{ "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": 0.009375, "completions/max_length": 1304.8, "completions/max_terminated_length": 1106.6, "completions/mean_length": 282.24375, "completions/mean_terminated_length": 270.57188110351564, "completions/min_length": 42.8, "completions/min_terminated_length": 42.8, "epoch": 0.07246376811594203, "grad_norm": 0.3630642890930176, "learning_rate": 5.714285714285714e-07, "loss": 0.0076, "num_tokens": 337130.0, "reward": 0.5359375, "reward_std": 0.1641431450843811, "rewards/accuracy_reward": 0.1375, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.934375, "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": 0.00625, "completions/max_length": 1316.6, "completions/max_terminated_length": 1149.2, "completions/mean_length": 276.371875, "completions/mean_terminated_length": 268.34995422363284, "completions/min_length": 37.0, "completions/min_terminated_length": 37.0, "epoch": 0.14492753623188406, "grad_norm": 0.40448635816574097, "learning_rate": 1e-06, "loss": 0.0108, "num_tokens": 676765.0, "reward": 0.515625, "reward_std": 0.10197109431028366, "rewards/accuracy_reward": 0.08125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.95, "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": 0.009375, "completions/max_length": 1222.8, "completions/max_terminated_length": 1032.6, "completions/mean_length": 261.6875, "completions/mean_terminated_length": 249.7670104980469, "completions/min_length": 35.4, "completions/min_terminated_length": 35.4, "epoch": 0.21739130434782608, "grad_norm": 0.39155879616737366, "learning_rate": 1e-06, "loss": 0.0102, "num_tokens": 1022165.0, "reward": 0.5609375, "reward_std": 0.1219540536403656, "rewards/accuracy_reward": 0.146875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.975, "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": 0.0125, "completions/max_length": 1364.2, "completions/max_terminated_length": 1197.6, "completions/mean_length": 244.296875, "completions/mean_terminated_length": 228.2646270751953, "completions/min_length": 34.8, "completions/min_terminated_length": 34.8, "epoch": 0.2898550724637681, "grad_norm": 0.5160397291183472, "learning_rate": 1e-06, "loss": 0.0215, "num_tokens": 1357132.0, "reward": 0.575, "reward_std": 0.14554107785224915, "rewards/accuracy_reward": 0.171875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.978125, "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": 0.0125, "completions/max_length": 1397.2, "completions/max_terminated_length": 1099.8, "completions/mean_length": 240.953125, "completions/mean_terminated_length": 224.53331298828124, "completions/min_length": 37.0, "completions/min_terminated_length": 37.0, "epoch": 0.36231884057971014, "grad_norm": 0.38764020800590515, "learning_rate": 1e-06, "loss": 0.0117, "num_tokens": 1686633.0, "reward": 0.5984375, "reward_std": 0.15249187350273133, "rewards/accuracy_reward": 0.21875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.978125, "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": 0.003125, "completions/max_length": 1133.0, "completions/max_terminated_length": 1013.0, "completions/mean_length": 201.953125, "completions/mean_terminated_length": 197.90258178710937, "completions/min_length": 29.8, "completions/min_terminated_length": 29.8, "epoch": 0.43478260869565216, "grad_norm": 0.34229469299316406, "learning_rate": 1e-06, "loss": 0.0153, "num_tokens": 1997466.0, "reward": 0.596875, "reward_std": 0.13894330859184265, "rewards/accuracy_reward": 0.203125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.990625, "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": 0.009375, "completions/max_length": 1245.6, "completions/max_terminated_length": 925.6, "completions/mean_length": 190.640625, "completions/mean_terminated_length": 177.88998718261718, "completions/min_length": 18.0, "completions/min_terminated_length": 18.0, "epoch": 0.5072463768115942, "grad_norm": 0.5255509614944458, "learning_rate": 1e-06, "loss": 0.0014, "num_tokens": 2316739.0, "reward": 0.6578125, "reward_std": 0.1760389268398285, "rewards/accuracy_reward": 0.325, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.990625, "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": 0.00625, "completions/max_length": 1343.6, "completions/max_terminated_length": 1012.8, "completions/mean_length": 