 14.22463768115942, "eval_completions/mean_length": 14.028079710144928, "eval_completions/mean_terminated_length": 14.028079710144928, "eval_completions/min_length": 14.0, "eval_completions/min_terminated_length": 14.0, "eval_loss": 0.0, "eval_num_tokens": 6714628.0, "eval_reward": 0.7862318840579711, "eval_reward_std": 0.0, "eval_rewards/accuracy_reward": 0.572463768115942, "eval_rewards/brier_reward": 0.0, "eval_rewards/confidence_one_or_zero": 0.0, "eval_rewards/format_reward": 1.0, "eval_rewards/mean_confidence_reward": 0.0, "eval_runtime": 27.9073, "eval_samples_per_second": 9.89, "eval_steps_per_second": 1.254, "step": 120 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.8115942028985508, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 6965428.0, "reward": 0.79375, "reward_std": 0.0, "rewards/accuracy_reward": 0.5875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 125 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.8840579710144927, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 7224156.0, "reward": 0.8, "reward_std": 0.0, "rewards/accuracy_reward": 0.6, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 130 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 1.9565217391304348, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 7473804.0, "reward": 0.79375, "reward_std": 0.0, "rewards/accuracy_reward": 0.5875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 135 }, { "clip_ratio/high_max": 0.0, "clip_ratio/high_mean": 0.0, "clip_ratio/low_mean": 0.0, "clip_ratio/low_min": 0.0, "clip_ratio/region_mean": 0.0, "completions/clipped_ratio": 0.0, "completions/max_length": 14.0, "completions/max_terminated_length": 14.0, "completions/mean_length": 14.0, "completions/mean_terminated_length": 14.0, "completions/min_length": 14.0, "completions/min_terminated_length": 14.0, "epoch": 2.0, "num_tokens": 7619180.0, "reward": 0.8854166666666666, "reward_std": 0.0, "rewards/accuracy_reward": 0.7708333333333334, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "rewards/mean_confidence_reward": 0.0, "step": 138, "total_flos": 0.0, "train_loss": 0.005852512148060445, "train_runtime": 1821.197, "train_samples_per_second": 1.212, "train_steps_per_second": 0.076 } ], "logging_steps": 5, "max_steps": 138, "num_input_tokens_seen": 7619180, "num_train_epochs": 2, "save_steps": 40, "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 }