{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.146484375, "eval_steps": 500, "global_step": 300, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00048828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.5023794770240784, "kl": 3.4439620971679688, "learning_rate": 0.0, "loss": 0.1378, "num_tokens": 2493.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 1 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0009765625, "frac_reward_zero_std": 0.0, "grad_norm": 17.32089614868164, "kl": 3.229800969362259, "learning_rate": 1.6666666666666668e-07, "loss": 0.1292, "num_tokens": 4878.0, "reward": 5.125, "reward_std": 1.4361406564712524, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 1.4361406564712524, "step": 2 }, { "completion_length": 3.5, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.5, "completions/mean_terminated_length": 3.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00146484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5975773930549622, "kl": 3.099585711956024, "learning_rate": 3.3333333333333335e-07, "loss": 0.124, "num_tokens": 7184.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 3 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.001953125, "frac_reward_zero_std": 0.0, "grad_norm": 13.477265357971191, "kl": 6.181465029716492, "learning_rate": 5.000000000000001e-07, "loss": 0.2473, "num_tokens": 9736.0, "reward": 5.125, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 0.75, "step": 4 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00244140625, "frac_reward_zero_std": 0.0, "grad_norm": 13.396224021911621, "kl": 3.2642417550086975, "learning_rate": 6.666666666666667e-07, "loss": 0.1306, "num_tokens": 12641.0, "reward": 5.125, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 0.75, "step": 5 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0029296875, "frac_reward_zero_std": 0.0, "grad_norm": 13.23792839050293, "kl": 2.4123340845108032, "learning_rate": 8.333333333333333e-07, "loss": 0.0965, "num_tokens": 15089.0, "reward": 6.25, "reward_std": 1.5, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 1.5, "step": 6 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.00341796875, "frac_reward_zero_std": 1.0, "grad_norm": 13.976696014404297, "kl": 4.631758064031601, "learning_rate": 1.0000000000000002e-06, "loss": 0.1853, "num_tokens": 17573.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 7 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.8367634415626526, "kl": 0.18378764390945435, "learning_rate": 1.1666666666666668e-06, "loss": 0.0074, "num_tokens": 19832.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 8 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00439453125, "frac_reward_zero_std": 0.0, "grad_norm": 21.79625701904297, "kl": 3.363227665424347, "learning_rate": 1.3333333333333334e-06, "loss": 0.1345, "num_tokens": 22019.0, "reward": 4.75, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 4.75, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 9 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0048828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.4207085072994232, "kl": 3.3578058183193207, "learning_rate": 1.5e-06, "loss": 0.1343, "num_tokens": 24459.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 10 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00537109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5501925945281982, "kl": 3.9019325375556946, "learning_rate": 1.6666666666666667e-06, "loss": 0.1561, "num_tokens": 26899.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 11 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.005859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.6654255390167236, "kl": 1.3104343861341476, "learning_rate": 1.8333333333333333e-06, "loss": 0.0524, "num_tokens": 29449.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 12 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00634765625, "frac_reward_zero_std": 1.0, "grad_norm": 1.428830623626709, "kl": 3.8021611273288727, "learning_rate": 2.0000000000000003e-06, "loss": 0.1521, "num_tokens": 31952.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 13 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0068359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.4458273947238922, "kl": 2.901183843612671, "learning_rate": 2.166666666666667e-06, "loss": 0.116, "num_tokens": 34297.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 14 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.00732421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.2321917712688446, "kl": 0.6252752877771854, "learning_rate": 2.3333333333333336e-06, "loss": 0.025, "num_tokens": 36552.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 15 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0078125, "frac_reward_zero_std": 0.0, "grad_norm": 13.081548690795898, "kl": 6.5625739097595215, "learning_rate": 2.5e-06, "loss": 0.2625, "num_tokens": 38896.0, "reward": 5.125, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 0.75, "step": 16 }, { "completion_length": 5.25, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.25, "completions/mean_terminated_length": 5.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.00830078125, "frac_reward_zero_std": 0.0, "grad_norm": 11.182866096496582, "kl": 2.416404664516449, "learning_rate": 2.666666666666667e-06, "loss": 0.0967, "num_tokens": 41209.0, "reward": 4.75, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 4.75, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 17 }, { "completion_length": 5.75, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 5.75, "completions/mean_terminated_length": 5.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0087890625, "frac_reward_zero_std": 0.0, "grad_norm": 12.929946899414062, "kl": 3.8091089129447937, "learning_rate": 2.8333333333333335e-06, "loss": 0.1524, "num_tokens": 43532.0, "reward": 5.125, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 0.75, "step": 18 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.00927734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.4657467305660248, "kl": 1.6181318908929825, "learning_rate": 3e-06, "loss": 0.0647, "num_tokens": 45781.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 19 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.009765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.6211950182914734, "kl": 3.865095466375351, "learning_rate": 3.1666666666666667e-06, "loss": 0.1546, "num_tokens": 48281.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 20 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.01025390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3958570659160614, "kl": 3.5635141730308533, "learning_rate": 3.3333333333333333e-06, "loss": 0.1425, "num_tokens": 50887.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 21 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0107421875, "frac_reward_zero_std": 0.0, "grad_norm": 12.463068962097168, "kl": 12.036278486251831, "learning_rate": 3.5e-06, "loss": 0.4815, "num_tokens": 53007.0, "reward": 6.25, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 22 }, { "completion_length": 3.5, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.5, "completions/mean_terminated_length": 3.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.01123046875, "frac_reward_zero_std": 1.0, "grad_norm": 1.8087347745895386, "kl": 4.94687682390213, "learning_rate": 3.6666666666666666e-06, "loss": 0.1979, "num_tokens": 55805.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 23 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.3324650526046753, "kl": 1.9915877282619476, "learning_rate": 3.833333333333334e-06, "loss": 0.0797, "num_tokens": 58175.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 24 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01220703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.5965218544006348, "kl": 6.541197180747986, "learning_rate": 4.000000000000001e-06, "loss": 0.2616, "num_tokens": 60743.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 25 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0126953125, "frac_reward_zero_std": 1.0, "grad_norm": 2.5730605125427246, "kl": 4.600815117359161, "learning_rate": 4.166666666666667e-06, "loss": 0.184, "num_tokens": 63313.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 26 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01318359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.48865750432014465, "kl": 5.737705945968628, "learning_rate": 4.333333333333334e-06, "loss": 0.2295, "num_tokens": 65949.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 27 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.013671875, "frac_reward_zero_std": 1.0, "grad_norm": 1.0922263860702515, "kl": 3.4447925984859467, "learning_rate": 4.5e-06, "loss": 0.1378, "num_tokens": 68352.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 28 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01416015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.31673523783683777, "kl": 8.99358320236206, "learning_rate": 4.666666666666667e-06, "loss": 0.3597, "num_tokens": 70797.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 29 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0146484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.32408228516578674, "kl": 6.888558983802795, "learning_rate": 4.833333333333333e-06, "loss": 0.2755, "num_tokens": 73192.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 30 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01513671875, "frac_reward_zero_std": 1.0, "grad_norm": 5.942379474639893, "kl": 5.723478376865387, "learning_rate": 5e-06, "loss": 0.2289, "num_tokens": 75996.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 31 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.45663902163505554, "kl": 3.852832078933716, "learning_rate": 4.981481481481482e-06, "loss": 0.1541, "num_tokens": 78451.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 32 }, { "completion_length": 5.5, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 5.5, "completions/mean_terminated_length": 5.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01611328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.413408488035202, "kl": 3.2340221405029297, "learning_rate": 4.962962962962964e-06, "loss": 0.1294, "num_tokens": 81037.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 33 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0166015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.4474402368068695, "kl": 1.7874335646629333, "learning_rate": 4.944444444444445e-06, "loss": 0.0715, "num_tokens": 83470.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 34 }, { "completion_length": 3.25, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.25, "completions/mean_terminated_length": 3.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.01708984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.43918073177337646, "kl": 11.389363706111908, "learning_rate": 4.925925925925926e-06, "loss": 0.4556, "num_tokens": 85619.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 35 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.017578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.8948702216148376, "kl": 1.6696243733167648, "learning_rate": 4.907407407407408e-06, "loss": 0.0668, "num_tokens": 88192.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 36 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01806640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.2700890004634857, "kl": 3.7178112268447876, "learning_rate": 4.888888888888889e-06, "loss": 0.1487, "num_tokens": 90320.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 37 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0185546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.371006578207016, "kl": 2.1654927730560303, "learning_rate": 4.870370370370371e-06, "loss": 0.0866, "num_tokens": 92618.