159.303125, "completions/mean_terminated_length": 150.57073669433595, "completions/min_length": 30.4, "completions/min_terminated_length": 30.4, "epoch": 0.5797101449275363, "grad_norm": 0.6401497721672058, "learning_rate": 1e-06, "loss": 0.0171, "num_tokens": 2623608.0, "reward": 0.6453125, "reward_std": 0.17053844630718232, "rewards/accuracy_reward": 0.30625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.984375, "rewards/mean_confidence_reward": 0.0, "step": 40 }, { "epoch": 0.5797101449275363, "eval_completions/clipped_ratio": 0.0, "eval_completions/max_length": 397.68840579710144, "eval_completions/max_terminated_length": 397.68840579710144, "eval_completions/mean_length": 139.5679347826087, "eval_completions/mean_terminated_length": 139.5679347826087, "eval_completions/min_length": 47.369565217391305, "eval_completions/min_terminated_length": 47.369565217391305, "eval_loss": 0.0, "eval_num_tokens": 2623608.0, "eval_reward": 0.6476449275362319, "eval_reward_std": 0.17102870237136233, "eval_rewards/accuracy_reward": 0.29891304347826086, "eval_rewards/brier_reward": 0.0, "eval_rewards/confidence_one_or_zero": 0.0, "eval_rewards/format_reward": 0.9963768115942029, "eval_rewards/mean_confidence_reward": 0.0, "eval_runtime": 290.7488, "eval_samples_per_second": 0.949, "eval_steps_per_second": 0.12, "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": 957.6, "completions/max_terminated_length": 957.6, "completions/mean_length": 131.75625, "completions/mean_terminated_length": 131.75625, "completions/min_length": 36.8, "completions/min_terminated_length": 36.8, "epoch": 0.6521739130434783, "grad_norm": 0.6761258244514465, "learning_rate": 1e-06, "loss": 0.0078, "num_tokens": 2910862.0, "reward": 0.6734375, "reward_std": 0.19236910045146943, "rewards/accuracy_reward": 0.346875, "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": 918.4, "completions/max_terminated_length": 918.4, "completions/mean_length": 128.084375, "completions/mean_terminated_length": 128.084375, "completions/min_length": 28.0, "completions/min_terminated_length": 28.0, "epoch": 0.7246376811594203, "grad_norm": 0.4991821050643921, "learning_rate": 1e-06, "loss": 0.0053, "num_tokens": 3198045.0, "reward": 0.7015625, "reward_std": 0.18611910045146943, "rewards/accuracy_reward": 0.403125, "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": 594.4, "completions/max_terminated_length": 594.4, "completions/mean_length": 104.95625, "completions/mean_terminated_length": 104.95625, "completions/min_length": 27.0, "completions/min_terminated_length": 27.0, "epoch": 0.7971014492753623, "grad_norm": 0.9114353656768799, "learning_rate": 1e-06, "loss": -0.0078, "num_tokens": 3474299.0, "reward": 0.715625, "reward_std": 0.16398502439260482, "rewards/accuracy_reward": 0.434375, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.996875, "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": 0.003125, "completions/max_length": 854.8, "completions/max_terminated_length": 692.4, "completions/mean_length": 111.109375, "completions/mean_terminated_length": 106.64737243652344, "completions/min_length": 31.2, "completions/min_terminated_length": 31.2, "epoch": 0.8695652173913043, "grad_norm": 0.7115863561630249, "learning_rate": 1e-06, "loss": 0.0002, "num_tokens": 3759690.0, "reward": 0.6875, "reward_std": 0.13854593336582183, "rewards/accuracy_reward": 0.38125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.99375, "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": 208.0, "completions/max_terminated_length": 208.0, "completions/mean_length": 76.95625, "completions/mean_terminated_length": 76.95625, "completions/min_length": 34.2, "completions/min_terminated_length": 34.2, "epoch": 0.9420289855072463, "grad_norm": 0.5785619020462036, "learning_rate": 1e-06, "loss": 0.0011, "num_tokens": 4039904.0, "reward": 0.709375, "reward_std": 0.12983438670635222, "rewards/accuracy_reward": 0.41875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "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": 171.8, "completions/max_terminated_length": 171.8, "completions/mean_length": 75.003125, "completions/mean_terminated_length": 75.003125, "completions/min_length": 36.6, "completions/min_terminated_length": 36.6, "epoch": 1.0144927536231885, "grad_norm": 0.5040161609649658, "learning_rate": 1e-06, "loss": -0.001, "num_tokens": 