diff --git "a/logs/train_log.jsonl" "b/logs/train_log.jsonl"
--- "a/logs/train_log.jsonl"
+++ "b/logs/train_log.jsonl"
@@ -21,3 +21,1042 @@
 {"update": 9, "steps_done": 18432, "mean_step_reward": 0.2662, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 354, "mean_ep_return": 1.545, "mean_ep_length": 5.8, "mean_collisions": 0.04, "mean_intent_rate": 0.256, "policy_loss": -0.021892, "value_loss": 4.490601, "entropy": 1.133845, "total_loss": 2.21207, "clip_frac": 0.118286, "sps": 28.5, "elapsed_s": 984.2}
 {"update": 10, "steps_done": 20480, "mean_step_reward": 0.2417, "min_step_reward": -10.5896, "max_step_reward": 2.5, "n_episodes_finished": 277, "mean_ep_return": 1.77, "mean_ep_length": 7.4, "mean_collisions": 0.08, "mean_intent_rate": 0.284, "policy_loss": -0.011659, "value_loss": 5.568001, "entropy": 1.090608, "total_loss": 2.761435, "clip_frac": 0.06311, "sps": 34.9, "elapsed_s": 1043.0}
 {"update": 11, "steps_done": 22528, "mean_step_reward": 0.3469, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 258, "mean_ep_return": 2.749, "mean_ep_length": 7.9, "mean_collisions": 0.05, "mean_intent_rate": 0.304, "policy_loss": -0.011639, "value_loss": 4.634015, "entropy": 1.023405, "total_loss": 2.295135, "clip_frac": 0.088135, "sps": 37.2, "elapsed_s": 1098.1}
+{"update": 12, "steps_done": 24576, "mean_step_reward": 0.3857, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 196, "mean_ep_return": 4.007, "mean_ep_length": 10.4, "mean_collisions": 0.03, "mean_intent_rate": 0.326, "policy_loss": -0.011827, "value_loss": 5.523948, "entropy": 1.009921, "total_loss": 2.740048, "clip_frac": 0.106323, "sps": 48.0, "elapsed_s": 1140.7}
+{"update": 13, "steps_done": 26624, "mean_step_reward": 0.3844, "min_step_reward": -10.5325, "max_step_reward": 2.5, "n_episodes_finished": 174, "mean_ep_return": 4.53, "mean_ep_length": 11.9, "mean_collisions": 0.03, "mean_intent_rate": 0.345, "policy_loss": -0.008322, "value_loss": 5.496592, "entropy": 0.958016, "total_loss": 2.730394, "clip_frac": 0.140137, "sps": 53.1, "elapsed_s": 1179.3}
+{"update": 14, "steps_done": 28672, "mean_step_reward": 0.4492, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 130, "mean_ep_return": 7.099, "mean_ep_length": 15.8, "mean_collisions": 0.06, "mean_intent_rate": 0.374, "policy_loss": -0.002666, "value_loss": 6.482279, "entropy": 0.874752, "total_loss": 3.229726, "clip_frac": 0.318115, "sps": 66.0, "elapsed_s": 1210.3}
+{"update": 15, "steps_done": 30720, "mean_step_reward": 0.4026, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 84, "mean_ep_return": 9.873, "mean_ep_length": 24.4, "mean_collisions": 0.11, "mean_intent_rate": 0.342, "policy_loss": 0.007893, "value_loss": 12.247667, "entropy": 0.824531, "total_loss": 6.123482, "clip_frac": 0.299194, "sps": 95.6, "elapsed_s": 1231.8}
+{"update": 16, "steps_done": 32768, "mean_step_reward": 0.3782, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 62, "mean_ep_return": 12.388, "mean_ep_length": 33.0, "mean_collisions": 0.13, "mean_intent_rate": 0.336, "policy_loss": -0.0051, "value_loss": 10.040691, "entropy": 0.765414, "total_loss": 5.007591, "clip_frac": 0.054199, "sps": 117.6, "elapsed_s": 1249.2}