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 38 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01904296875, "frac_reward_zero_std": 0.0, "grad_norm": 15.53289794921875, "kl": 4.967282056808472, "learning_rate": 4.851851851851852e-06, "loss": 0.1987, "num_tokens": 94780.0, "reward": 6.25, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 39 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.01953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.2934619188308716, "kl": 8.455414772033691, "learning_rate": 4.833333333333333e-06, "loss": 0.3382, "num_tokens": 97056.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 40 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02001953125, "frac_reward_zero_std": 0.0, "grad_norm": 13.764455795288086, "kl": 1.0491378605365753, "learning_rate": 4.814814814814815e-06, "loss": 0.042, "num_tokens": 99276.0, "reward": 5.875, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.875, "rewards/showdown_reward_func/std": 0.75, "step": 41 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0205078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.3336067497730255, "kl": 10.553035259246826, "learning_rate": 4.796296296296297e-06, "loss": 0.4221, "num_tokens": 101660.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 42 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02099609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.39892303943634033, "kl": 1.0458447635173798, "learning_rate": 4.777777777777778e-06, "loss": 0.0418, "num_tokens": 104092.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 43 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.021484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.6236617565155029, "kl": 2.4145063161849976, "learning_rate": 4.75925925925926e-06, "loss": 0.0966, "num_tokens": 106504.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 44 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.02197265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.023313550278544426, "kl": 0.057367634028196335, "learning_rate": 4.7407407407407415e-06, "loss": 0.0023, "num_tokens": 108832.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 45 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0224609375, "frac_reward_zero_std": 1.0, "grad_norm": 1.156087875366211, "kl": 7.865663051605225, "learning_rate": 4.722222222222222e-06, "loss": 0.3146, "num_tokens": 111168.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 46 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02294921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.032481029629707336, "kl": 0.0028364360332489014, "learning_rate": 4.703703703703704e-06, "loss": 0.0001, "num_tokens": 113564.0, "reward": 8.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 8.0, "rewards/showdown_reward_func/std": 0.0, "step": 47 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0234375, "frac_reward_zero_std": 1.0, "grad_norm": 3.7456295490264893, "kl": 1.06026192009449, "learning_rate": 4.6851851851851855e-06, "loss": 0.0424, "num_tokens": 116096.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 48 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02392578125, "frac_reward_zero_std": 0.0, "grad_norm": 13.377860069274902, "kl": 11.406485557556152, "learning_rate": 4.666666666666667e-06, "loss": 0.4563, "num_tokens": 118564.0, "reward": 6.625, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 6.625, "rewards/showdown_reward_func/std": 0.75, "step": 49 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0244140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.8176776766777039, "kl": 10.981850624084473, "learning_rate": 4.648148148148148e-06, "loss": 0.4393, "num_tokens": 121208.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 50 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02490234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.008365115150809288, "kl": 0.073294997215271, "learning_rate": 4.62962962962963e-06, "loss": 0.0029, "num_tokens": 123516.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 51 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.025390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.30824124813079834, "kl": 8.523558616638184, "learning_rate": 4.611111111111112e-06, "loss": 0.3409, "num_tokens": 125936.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 52 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02587890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.553617537021637, "kl": 5.872042238712311, "learning_rate": 4.592592592592593e-06, "loss": 0.2349, "num_tokens": 128184.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 53 }, { "completion_length": 5.75, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.75, "completions/mean_terminated_length": 5.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0263671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.18106742203235626, "kl": 0.5723884664475918, "learning_rate": 4.5740740740740745e-06, "loss": 0.0229, "num_tokens": 130883.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 54 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.02685546875, "frac_reward_zero_std": 1.0, "grad_norm": 5.008867263793945, "kl": 3.4262692779302597, "learning_rate": 4.555555555555556e-06, "loss": 0.1371, "num_tokens": 133135.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 55 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.4350966513156891, "kl": 7.383725523948669, "learning_rate": 4.537037037037038e-06, "loss": 0.2953, "num_tokens": 135435.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 56 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02783203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.0043317885138094425, "kl": 0.013385593891143799, "learning_rate": 4.5185185185185185e-06, "loss": 0.0005, "num_tokens": 137703.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 57 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0283203125, "frac_reward_zero_std": 1.0, "grad_norm": 2.3241515159606934, "kl": 2.9452744722366333, "learning_rate": 4.5e-06, "loss": 0.1178, "num_tokens": 140202.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 58 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.02880859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.6177461743354797, "kl": 4.123246252536774, "learning_rate": 4.481481481481482e-06, "loss": 0.1649, "num_tokens": 142585.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 59 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.029296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.005939013324677944, "kl": 1.6706511974334717, "learning_rate": 4.462962962962963e-06, "loss": 0.0668, "num_tokens": 144681.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 60 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.02978515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.07706946134567261, "kl": 3.4405479431152344, "learning_rate": 4.444444444444444e-06, "loss": 0.1376, "num_tokens": 147097.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 61 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0302734375, "frac_reward_zero_std": 0.0, "grad_norm": 6.611453533172607, "kl": 0.8232481330633163, "learning_rate": 4.425925925925927e-06, "loss": 0.0329, "num_tokens": 149553.0, "reward": 6.25, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 62 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.03076171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.030268417671322823, "kl": 0.006503701210021973, "learning_rate": 4.407407407407408e-06, "loss": 0.0003, "num_tokens": 152065.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 63 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.03125, "frac_reward_zero_std": 1.0, "grad_norm": 0.4905981421470642, "kl": 6.123928725719452, "learning_rate": 4.388888888888889e-06, "loss": 0.245, "num_tokens": 154267.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 64 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03173828125, "frac_reward_zero_std": 0.0, "grad_norm": 14.905092239379883, "kl": 0.4855745881795883, "learning_rate": 4.370370370370371e-06, "loss": 0.0194, "num_tokens": 156575.0, "reward": 5.875, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.875, "rewards/showdown_reward_func/std": 0.75, "step": 65 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0322265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.29673489928245544, "kl": 1.1488926261663437, "learning_rate": 4.351851851851852e-06, "loss": 0.046, "num_tokens": 158963.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 66 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03271484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.05987896770238876, "kl": 2.526663154363632, "learning_rate": 4.333333333333334e-06, "loss": 0.1011, "num_tokens": 161462.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 67 }, { "completion_length": 3.25, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.25, "completions/mean_terminated_length": 3.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.033203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.297637939453125, "kl": 2.736666738986969, "learning_rate": 4.314814814814815e-06, "loss": 0.1095, "num_tokens": 163783.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 68 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03369140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.59696364402771, "kl": 3.9765483140945435, "learning_rate": 4.296296296296296e-06, "loss": 0.1591, "num_tokens": 166302.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 69 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0341796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.8638142943382263, "kl": 3.8630095422267914, "learning_rate": 4.277777777777778e-06, "loss": 0.1545, "num_tokens": 168947.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 70 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03466796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.2229151874780655, "kl": 3.1161839962005615, "learning_rate": 4.2592592592592596e-06, "loss": 0.1246, "num_tokens": 171571.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 71 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3344535827636719, "kl": 6.6893035769462585, "learning_rate": 4.240740740740741e-06, "loss": 0.2676, "num_tokens": 174133.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 72 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03564453125, "frac_reward_zero_std": 0.0, "grad_norm": 15.90652847290039, "kl": 1.6870811581611633, "learning_rate": 4.222222222222223e-06, "loss": 0.0675, "num_tokens": 176681.0, "reward": 6.25, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 73 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0361328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.6609954833984375, "kl": 1.1780510246753693, "learning_rate": 4.2037037037037045e-06, "loss": 0.0471, "num_tokens": 178973.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 74 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.03662109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.21606554090976715, "kl": 2.419235795736313, "learning_rate": 4.185185185185185e-06, "loss": 0.0968, "num_tokens": 181285.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 75 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.037109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.35656028985977173, "kl": 5.846519470214844, "learning_rate": 4.166666666666667e-06, "loss": 0.2339, "num_tokens": 183609.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 76 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03759765625, "frac_reward_zero_std": 0.0, "grad_norm": 18.146257400512695, "kl": 3.849558472633362, "learning_rate": 4.1481481481481485e-06, "loss": 0.154, "num_tokens": 185849.0, "reward": 5.875, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.875, "rewards/showdown_reward_func/std": 0.75, "step": 77 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0380859375, "frac_reward_zero_std": 1.0, "grad_norm": 5.484549045562744, "kl": 1.8681082129478455, "learning_rate": 4.12962962962963e-06, "loss": 0.0747, "num_tokens": 188401.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 78 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.03857421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.0007116745109669864, "kl": 0.0007021427154541016, "learning_rate": 4.111111111111111e-06, "loss": 0.0, "num_tokens": 190909.