4310042.0, "reward": 0.7390625, "reward_std": 0.115884068608284, "rewards/accuracy_reward": 0.48125, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.996875, "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": 236.6, "completions/max_terminated_length": 236.6, "completions/mean_length": 79.35625, "completions/mean_terminated_length": 79.35625, "completions/min_length": 33.8, "completions/min_terminated_length": 33.8, "epoch": 1.0869565217391304, "grad_norm": 0.46502524614334106, "learning_rate": 1e-06, "loss": 0.0006, "num_tokens": 4593268.0, "reward": 0.74375, "reward_std": 0.09110843911767005, "rewards/accuracy_reward": 0.4875, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 1.0, "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": 251.2, "completions/max_terminated_length": 251.2, "completions/mean_length": 83.08125, "completions/mean_terminated_length": 83.08125, "completions/min_length": 34.4, "completions/min_terminated_length": 34.4, "epoch": 1.1594202898550725, "grad_norm": 0.27010294795036316, "learning_rate": 1e-06, "loss": 0.0009, "num_tokens": 4866926.0, "reward": 0.7703125, "reward_std": 0.06034187749028206, "rewards/accuracy_reward": 0.540625, "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": 168.1014492753623, "eval_completions/max_terminated_length": 168.1014492753623, "eval_completions/mean_length": 88.21286231884058, "eval_completions/mean_terminated_length": 88.21286231884058, "eval_completions/min_length": 51.594202898550726, "eval_completions/min_terminated_length": 51.594202898550726, "eval_loss": 0.0, "eval_num_tokens": 4866926.0, "eval_reward": 0.7581521739130435, "eval_reward_std": 0.05643967485082322, "eval_rewards/accuracy_reward": 0.5163043478260869, "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": 134.5955, "eval_samples_per_second": 2.051, "eval_steps_per_second": 0.26, "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": 331.0, "completions/max_terminated_length": 331.0, "completions/mean_length": 86.365625, "completions/mean_terminated_length": 86.365625, "completions/min_length": 38.6, "completions/min_terminated_length": 38.6, "epoch": 1.2318840579710144, "grad_norm": 0.32811790704727173, "learning_rate": 1e-06, "loss": 0.0014, "num_tokens": 5147879.0, "reward": 0.765625, "reward_std": 0.066108438372612, "rewards/accuracy_reward": 0.53125, "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": 460.0, "completions/max_terminated_length": 460.0, "completions/mean_length": 91.74375, "completions/mean_terminated_length": 91.74375, "completions/min_length": 38.8, "completions/min_terminated_length": 38.8, "epoch": 1.3043478260869565, "grad_norm": 0.302518755197525, "learning_rate": 1e-06, "loss": 0.0009, "num_tokens": 5435417.0, "reward": 0.771875, "reward_std": 0.0375, "rewards/accuracy_reward": 0.54375, "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": 361.6, "completions/max_terminated_length": 361.6, "completions/mean_length": 89.4875, "completions/mean_terminated_length": 89.4875, "completions/min_length": 38.4, "completions/min_terminated_length": 38.4, "epoch": 1.3768115942028984, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0005, "num_tokens": 5710305.0, "reward": 0.75, "reward_std": 0.0125, "rewards/accuracy_reward": 0.5, "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": 0.003125, "completions/max_length": 499.8, "completions/max_terminated_length": 355.2, "completions/mean_length": 93.15625, "completions/mean_terminated_length": 88.68184661865234, "completions/min_length": 36.2, "completions/min_terminated_length": 36.2, "epoch": 1.4492753623188406, "grad_norm": 0.3082229197025299, "learning_rate": 1e-06, "loss": 0.0026, "num_tokens": 5996995.0, "reward": 0.73125, "reward_std": 0.01875, "rewards/accuracy_reward": 0.465625, "rewards/brier_reward": 0.0, "rewards/confidence_one_or_zero": 0.0, "rewards/format_reward": 0.996875, "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": 645.8, "completions/max_terminated_length": 645.8, "completions/mean_length": 97.765625, "completions/mean_terminated_length": 97.765625, "completions/min_length": 34.4, "completions/min_terminated_length": 34.4, "epoch": 1.5217391304347827, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": -0.0013, "num_tokens": 6281216.0, "reward": 0.796875, "reward_std": 0.0125, "rewards/accuracy_reward": 