+{"update": 17, "steps_done": 34816, "mean_step_reward": 0.3308, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 69, "mean_ep_return": 9.619, "mean_ep_length": 29.5, "mean_collisions": 0.13, "mean_intent_rate": 0.329, "policy_loss": -0.00466, "value_loss": 11.27416, "entropy": 0.711745, "total_loss": 5.625303, "clip_frac": 0.074219, "sps": 113.9, "elapsed_s": 1267.2}
+{"update": 18, "steps_done": 36864, "mean_step_reward": 0.3387, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 41, "mean_ep_return": 17.109, "mean_ep_length": 49.1, "mean_collisions": 0.12, "mean_intent_rate": 0.319, "policy_loss": -0.003621, "value_loss": 11.207777, "entropy": 0.691627, "total_loss": 5.593351, "clip_frac": 0.043091, "sps": 155.0, "elapsed_s": 1280.4}
+{"update": 19, "steps_done": 38912, "mean_step_reward": 0.355, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 49, "mean_ep_return": 14.972, "mean_ep_length": 42.1, "mean_collisions": 0.12, "mean_intent_rate": 0.317, "policy_loss": -0.003374, "value_loss": 10.967927, "entropy": 0.66069, "total_loss": 5.473983, "clip_frac": 0.062988, "sps": 138.9, "elapsed_s": 1295.1}
+{"update": 20, "steps_done": 40960, "mean_step_reward": 0.3722, "min_step_reward": -10.5807, "max_step_reward": 2.5, "n_episodes_finished": 38, "mean_ep_return": 19.363, "mean_ep_length": 52.8, "mean_collisions": 0.18, "mean_intent_rate": 0.314, "policy_loss": -0.000955, "value_loss": 14.071788, "entropy": 0.703032, "total_loss": 7.027909, "clip_frac": 0.078735, "sps": 165.8, "elapsed_s": 1307.5}
+{"update": 1, "steps_done": 2048, "mean_step_reward": -0.4067, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 380, "mean_ep_return": -2.196, "mean_ep_length": 5.4, "mean_collisions": 0.26, "mean_intent_rate": 0.142, "policy_loss": -0.01167, "value_loss": 19.715629, "entropy": 1.605821, "total_loss": 9.830086, "clip_frac": 0.054199, "sps": 26.8, "elapsed_s": 76.3}
+{"update": 2, "steps_done": 4096, "mean_step_reward": -0.2433, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 473, "mean_ep_return": -1.051, "mean_ep_length": 4.3, "mean_collisions": 0.12, "mean_intent_rate": 0.15, "policy_loss": -0.020159, "value_loss": 8.248246, "entropy": 1.554832, "total_loss": 4.088416, "clip_frac": 0.326172, "sps": 20.8, "elapsed_s": 174.6}
+{"update": 3, "steps_done": 6144, "mean_step_reward": -0.2488, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 638, "mean_ep_return": -0.798, "mean_ep_length": 3.2, "mean_collisions": 0.08, "mean_intent_rate": 0.122, "policy_loss": -0.019122, "value_loss": 5.589961, "entropy": 1.48965, "total_loss": 2.760962, "clip_frac": 0.25708, "sps": 16.0, "elapsed_s": 302.3}
+{"update": 4, "steps_done": 8192, "mean_step_reward": -0.2106, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 710, "mean_ep_return": -0.607, "mean_ep_length": 2.9, "mean_collisions": 0.07, "mean_intent_rate": 0.128, "policy_loss": -0.020921, "value_loss": 4.639985, "entropy": 1.418522, "total_loss": 2.284887, "clip_frac": 0.197876, "sps": 14.4, "elapsed_s": 444.4}
+{"update": 1, "steps_done": 2048, "mean_step_reward": -0.4067, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 380, "mean_ep_return": -2.196, "mean_ep_length": 5.4, "mean_collisions": 0.26, "mean_intent_rate": 0.142, "policy_loss": -0.01167, "value_loss": 19.715627, "entropy": 1.605821, "total_loss": 9.830085, "clip_frac": 0.054199, "sps": 27.0, "elapsed_s": 75.7}