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 79 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.34523677825927734, "kl": 6.302359104156494, "learning_rate": 4.092592592592593e-06, "loss": 0.2521, "num_tokens": 193332.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 80 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.03955078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.4496091306209564, "kl": 0.33285149931907654, "learning_rate": 4.074074074074074e-06, "loss": 0.0133, "num_tokens": 195784.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 81 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0400390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.35587313771247864, "kl": 1.443438082933426, "learning_rate": 4.055555555555556e-06, "loss": 0.0577, "num_tokens": 198256.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 82 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.04052734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.015585344284772873, "kl": 2.6228585243225098, "learning_rate": 4.037037037037037e-06, "loss": 0.1049, "num_tokens": 200724.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 83 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.041015625, "frac_reward_zero_std": 0.0, "grad_norm": 24.365297317504883, "kl": 0.22576447576284409, "learning_rate": 4.018518518518519e-06, "loss": 0.009, "num_tokens": 202959.0, "reward": 5.875, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.875, "rewards/showdown_reward_func/std": 0.75, "step": 84 }, { "completion_length": 5.75, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 5.75, "completions/mean_terminated_length": 5.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.04150390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.24424317479133606, "kl": 3.225794494152069, "learning_rate": 4.000000000000001e-06, "loss": 0.129, "num_tokens": 205410.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 85 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0419921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.13267649710178375, "kl": 5.011456370353699, "learning_rate": 3.9814814814814814e-06, "loss": 0.2005, "num_tokens": 207588.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 86 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.04248046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.7399712800979614, "kl": 1.0110120298340917, "learning_rate": 3.962962962962963e-06, "loss": 0.0404, "num_tokens": 210028.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 87 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.04296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.002878983039408922, "kl": 0.6789208054542542, "learning_rate": 3.944444444444445e-06, "loss": 0.0272, "num_tokens": 212476.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 88 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.04345703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.41850200295448303, "kl": 1.0238634496927261, "learning_rate": 3.925925925925926e-06, "loss": 0.041, "num_tokens": 215123.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 89 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0439453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.02383102849125862, "kl": 3.526920795440674, "learning_rate": 3.907407407407408e-06, "loss": 0.1411, "num_tokens": 217599.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 90 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.04443359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.03795033320784569, "kl": 1.424944519996643, "learning_rate": 3.88888888888889e-06, "loss": 0.057, "num_tokens": 219991.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 91 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.044921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.23869675397872925, "kl": 2.8319059014320374, "learning_rate": 3.87037037037037e-06, "loss": 0.1133, "num_tokens": 222510.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 92 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.04541015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.5452110171318054, "kl": 2.0515754222869873, "learning_rate": 3.851851851851852e-06, "loss": 0.0821, "num_tokens": 224778.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 93 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0458984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.010540379211306572, "kl": 0.589465320110321, "learning_rate": 3.833333333333334e-06, "loss": 0.0236, "num_tokens": 227266.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 94 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.04638671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.0332580991089344, "kl": 7.763068675994873, "learning_rate": 3.814814814814815e-06, "loss": 0.3105, "num_tokens": 229870.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 95 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.152862548828125, "kl": 4.540526390075684, "learning_rate": 3.796296296296297e-06, "loss": 0.1816, "num_tokens": 232278.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 96 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.04736328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.04859086871147156, "kl": 7.060283660888672, "learning_rate": 3.777777777777778e-06, "loss": 0.2824, "num_tokens": 234798.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 97 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0478515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.7969298958778381, "kl": 5.902980804443359, "learning_rate": 3.7592592592592597e-06, "loss": 0.2361, "num_tokens": 236982.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 98 }, { "completion_length": 3.5, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.5, "completions/mean_terminated_length": 3.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.04833984375, "frac_reward_zero_std": 1.0, "grad_norm": 1.0662403106689453, "kl": 6.604215621948242, "learning_rate": 3.740740740740741e-06, "loss": 0.2642, "num_tokens": 239280.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 99 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.048828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.025562407448887825, "kl": 0.0035070478916168213, "learning_rate": 3.7222222222222225e-06, "loss": 0.0001, "num_tokens": 241564.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 100 }, { "completion_length": 3.0, "completions/clipped_ratio": 0.0, "completions/max_length": 3.0, "completions/max_terminated_length": 3.0, "completions/mean_length": 3.0, "completions/mean_terminated_length": 3.0, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.04931640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.43693795800209045, "kl": 1.8416004180908203, "learning_rate": 3.7037037037037037e-06, "loss": 0.0737, "num_tokens": 243976.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 101 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0498046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.32465389370918274, "kl": 6.581436075270176, "learning_rate": 3.6851851851851854e-06, "loss": 0.2633, "num_tokens": 246639.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 102 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05029296875, "frac_reward_zero_std": 0.0, "grad_norm": 12.13711166381836, "kl": 4.830296277999878, "learning_rate": 3.6666666666666666e-06, "loss": 0.1932, "num_tokens": 248971.0, "reward": 6.625, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 6.625, "rewards/showdown_reward_func/std": 0.75, "step": 103 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.05078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.0854559913277626, "kl": 1.377313494682312, "learning_rate": 3.6481481481481486e-06, "loss": 0.0551, "num_tokens": 251587.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 104 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.05126953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.09630150347948074, "kl": 1.3836550116539001, "learning_rate": 3.6296296296296302e-06, "loss": 0.0553, "num_tokens": 254187.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 105 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0517578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.20805594325065613, "kl": 2.479797899723053, "learning_rate": 3.6111111111111115e-06, "loss": 0.0992, "num_tokens": 256497.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 106 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.05224609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.0158888828009367, "kl": 0.6097385883331299, "learning_rate": 3.592592592592593e-06, "loss": 0.0244, "num_tokens": 259201.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 107 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.052734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.47654053568840027, "kl": 1.5832329466938972, "learning_rate": 3.5740740740740743e-06, "loss": 0.0633, "num_tokens": 261632.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 108 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.05322265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.001356764230877161, "kl": 0.0040444135665893555, "learning_rate": 3.555555555555556e-06, "loss": 0.0002, "num_tokens": 264236.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 109 }, { "completion_length": 3.5, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 3.5, "completions/mean_terminated_length": 3.5, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.0537109375, "frac_reward_zero_std": 0.0, "grad_norm": 24.048017501831055, "kl": 1.570328811649233, "learning_rate": 3.537037037037037e-06, "loss": 0.0628, "num_tokens": 266522.0, "reward": 6.25, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 6.25, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 110 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05419921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.4605531692504883, "kl": 1.2849935293197632, "learning_rate": 3.5185185185185187e-06, "loss": 0.0514, "num_tokens": 269188.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 111 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.6639642119407654, "kl": 1.6561651825904846, "learning_rate": 3.5e-06, "loss": 0.0662, "num_tokens": 271620.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 112 }, { "completion_length": 3.75, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 3.75, "completions/mean_terminated_length": 3.75, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.05517578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.6005472540855408, "kl": 8.446895480155945, "learning_rate": 3.481481481481482e-06, "loss": 0.3379, "num_tokens": 274055.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 113 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0556640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3825991749763489, "kl": 0.38878414779901505, "learning_rate": 3.4629629629629628e-06, "loss": 0.0156, "num_tokens": 276375.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 114 }, { "completion_length": 4.25, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.25, "completions/mean_terminated_length": 4.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05615234375, "frac_reward_zero_std": 1.0, "grad_norm": 1.662015676498413, "kl": 4.800809904932976, "learning_rate": 3.444444444444445e-06, "loss": 0.192, "num_tokens": 278672.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 115 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.056640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.21853791177272797, "kl": 5.275329113006592, "learning_rate": 3.4259259259259265e-06, "loss": 0.211, "num_tokens": 281016.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 116 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05712890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.5037992596626282, "kl": 1.0388526264578104, "learning_rate": 3.4074074074074077e-06, "loss": 0.0416, "num_tokens": 283811.