0.59375, "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": 648.8, "completions/max_terminated_length": 648.8, "completions/mean_length": 104.1125, "completions/mean_terminated_length": 104.1125, "completions/min_length": 42.0, "completions/min_terminated_length": 42.0, "epoch": 1.5942028985507246, "grad_norm": 0.2253188043832779, "learning_rate": 1e-06, "loss": 0.0002, "num_tokens": 6565188.0, "reward": 0.7609375, "reward_std": 0.003125, "rewards/accuracy_reward": 0.521875, "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": 536.0, "completions/max_terminated_length": 536.0, "completions/mean_length": 96.346875, "completions/mean_terminated_length": 96.346875, "completions/min_length": 41.4, "completions/min_terminated_length": 41.4, "epoch": 1.6666666666666665, "grad_norm": 0.19658243656158447, "learning_rate": 1e-06, "loss": -0.0002, "num_tokens": 6847259.0, "reward": 0.759375, "reward_std": 0.00625, "rewards/accuracy_reward": 0.51875, "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": 545.4, "completions/max_terminated_length": 545.4, "completions/mean_length": 100.046875, "completions/mean_terminated_length": 100.046875, "completions/min_length": 35.6, "completions/min_terminated_length": 35.6, "epoch": 1.7391304347826086, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0015, "num_tokens": 7136738.0, "reward": 0.759375, "reward_std": 0.00625, "rewards/accuracy_reward": 0.51875, "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": 207.07971014492753, "eval_completions/max_terminated_length": 207.07971014492753, "eval_completions/mean_length": 99.1875, "eval_completions/mean_terminated_length": 99.1875, "eval_completions/min_length": 53.3768115942029, "eval_completions/min_terminated_length": 53.3768115942029, "eval_loss": 0.0, "eval_num_tokens": 7136738.0, "eval_reward": 0.7549818840579711, "eval_reward_std": 0.004528985507246377, "eval_rewards/accuracy_reward": 0.5108695652173914, "eval_rewards/brier_reward": 0.0, "eval_rewards/confidence_one_or_zero": 0.0, "eval_rewards/format_reward": 0.9990942028985508, "eval_rewards/mean_confidence_reward": 0.0, "eval_runtime": 161.768, "eval_samples_per_second": 1.706, "eval_steps_per_second": 0.216, "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": 564.6, "completions/max_terminated_length": 564.6, "completions/mean_length": 95.471875, "completions/mean_terminated_length": 95.471875, "completions/min_length": 41.0, "completions/min_terminated_length": 41.0, "epoch": 1.8115942028985508, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": -0.0001, "num_tokens": 7413929.0, "reward": 0.75, "reward_std": 0.00625, "rewards/accuracy_reward": 0.5, "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": 510.2, "completions/max_terminated_length": 510.2, "completions/mean_length": 94.475, "completions/mean_terminated_length": 94.475, "completions/min_length": 35.6, "completions/min_terminated_length": 35.6, "epoch": 1.8840579710144927, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0, "num_tokens": 7698729.0, "reward": 0.78125, "reward_std": 0.0, "rewards/accuracy_reward": 0.5625, "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": 625.8, "completions/max_terminated_length": 625.8, "completions/mean_length": 100.25, "completions/mean_terminated_length": 100.25, "completions/min_length": 40.2, "completions/min_terminated_length": 40.2, "epoch": 1.9565217391304348, "grad_norm": 0.0, "learning_rate": 1e-06, "loss": 0.0003, "num_tokens": 7976297.0, "reward": 0.7578125, "reward_std": 0.009375, "rewards/accuracy_reward": 0.515625, "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": 441.6666666666667, "completions/max_terminated_length": 441.6666666666667, "completions/mean_length": 103.31770833333333, "completions/mean_terminated_length": 103.31770833333333, "completions/min_length": 44.666666666666664, "completions/min_terminated_length": 44.666666666666664, "epoch": 2.0, "num_tokens": 8138590.0, "reward": 0.7604166666666666, "reward_std": 0.0, "rewards/accuracy_reward": 0.5208333333333334, "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": 0.003933925466760453, "train_runtime": 2761.4112, "train_samples_per_second": 0.8, "train_steps_per_second": 0.05 } ], "logging_steps": 5, "max_steps": 138, "num_input_tokens_seen": 8138590, "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 }