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 117 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0576171875, "frac_reward_zero_std": 1.0, "grad_norm": 1.1381536722183228, "kl": 1.1480399072170258, "learning_rate": 3.3888888888888893e-06, "loss": 0.0459, "num_tokens": 285841.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 118 }, { "completion_length": 4.5, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.5, "completions/mean_terminated_length": 4.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05810546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.681000828742981, "kl": 2.220241963863373, "learning_rate": 3.3703703703703705e-06, "loss": 0.0888, "num_tokens": 288163.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 119 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3646780252456665, "kl": 0.7028678879141808, "learning_rate": 3.351851851851852e-06, "loss": 0.0281, "num_tokens": 291151.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 120 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.05908203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.1152932420372963, "kl": 0.03880554437637329, "learning_rate": 3.3333333333333333e-06, "loss": 0.0016, "num_tokens": 293619.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 121 }, { "completion_length": 5.25, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.25, "completions/mean_terminated_length": 5.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0595703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.3502674102783203, "kl": 1.8929688036441803, "learning_rate": 3.314814814814815e-06, "loss": 0.0757, "num_tokens": 296044.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 122 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.06005859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5672823190689087, "kl": 4.6738598346710205, "learning_rate": 3.296296296296296e-06, "loss": 0.187, "num_tokens": 298251.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 123 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.060546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.2383938580751419, "kl": 2.899338483810425, "learning_rate": 3.277777777777778e-06, "loss": 0.116, "num_tokens": 300551.0, "reward": 7.0, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 7.0, "rewards/showdown_reward_func/std": 0.0, "step": 124 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06103515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.021955661475658417, "kl": 5.0965962409973145, "learning_rate": 3.25925925925926e-06, "loss": 0.2039, "num_tokens": 302871.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 125 }, { "completion_length": 6.75, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 6.75, "completions/mean_terminated_length": 6.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.0615234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.19196857511997223, "kl": 2.3878124803304672, "learning_rate": 3.240740740740741e-06, "loss": 0.0955, "num_tokens": 305426.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 126 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06201171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.0334867425262928, "kl": 1.4965951442718506, "learning_rate": 3.2222222222222227e-06, "loss": 0.0599, "num_tokens": 307790.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 127 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.0625, "frac_reward_zero_std": 1.0, "grad_norm": 0.09326554834842682, "kl": 3.4022045135498047, "learning_rate": 3.203703703703704e-06, "loss": 0.1361, "num_tokens": 310490.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 128 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06298828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.02333025075495243, "kl": 2.7339894771575928, "learning_rate": 3.1851851851851855e-06, "loss": 0.1094, "num_tokens": 312738.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 129 }, { "completion_length": 6.5, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 6.5, "completions/mean_terminated_length": 6.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.0634765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.231577530503273, "kl": 4.253310561180115, "learning_rate": 3.1666666666666667e-06, "loss": 0.1701, "num_tokens": 314992.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 130 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.06396484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.010527431964874268, "kl": 1.153098464012146, "learning_rate": 3.1481481481481483e-06, "loss": 0.0461, "num_tokens": 317468.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 131 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.064453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.00619816267862916, "kl": 5.526472091674805, "learning_rate": 3.1296296296296295e-06, "loss": 0.2211, "num_tokens": 319836.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 132 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06494140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.028091223910450935, "kl": 2.9685192108154297, "learning_rate": 3.1111111111111116e-06, "loss": 0.1187, "num_tokens": 322300.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 133 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0654296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.3335534632205963, "kl": 5.209336638450623, "learning_rate": 3.0925925925925928e-06, "loss": 0.2084, "num_tokens": 324767.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 134 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06591796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.007512012962251902, "kl": 2.894021987915039, "learning_rate": 3.0740740740740744e-06, "loss": 0.1158, "num_tokens": 327431.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 135 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.06640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.04429686814546585, "kl": 1.9940108060836792, "learning_rate": 3.055555555555556e-06, "loss": 0.0798, "num_tokens": 329847.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 136 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.06689453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.21994587779045105, "kl": 5.019688606262207, "learning_rate": 3.0370370370370372e-06, "loss": 0.2008, "num_tokens": 332211.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 137 }, { "completion_length": 6.25, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 6.25, "completions/mean_terminated_length": 6.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0673828125, "frac_reward_zero_std": 1.0, "grad_norm": 1.2869529724121094, "kl": 4.645696938037872, "learning_rate": 3.018518518518519e-06, "loss": 0.1858, "num_tokens": 334808.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 138 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.06787109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.026302624493837357, "kl": 2.2573835849761963, "learning_rate": 3e-06, "loss": 0.0903, "num_tokens": 337460.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 139 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.068359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.0010225351434201002, "kl": 0.020522713661193848, "learning_rate": 2.9814814814814817e-06, "loss": 0.0008, "num_tokens": 339736.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 140 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.06884765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.29835352301597595, "kl": 2.5923999547958374, "learning_rate": 2.962962962962963e-06, "loss": 0.1037, "num_tokens": 342220.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 141 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.0693359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.0008473648340441287, "kl": 7.624924182891846e-05, "learning_rate": 2.944444444444445e-06, "loss": 0.0, "num_tokens": 344508.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 142 }, { "completion_length": 6.5, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 6.5, "completions/mean_terminated_length": 6.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.06982421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.08786490559577942, "kl": 3.773865282535553, "learning_rate": 2.9259259259259257e-06, "loss": 0.151, "num_tokens": 346754.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 143 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.18450479209423065, "kl": 4.580117702484131, "learning_rate": 2.907407407407408e-06, "loss": 0.1832, "num_tokens": 349339.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 144 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.07080078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.21155954897403717, "kl": 4.817610740661621, "learning_rate": 2.888888888888889e-06, "loss": 0.1927, "num_tokens": 351831.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 145 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.0712890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.4896208643913269, "kl": 3.9626405239105225, "learning_rate": 2.8703703703703706e-06, "loss": 0.1585, "num_tokens": 354351.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 146 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.07177734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3946496248245239, "kl": 1.2275368142873049, "learning_rate": 2.8518518518518522e-06, "loss": 0.0491, "num_tokens": 356726.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 147 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.072265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3141247034072876, "kl": 4.836394190788269, "learning_rate": 2.8333333333333335e-06, "loss": 0.1935, "num_tokens": 359166.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 148 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.07275390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.08556187152862549, "kl": 0.9447380900382996, "learning_rate": 2.814814814814815e-06, "loss": 0.0378, "num_tokens": 361574.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 149 }, { "completion_length": 6.25, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 6.25, "completions/mean_terminated_length": 6.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0732421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.2837235629558563, "kl": 5.791716158390045, "learning_rate": 2.7962962962962963e-06, "loss": 0.2317, "num_tokens": 363803.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 150 }, { "completion_length": 9.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.5, "completions/mean_terminated_length": 9.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.07373046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.27554503083229065, "kl": 6.049068570137024, "learning_rate": 2.7777777777777783e-06, "loss": 0.242, "num_tokens": 366245.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 151 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.07421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.16460341215133667, "kl": 2.872504413127899, "learning_rate": 2.759259259259259e-06, "loss": 0.1149, "num_tokens": 368615.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 152 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.07470703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.4346492290496826, "kl": 1.19341541826725, "learning_rate": 2.740740740740741e-06, "loss": 0.0477, "num_tokens": 371099.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 153 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.0751953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.17994923889636993, "kl": 3.9729872047901154, "learning_rate": 2.7222222222222224e-06, "loss": 0.1589, "num_tokens": 373617.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 154 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 11.0, "completions/max_terminated_length": 11.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.07568359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5378056168556213, "kl": 4.058318495750427, "learning_rate": 2.703703703703704e-06, "loss": 0.1623, "num_tokens": 376401.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 155 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.076171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.0004746035556308925, "kl": 1.4334917068481445e-05, "learning_rate": 2.6851851851851856e-06, "loss": 0.0, "num_tokens": 378733.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 156 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.07666015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.14175662398338318, "kl": 2.4374873638153076, "learning_rate": 2.666666666666667e-06, "loss": 0.0975, "num_tokens": 381341.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 157 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0771484375, "frac_reward_zero_std": 0.0, "grad_norm": 12.361019134521484, "kl": 5.399029493331909, "learning_rate": 2.6481481481481485e-06, "loss": 0.216, "num_tokens": 383735.0, "reward": 5.75, "reward_std": 0.5, "rewards/showdown_reward_func/mean": 5.75, "rewards/showdown_reward_func/std": 0.5, "step": 158 }, { "completion_length": 9.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.75, "completions/mean_terminated_length": 9.75, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.07763671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.12494304031133652, "kl": 5.293493807315826, "learning_rate": 2.6296296296296297e-06, "loss": 0.2117, "num_tokens": 386390.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 159 }, { "completion_length": 10.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 10.0, "completions/mean_terminated_length": 10.0, "completions/min_length": 10.0, "completions/min_terminated_length": 10.0, "epoch": 0.078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.13608790934085846, "kl": 3.48713755607605, "learning_rate": 2.6111111111111113e-06, "loss": 0.1395, "num_tokens": 389134.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 160 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.07861328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.1899229735136032, "kl": 2.72549307346344, "learning_rate": 2.5925925925925925e-06, "loss": 0.109, "num_tokens": 391518.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 161 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0791015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.0027998480945825577, "kl": 0.007664513774216175, "learning_rate": 2.5740740740740745e-06, "loss": 0.0003, "num_tokens": 393778.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 162 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.07958984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.09515848755836487, "kl": 4.079674243927002, "learning_rate": 2.5555555555555557e-06, "loss": 0.1632, "num_tokens": 396310.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 163 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.080078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.15099413692951202, "kl": 1.5627632736577652, "learning_rate": 2.5370370370370374e-06, "loss": 0.0625, "num_tokens": 398730.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 164 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08056640625, "frac_reward_zero_std": 1.0, "grad_norm": 1.2412798404693604, "kl": 12.494128942489624, "learning_rate": 2.5185185185185186e-06, "loss": 0.4998, "num_tokens": 401216.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 165 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.0810546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.23200121521949768, "kl": 5.383057713508606, "learning_rate": 2.5e-06, "loss": 0.2153, "num_tokens": 403604.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 166 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08154296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.010226655751466751, "kl": 1.0866353511810303, "learning_rate": 2.481481481481482e-06, "loss": 0.0435, "num_tokens": 406064.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 167 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.3672674894332886, "kl": 4.310274720191956, "learning_rate": 2.462962962962963e-06, "loss": 0.1724, "num_tokens": 408540.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 168 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.08251953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.17687800526618958, "kl": 3.8067517280578613, "learning_rate": 2.4444444444444447e-06, "loss": 0.1523, "num_tokens": 410836.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 169 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.0830078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.006304240319877863, "kl": 1.336727261543274, "learning_rate": 2.425925925925926e-06, "loss": 0.0535, "num_tokens": 413372.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 170 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08349609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.1857936680316925, "kl": 0.6394333839416504, "learning_rate": 2.4074074074074075e-06, "loss": 0.0256, "num_tokens": 415696.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 171 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.083984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.10984928160905838, "kl": 4.449400186538696, "learning_rate": 2.388888888888889e-06, "loss": 0.178, "num_tokens": 418340.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 172 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08447265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.01759975403547287, "kl": 0.4628032147884369, "learning_rate": 2.3703703703703707e-06, "loss": 0.0185, "num_tokens": 420904.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 173 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.0849609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3516920208930969, "kl": 7.578170537948608, "learning_rate": 2.351851851851852e-06, "loss": 0.3031, "num_tokens": 423124.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 174 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.08544921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.12523461878299713, "kl": 0.11241680383682251, "learning_rate": 2.3333333333333336e-06, "loss": 0.0045, "num_tokens": 425420.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 175 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.33217379450798035, "kl": 0.9992741048336029, "learning_rate": 2.314814814814815e-06, "loss": 0.04, "num_tokens": 427859.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 176 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08642578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.25113457441329956, "kl": 2.7541376408189535, "learning_rate": 2.2962962962962964e-06, "loss": 0.1102, "num_tokens": 430192.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 177 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0869140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.10678017139434814, "kl": 1.9695679247379303, "learning_rate": 2.277777777777778e-06, "loss": 0.0788, "num_tokens": 432456.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 178 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08740234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.9062843918800354, "kl": 6.54217004776001, "learning_rate": 2.2592592592592592e-06, "loss": 0.2617, "num_tokens": 434756.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 179 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.087890625, "frac_reward_zero_std": 1.0, "grad_norm": 3.1022541406855453e-06, "kl": 2.2451083623309387e-06, "learning_rate": 2.240740740740741e-06, "loss": 0.0, "num_tokens": 437008.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 180 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.08837890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.06636836379766464, "kl": 2.3290655314922333, "learning_rate": 2.222222222222222e-06, "loss": 0.0932, "num_tokens": 439356.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 181 }, { "completion_length": 6.25, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 6.25, "completions/mean_terminated_length": 6.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.0888671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.4140000343322754, "kl": 2.4567487835884094, "learning_rate": 2.203703703703704e-06, "loss": 0.0983, "num_tokens": 441645.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 182 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.08935546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.14062437415122986, "kl": 3.0142419040203094, "learning_rate": 2.1851851851851853e-06, "loss": 0.1206, "num_tokens": 444161.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 183 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.08984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.03000878542661667, "kl": 2.3112518787384033, "learning_rate": 2.166666666666667e-06, "loss": 0.0925, "num_tokens": 446773.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 184 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.09033203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.6564126014709473, "kl": 6.488188743591309, "learning_rate": 2.148148148148148e-06, "loss": 0.2595, "num_tokens": 449081.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 185 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0908203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.5221624374389648, "kl": 5.600887298583984, "learning_rate": 2.1296296296296298e-06, "loss": 0.224, "num_tokens": 451325.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 186 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09130859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.12265491485595703, "kl": 0.01206827163696289, "learning_rate": 2.1111111111111114e-06, "loss": 0.0005, "num_tokens": 453581.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 187 }, { "completion_length": 12.0, "completions/clipped_ratio": 0.0, "completions/max_length": 13.0, "completions/max_terminated_length": 13.0, "completions/mean_length": 12.0, "completions/mean_terminated_length": 12.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.091796875, "frac_reward_zero_std": 0.0, "grad_norm": 4.592714786529541, "kl": 3.3034667372703552, "learning_rate": 2.0925925925925926e-06, "loss": 0.1321, "num_tokens": 455817.0, "reward": 5.125, "reward_std": 0.75, "rewards/showdown_reward_func/mean": 5.125, "rewards/showdown_reward_func/std": 0.75, "step": 188 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.09228515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.10360771417617798, "kl": 4.065852105617523, "learning_rate": 2.0740740740740742e-06, "loss": 0.1626, "num_tokens": 458038.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 189 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0927734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3856542110443115, "kl": 3.8794945627450943, "learning_rate": 2.0555555555555555e-06, "loss": 0.1552, "num_tokens": 460442.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 190 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.09326171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.12536181509494781, "kl": 4.723106622695923, "learning_rate": 2.037037037037037e-06, "loss": 0.1889, "num_tokens": 462776.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 191 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09375, "frac_reward_zero_std": 1.0, "grad_norm": 0.11561095714569092, "kl": 6.818011999130249, "learning_rate": 2.0185185185185187e-06, "loss": 0.2727, "num_tokens": 465192.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 192 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.09423828125, "frac_reward_zero_std": 0.0, "grad_norm": 9.071577072143555, "kl": 2.3952170610427856, "learning_rate": 2.0000000000000003e-06, "loss": 0.0958, "num_tokens": 467806.0, "reward": 4.75, "reward_std": 0.8660253882408142, "rewards/showdown_reward_func/mean": 4.75, "rewards/showdown_reward_func/std": 0.8660253882408142, "step": 193 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.0947265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.23048385977745056, "kl": 4.108124792575836, "learning_rate": 1.9814814814814815e-06, "loss": 0.1643, "num_tokens": 470418.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 194 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09521484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5544969439506531, "kl": 3.145868271589279, "learning_rate": 1.962962962962963e-06, "loss": 0.1258, "num_tokens": 472910.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 195 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.095703125, "frac_reward_zero_std": 1.0, "grad_norm": 1.1315934658050537, "kl": 3.615946650505066, "learning_rate": 1.944444444444445e-06, "loss": 0.1446, "num_tokens": 475285.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 196 }, { "completion_length": 6.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.75, "completions/mean_terminated_length": 6.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.09619140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.5950599312782288, "kl": 1.6280896365642548, "learning_rate": 1.925925925925926e-06, "loss": 0.0651, "num_tokens": 477776.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 197 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.0966796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.26538848876953125, "kl": 3.717039406299591, "learning_rate": 1.9074074074074076e-06, "loss": 0.1487, "num_tokens": 480160.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 198 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09716796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.5169658660888672, "kl": 4.131194859743118, "learning_rate": 1.888888888888889e-06, "loss": 0.1652, "num_tokens": 482681.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 199 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.05411991849541664, "kl": 1.0043045997663285, "learning_rate": 1.8703703703703705e-06, "loss": 0.0402, "num_tokens": 485081.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 200 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.09814453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.10347495973110199, "kl": 6.020248115062714, "learning_rate": 1.8518518518518519e-06, "loss": 0.2408, "num_tokens": 487372.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 201 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.0986328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.20395448803901672, "kl": 5.228438138961792, "learning_rate": 1.8333333333333333e-06, "loss": 0.2091, "num_tokens": 490028.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 202 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.09912109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.9715097546577454, "kl": 4.245141506195068, "learning_rate": 1.8148148148148151e-06, "loss": 0.1698, "num_tokens": 492496.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 203 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 11.0, "completions/max_terminated_length": 11.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.099609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.25802063941955566, "kl": 3.3548801988363266, "learning_rate": 1.7962962962962965e-06, "loss": 0.1342, "num_tokens": 494983.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 204 }, { "completion_length": 10.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 10.0, "completions/mean_terminated_length": 10.0, "completions/min_length": 10.0, "completions/min_terminated_length": 10.0, "epoch": 0.10009765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.07226474583148956, "kl": 4.573369026184082, "learning_rate": 1.777777777777778e-06, "loss": 0.1829, "num_tokens": 497427.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 205 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1005859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.22307896614074707, "kl": 3.8933595418930054, "learning_rate": 1.7592592592592594e-06, "loss": 0.1557, "num_tokens": 499783.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 206 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.10107421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.5825552344322205, "kl": 3.364398956298828, "learning_rate": 1.740740740740741e-06, "loss": 0.1346, "num_tokens": 502631.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 207 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.1015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.2188757359981537, "kl": 5.777818143367767, "learning_rate": 1.7222222222222224e-06, "loss": 0.2311, "num_tokens": 504923.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 208 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.10205078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.33492913842201233, "kl": 5.460749328136444, "learning_rate": 1.7037037037037038e-06, "loss": 0.2184, "num_tokens": 507384.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 209 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.1025390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.10738930106163025, "kl": 4.29905366897583, "learning_rate": 1.6851851851851852e-06, "loss": 0.172, "num_tokens": 509858.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 210 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.10302734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.5743770003318787, "kl": 3.2954869978129864, "learning_rate": 1.6666666666666667e-06, "loss": 0.1318, "num_tokens": 512344.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 211 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.103515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3180544972419739, "kl": 2.7259351313114166, "learning_rate": 1.648148148148148e-06, "loss": 0.109, "num_tokens": 514807.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 212 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.10400390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.1163901835680008, "kl": 7.028075337409973, "learning_rate": 1.62962962962963e-06, "loss": 0.2811, "num_tokens": 517064.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 213 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.1044921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.3700006604194641, "kl": 4.933626249432564, "learning_rate": 1.6111111111111113e-06, "loss": 0.1973, "num_tokens": 519568.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 214 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.10498046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.27510184049606323, "kl": 7.494894504547119, "learning_rate": 1.5925925925925927e-06, "loss": 0.2998, "num_tokens": 521998.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 215 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.10546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.7730336785316467, "kl": 2.871400460600853, "learning_rate": 1.5740740740740742e-06, "loss": 0.1149, "num_tokens": 524503.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 216 }, { "completion_length": 6.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.75, "completions/mean_terminated_length": 6.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.10595703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.22861027717590332, "kl": 2.7901344895362854, "learning_rate": 1.5555555555555558e-06, "loss": 0.1116, "num_tokens": 526858.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 217 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1064453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.8214117288589478, "kl": 4.741910457611084, "learning_rate": 1.5370370370370372e-06, "loss": 0.1897, "num_tokens": 529378.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 218 }, { "completion_length": 5.25, "completions/clipped_ratio": 0.0, "completions/max_length": 6.0, "completions/max_terminated_length": 6.0, "completions/mean_length": 5.25, "completions/mean_terminated_length": 5.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.10693359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.16543854773044586, "kl": 1.38276606798172, "learning_rate": 1.5185185185185186e-06, "loss": 0.0553, "num_tokens": 531859.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 219 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.107421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.8888351917266846, "kl": 2.417250782251358, "learning_rate": 1.5e-06, "loss": 0.0967, "num_tokens": 534143.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 220 }, { "completion_length": 5.25, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 5.25, "completions/mean_terminated_length": 5.25, "completions/min_length": 3.0, "completions/min_terminated_length": 3.0, "epoch": 0.10791015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.3612446188926697, "kl": 7.537065202370286, "learning_rate": 1.4814814814814815e-06, "loss": 0.3015, "num_tokens": 536728.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 221 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1083984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.15502597391605377, "kl": 4.3510245978832245, "learning_rate": 1.4629629629629629e-06, "loss": 0.174, "num_tokens": 539256.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 222 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.10888671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.0940648764371872, "kl": 7.621327519416809, "learning_rate": 1.4444444444444445e-06, "loss": 0.3049, "num_tokens": 541512.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 223 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3595089912414551, "kl": 2.116122618317604, "learning_rate": 1.4259259259259261e-06, "loss": 0.0846, "num_tokens": 543924.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 224 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.10986328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.11729490011930466, "kl": 9.008839130401611, "learning_rate": 1.4074074074074075e-06, "loss": 0.3604, "num_tokens": 546370.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 225 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 11.0, "completions/max_terminated_length": 11.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.1103515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.2782983183860779, "kl": 5.948825657367706, "learning_rate": 1.3888888888888892e-06, "loss": 0.238, "num_tokens": 548781.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 226 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.11083984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.30998173356056213, "kl": 4.765108704566956, "learning_rate": 1.3703703703703706e-06, "loss": 0.1906, "num_tokens": 551345.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 227 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.111328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.15352797508239746, "kl": 2.994396686553955, "learning_rate": 1.351851851851852e-06, "loss": 0.1198, "num_tokens": 554001.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 228 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.11181640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.2045368254184723, "kl": 4.165620267391205, "learning_rate": 1.3333333333333334e-06, "loss": 0.1666, "num_tokens": 556862.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 229 }, { "completion_length": 5.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 5.25, "completions/mean_terminated_length": 5.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.1123046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.36610859632492065, "kl": 3.9922608733177185, "learning_rate": 1.3148148148148148e-06, "loss": 0.1597, "num_tokens": 559231.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 230 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.11279296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.24757589399814606, "kl": 6.000028967857361, "learning_rate": 1.2962962962962962e-06, "loss": 0.24, "num_tokens": 561900.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 231 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.11328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.23765164613723755, "kl": 5.253152847290039, "learning_rate": 1.2777777777777779e-06, "loss": 0.2101, "num_tokens": 564416.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 232 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.11376953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.12813617289066315, "kl": 5.701294034719467, "learning_rate": 1.2592592592592593e-06, "loss": 0.2281, "num_tokens": 567081.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 233 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.1142578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.09155172854661942, "kl": 7.213868856430054, "learning_rate": 1.240740740740741e-06, "loss": 0.2886, "num_tokens": 569281.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 234 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.11474609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.10943949222564697, "kl": 8.968728303909302, "learning_rate": 1.2222222222222223e-06, "loss": 0.3587, "num_tokens": 571610.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 235 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.115234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.08101601898670197, "kl": 7.308959007263184, "learning_rate": 1.2037037037037037e-06, "loss": 0.2924, "num_tokens": 573926.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 236 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.11572265625, "frac_reward_zero_std": 1.0, "grad_norm": 0.09388670325279236, "kl": 4.292471170425415, "learning_rate": 1.1851851851851854e-06, "loss": 0.1717, "num_tokens": 576566.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 237 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1162109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.22590592503547668, "kl": 6.105626165866852, "learning_rate": 1.1666666666666668e-06, "loss": 0.2442, "num_tokens": 578700.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 238 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.11669921875, "frac_reward_zero_std": 1.0, "grad_norm": 0.513176441192627, "kl": 2.6796352565288544, "learning_rate": 1.1481481481481482e-06, "loss": 0.1072, "num_tokens": 581246.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 239 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.1171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.22581928968429565, "kl": 6.670910835266113, "learning_rate": 1.1296296296296296e-06, "loss": 0.2668, "num_tokens": 583842.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 240 }, { "completion_length": 9.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.5, "completions/mean_terminated_length": 9.5, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.11767578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.35623806715011597, "kl": 7.097242712974548, "learning_rate": 1.111111111111111e-06, "loss": 0.2839, "num_tokens": 586256.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 241 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.1181640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.24214425683021545, "kl": 2.4126784801483154, "learning_rate": 1.0925925925925927e-06, "loss": 0.0965, "num_tokens": 588639.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 242 }, { "completion_length": 10.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 10.0, "completions/mean_terminated_length": 10.0, "completions/min_length": 10.0, "completions/min_terminated_length": 10.0, "epoch": 0.11865234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.09642838686704636, "kl": 5.443140506744385, "learning_rate": 1.074074074074074e-06, "loss": 0.2177, "num_tokens": 591155.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 243 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.119140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.6441452503204346, "kl": 4.6939539313316345, "learning_rate": 1.0555555555555557e-06, "loss": 0.1878, "num_tokens": 593463.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 244 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.11962890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.35765326023101807, "kl": 3.942962110042572, "learning_rate": 1.0370370370370371e-06, "loss": 0.1577, "num_tokens": 595980.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 245 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1201171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.06273625791072845, "kl": 3.8977761268615723, "learning_rate": 1.0185185185185185e-06, "loss": 0.1559, "num_tokens": 598352.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 246 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.12060546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.1850065290927887, "kl": 4.137905716896057, "learning_rate": 1.0000000000000002e-06, "loss": 0.1655, "num_tokens": 600911.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 247 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.12109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.1793782263994217, "kl": 5.5979530811309814, "learning_rate": 9.814814814814816e-07, "loss": 0.2239, "num_tokens": 603237.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 248 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.12158203125, "frac_reward_zero_std": 1.0, "grad_norm": 0.17351923882961273, "kl": 5.10476279258728, "learning_rate": 9.62962962962963e-07, "loss": 0.2042, "num_tokens": 605841.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 249 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.1220703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.16786052286624908, "kl": 4.248399376869202, "learning_rate": 9.444444444444445e-07, "loss": 0.1699, "num_tokens": 608172.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 250 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.12255859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.11953893303871155, "kl": 8.075819969177246, "learning_rate": 9.259259259259259e-07, "loss": 0.323, "num_tokens": 610604.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 251 }, { "completion_length": 9.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.25, "completions/mean_terminated_length": 9.25, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.123046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.31080305576324463, "kl": 5.813385963439941, "learning_rate": 9.074074074074076e-07, "loss": 0.2325, "num_tokens": 613045.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 252 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.12353515625, "frac_reward_zero_std": 1.0, "grad_norm": 0.5183307528495789, "kl": 2.368314176797867, "learning_rate": 8.88888888888889e-07, "loss": 0.0947, "num_tokens": 615521.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 253 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.1240234375, "frac_reward_zero_std": 1.0, "grad_norm": 0.32232245802879333, "kl": 7.091058850288391, "learning_rate": 8.703703703703705e-07, "loss": 0.2836, "num_tokens": 617980.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 254 }, { "completion_length": 10.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 10.0, "completions/mean_terminated_length": 10.0, "completions/min_length": 10.0, "completions/min_terminated_length": 10.0, "epoch": 0.12451171875, "frac_reward_zero_std": 1.0, "grad_norm": 0.06373075395822525, "kl": 3.709517478942871, "learning_rate": 8.518518518518519e-07, "loss": 0.1484, "num_tokens": 620432.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 255 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.125, "frac_reward_zero_std": 1.0, "grad_norm": 0.0952451229095459, "kl": 2.6823215782642365, "learning_rate": 8.333333333333333e-07, "loss": 0.1073, "num_tokens": 622981.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 256 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.12548828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.21413163840770721, "kl": 2.692964404821396, "learning_rate": 8.14814814814815e-07, "loss": 0.1077, "num_tokens": 625373.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 257 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.1259765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.09951522201299667, "kl": 5.755279541015625, "learning_rate": 7.962962962962964e-07, "loss": 0.2302, "num_tokens": 627979.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 258 }, { "completion_length": 9.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.75, "completions/mean_terminated_length": 9.75, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.12646484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.07772009819746017, "kl": 4.023648262023926, "learning_rate": 7.777777777777779e-07, "loss": 0.1609, "num_tokens": 630386.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 259 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.126953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.41609689593315125, "kl": 4.146004110574722, "learning_rate": 7.592592592592593e-07, "loss": 0.1658, "num_tokens": 632752.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 260 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.12744140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.0842471793293953, "kl": 6.2071133852005005, "learning_rate": 7.407407407407407e-07, "loss": 0.2483, "num_tokens": 634975.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 261 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.1279296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.1591452658176422, "kl": 8.810772180557251, "learning_rate": 7.222222222222222e-07, "loss": 0.3524, "num_tokens": 637842.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 262 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.12841796875, "frac_reward_zero_std": 1.0, "grad_norm": 0.3712395131587982, "kl": 5.303708553314209, "learning_rate": 7.037037037037038e-07, "loss": 0.2121, "num_tokens": 640270.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 263 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.12890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.20163346827030182, "kl": 7.655542016029358, "learning_rate": 6.851851851851853e-07, "loss": 0.3062, "num_tokens": 642514.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 264 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.12939453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.16366910934448242, "kl": 8.202631950378418, "learning_rate": 6.666666666666667e-07, "loss": 0.3281, "num_tokens": 645306.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 265 }, { "completion_length": 6.0, "completions/clipped_ratio": 0.0, "completions/max_length": 8.0, "completions/max_terminated_length": 8.0, "completions/mean_length": 6.0, "completions/mean_terminated_length": 6.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.1298828125, "frac_reward_zero_std": 1.0, "grad_norm": 0.7851002216339111, "kl": 5.521994948387146, "learning_rate": 6.481481481481481e-07, "loss": 0.2209, "num_tokens": 648158.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 266 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 7.0, "completions/min_terminated_length": 7.0, "epoch": 0.13037109375, "frac_reward_zero_std": 1.0, "grad_norm": 0.1023290753364563, "kl": 4.904609680175781, "learning_rate": 6.296296296296296e-07, "loss": 0.1962, "num_tokens": 650390.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 267 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.130859375, "frac_reward_zero_std": 1.0, "grad_norm": 0.3921956419944763, "kl": 4.057255297899246, "learning_rate": 6.111111111111112e-07, "loss": 0.1623, "num_tokens": 652770.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 268 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.13134765625, "frac_reward_zero_std": 1.0, "grad_norm": 3.2410147190093994, "kl": 5.109633386135101, "learning_rate": 5.925925925925927e-07, "loss": 0.2044, "num_tokens": 655083.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 269 }, { "completion_length": 9.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 9.0, "completions/mean_terminated_length": 9.0, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.1318359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.1031208410859108, "kl": 4.560241222381592, "learning_rate": 5.740740740740741e-07, "loss": 0.1824, "num_tokens": 657595.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 270 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.13232421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.14495760202407837, "kl": 5.710577346384525, "learning_rate": 5.555555555555555e-07, "loss": 0.2284, "num_tokens": 659837.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 271 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.1328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.332975834608078, "kl": 4.155658662319183, "learning_rate": 5.37037037037037e-07, "loss": 0.1662, "num_tokens": 662597.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 272 }, { "completion_length": 8.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.75, "completions/mean_terminated_length": 8.75, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.13330078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.12158675491809845, "kl": 4.373098373413086, "learning_rate": 5.185185185185186e-07, "loss": 0.1749, "num_tokens": 664960.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 273 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.1337890625, "frac_reward_zero_std": 1.0, "grad_norm": 0.14800402522087097, "kl": 4.969854176044464, "learning_rate": 5.000000000000001e-07, "loss": 0.1988, "num_tokens": 667542.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 274 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.13427734375, "frac_reward_zero_std": 1.0, "grad_norm": 0.6896809935569763, "kl": 5.0034394562244415, "learning_rate": 4.814814814814815e-07, "loss": 0.2001, "num_tokens": 670083.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 275 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.134765625, "frac_reward_zero_std": 1.0, "grad_norm": 0.07263071835041046, "kl": 5.373521208763123, "learning_rate": 4.6296296296296297e-07, "loss": 0.2149, "num_tokens": 672559.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 276 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.13525390625, "frac_reward_zero_std": 1.0, "grad_norm": 0.043924469500780106, "kl": 3.768325090408325, "learning_rate": 4.444444444444445e-07, "loss": 0.1507, "num_tokens": 674851.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 277 }, { "completion_length": 9.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.5, "completions/mean_terminated_length": 9.5, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.1357421875, "frac_reward_zero_std": 1.0, "grad_norm": 0.30079495906829834, "kl": 6.40265679359436, "learning_rate": 4.2592592592592596e-07, "loss": 0.2561, "num_tokens": 677205.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 278 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.13623046875, "frac_reward_zero_std": 1.0, "grad_norm": 0.23181484639644623, "kl": 4.723723351955414, "learning_rate": 4.074074074074075e-07, "loss": 0.1889, "num_tokens": 679602.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 279 }, { "completion_length": 8.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.5, "completions/mean_terminated_length": 8.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.13671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.1326865553855896, "kl": 4.769572019577026, "learning_rate": 3.8888888888888895e-07, "loss": 0.1908, "num_tokens": 681888.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 280 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.13720703125, "frac_reward_zero_std": 1.0, "grad_norm": 0.22778980433940887, "kl": 3.12056964635849, "learning_rate": 3.7037037037037036e-07, "loss": 0.1248, "num_tokens": 684187.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 281 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1376953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.24375465512275696, "kl": 4.708740234375, "learning_rate": 3.518518518518519e-07, "loss": 0.1883, "num_tokens": 686588.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 282 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.13818359375, "frac_reward_zero_std": 1.0, "grad_norm": 0.19137854874134064, "kl": 4.5213563442230225, "learning_rate": 3.3333333333333335e-07, "loss": 0.1809, "num_tokens": 689153.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 283 }, { "completion_length": 6.5, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.5, "completions/mean_terminated_length": 6.5, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.138671875, "frac_reward_zero_std": 1.0, "grad_norm": 4.8558197021484375, "kl": 3.394066274166107, "learning_rate": 3.148148148148148e-07, "loss": 0.1358, "num_tokens": 691439.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 284 }, { "completion_length": 8.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 8.25, "completions/mean_terminated_length": 8.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.13916015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.32645174860954285, "kl": 5.782284701243043, "learning_rate": 2.9629629629629634e-07, "loss": 0.2313, "num_tokens": 693904.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 285 }, { "completion_length": 9.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.5, "completions/mean_terminated_length": 9.5, "completions/min_length": 8.0, "completions/min_terminated_length": 8.0, "epoch": 0.1396484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.11963459104299545, "kl": 5.183718681335449, "learning_rate": 2.7777777777777776e-07, "loss": 0.2073, "num_tokens": 696366.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 286 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.14013671875, "frac_reward_zero_std": 1.0, "grad_norm": 0.4972098171710968, "kl": 4.34329429268837, "learning_rate": 2.592592592592593e-07, "loss": 0.1737, "num_tokens": 698688.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 287 }, { "completion_length": 9.25, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 9.25, "completions/mean_terminated_length": 9.25, "completions/min_length": 9.0, "completions/min_terminated_length": 9.0, "epoch": 0.140625, "frac_reward_zero_std": 1.0, "grad_norm": 0.10300704836845398, "kl": 7.026785612106323, "learning_rate": 2.4074074074074075e-07, "loss": 0.2811, "num_tokens": 701137.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 288 }, { "completion_length": 7.5, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.5, "completions/mean_terminated_length": 7.5, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.14111328125, "frac_reward_zero_std": 1.0, "grad_norm": 0.09733907133340836, "kl": 5.699796557426453, "learning_rate": 2.2222222222222224e-07, "loss": 0.228, "num_tokens": 703683.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 289 }, { "completion_length": 6.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.75, "completions/mean_terminated_length": 6.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1416015625, "frac_reward_zero_std": 1.0, "grad_norm": 0.392020046710968, "kl": 6.196896195411682, "learning_rate": 2.0370370370370374e-07, "loss": 0.2479, "num_tokens": 705978.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 290 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 10.0, "completions/max_terminated_length": 10.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.14208984375, "frac_reward_zero_std": 1.0, "grad_norm": 0.12707433104515076, "kl": 5.11537766456604, "learning_rate": 1.8518518518518518e-07, "loss": 0.2046, "num_tokens": 708373.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 291 }, { "completion_length": 6.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.25, "completions/mean_terminated_length": 6.25, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.142578125, "frac_reward_zero_std": 1.0, "grad_norm": 0.3045845329761505, "kl": 7.102207660675049, "learning_rate": 1.6666666666666668e-07, "loss": 0.2841, "num_tokens": 711054.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 292 }, { "completion_length": 7.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.75, "completions/mean_terminated_length": 7.75, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.14306640625, "frac_reward_zero_std": 1.0, "grad_norm": 0.11937805265188217, "kl": 5.417436480522156, "learning_rate": 1.4814814814814817e-07, "loss": 0.2167, "num_tokens": 713453.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 293 }, { "completion_length": 7.25, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.25, "completions/mean_terminated_length": 7.25, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.1435546875, "frac_reward_zero_std": 1.0, "grad_norm": 0.2303619086742401, "kl": 3.6886804699897766, "learning_rate": 1.2962962962962964e-07, "loss": 0.1475, "num_tokens": 715750.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 294 }, { "completion_length": 4.75, "completions/clipped_ratio": 0.0, "completions/max_length": 7.0, "completions/max_terminated_length": 7.0, "completions/mean_length": 4.75, "completions/mean_terminated_length": 4.75, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.14404296875, "frac_reward_zero_std": 1.0, "grad_norm": 0.3920138478279114, "kl": 3.8468642830848694, "learning_rate": 1.1111111111111112e-07, "loss": 0.1539, "num_tokens": 718181.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 295 }, { "completion_length": 8.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 8.0, "completions/mean_terminated_length": 8.0, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.14453125, "frac_reward_zero_std": 1.0, "grad_norm": 0.5313488841056824, "kl": 6.366184711456299, "learning_rate": 9.259259259259259e-08, "loss": 0.2546, "num_tokens": 720365.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 296 }, { "completion_length": 7.0, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 7.0, "completions/mean_terminated_length": 7.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.14501953125, "frac_reward_zero_std": 1.0, "grad_norm": 0.2511424720287323, "kl": 3.7885043919086456, "learning_rate": 7.407407407407409e-08, "loss": 0.1515, "num_tokens": 722869.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 297 }, { "completion_length": 4.0, "completions/clipped_ratio": 0.0, "completions/max_length": 4.0, "completions/max_terminated_length": 4.0, "completions/mean_length": 4.0, "completions/mean_terminated_length": 4.0, "completions/min_length": 4.0, "completions/min_terminated_length": 4.0, "epoch": 0.1455078125, "frac_reward_zero_std": 1.0, "grad_norm": 0.08439668267965317, "kl": 3.0557634830474854, "learning_rate": 5.555555555555556e-08, "loss": 0.1222, "num_tokens": 725309.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 298 }, { "completion_length": 5.0, "completions/clipped_ratio": 0.0, "completions/max_length": 5.0, "completions/max_terminated_length": 5.0, "completions/mean_length": 5.0, "completions/mean_terminated_length": 5.0, "completions/min_length": 5.0, "completions/min_terminated_length": 5.0, "epoch": 0.14599609375, "frac_reward_zero_std": 1.0, "grad_norm": 0.12893590331077576, "kl": 1.9960243701934814, "learning_rate": 3.703703703703704e-08, "loss": 0.0798, "num_tokens": 727741.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 299 }, { "completion_length": 6.75, "completions/clipped_ratio": 0.0, "completions/max_length": 9.0, "completions/max_terminated_length": 9.0, "completions/mean_length": 6.75, "completions/mean_terminated_length": 6.75, "completions/min_length": 6.0, "completions/min_terminated_length": 6.0, "epoch": 0.146484375, "frac_reward_zero_std": 1.0, "grad_norm": 0.568265974521637, "kl": 4.718587636947632, "learning_rate": 1.851851851851852e-08, "loss": 0.1887, "num_tokens": 729932.0, "reward": 5.5, "reward_std": 0.0, "rewards/showdown_reward_func/mean": 5.5, "rewards/showdown_reward_func/std": 0.0, "step": 300 } ], "logging_steps": 1, "max_steps": 300, "num_input_tokens_seen": 729932, "num_train_epochs": 1, "save_steps": 50, "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": 1, "trial_name": null, "trial_params": null }