+{"update": 5, "steps_done": 10240, "mean_step_reward": -0.0937, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 747, "mean_ep_return": -0.26, "mean_ep_length": 2.7, "mean_collisions": 0.04, "mean_intent_rate": 0.137, "policy_loss": -0.020439, "value_loss": 2.785062, "entropy": 1.40261, "total_loss": 1.358066, "clip_frac": 0.165405, "sps": 13.4, "elapsed_s": 596.7}
+{"update": 6, "steps_done": 12288, "mean_step_reward": 0.0216, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 624, "mean_ep_return": 0.073, "mean_ep_length": 3.3, "mean_collisions": 0.04, "mean_intent_rate": 0.173, "policy_loss": -0.021561, "value_loss": 3.313721, "entropy": 1.358029, "total_loss": 1.62172, "clip_frac": 0.142944, "sps": 15.8, "elapsed_s": 726.1}
+{"update": 1, "steps_done": 2048, "mean_step_reward": -0.4067, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 380, "mean_ep_return": -2.196, "mean_ep_length": 5.4, "mean_collisions": 0.26, "mean_intent_rate": 0.142, "policy_loss": -0.01167, "value_loss": 19.715641, "entropy": 1.605821, "total_loss": 9.830092, "clip_frac": 0.054199, "sps": 340.2, "elapsed_s": 6.0}
+{"update": 2, "steps_done": 4096, "mean_step_reward": -0.2433, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 473, "mean_ep_return": -1.051, "mean_ep_length": 4.3, "mean_collisions": 0.12, "mean_intent_rate": 0.15, "policy_loss": -0.020157, "value_loss": 8.248294, "entropy": 1.554779, "total_loss": 4.088442, "clip_frac": 0.327515, "sps": 381.9, "elapsed_s": 11.4}
+{"update": 3, "steps_done": 6144, "mean_step_reward": -0.2488, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 638, "mean_ep_return": -0.798, "mean_ep_length": 3.2, "mean_collisions": 0.08, "mean_intent_rate": 0.122, "policy_loss": -0.019061, "value_loss": 5.590029, "entropy": 1.489708, "total_loss": 2.761056, "clip_frac": 0.260254, "sps": 369.1, "elapsed_s": 16.9}
+{"update": 4, "steps_done": 8192, "mean_step_reward": -0.2273, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 710, "mean_ep_return": -0.656, "mean_ep_length": 2.9, "mean_collisions": 0.07, "mean_intent_rate": 0.127, "policy_loss": -0.021522, "value_loss": 4.8382, "entropy": 1.418572, "total_loss": 2.383392, "clip_frac": 0.241089, "sps": 357.0, "elapsed_s": 22.7}
+{"update": 5, "steps_done": 10240, "mean_step_reward": -0.0836, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 762, "mean_ep_return": -0.224, "mean_ep_length": 2.7, "mean_collisions": 0.03, "mean_intent_rate": 0.129, "policy_loss": -0.015987, "value_loss": 3.646871, "entropy": 1.390213, "total_loss": 1.793547, "clip_frac": 0.153076, "sps": 357.9, "elapsed_s": 28.4}
+{"update": 6, "steps_done": 12288, "mean_step_reward": 0.0195, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 625, "mean_ep_return": 0.056, "mean_ep_length": 3.3, "mean_collisions": 0.04, "mean_intent_rate": 0.168, "policy_loss": -0.022837, "value_loss": 2.953562, "entropy": 1.351083, "total_loss": 1.440433, "clip_frac": 0.169189, "sps": 365.3, "elapsed_s": 34.0}
+{"update": 7, "steps_done": 14336, "mean_step_reward": 0.1334, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 558, "mean_ep_return": 0.491, "mean_ep_length": 3.7, "mean_collisions": 0.03, "mean_intent_rate": 0.182, "policy_loss": -0.036337, "value_loss": 3.615671, "entropy": 1.26046, "total_loss": 1.758893, "clip_frac": 0.404541, "sps": 371.0, "elapsed_s": 39.5}
+{"update": 8, "steps_done": 16384, "mean_step_reward": 0.2251, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 426, "mean_ep_return": 1.092, "mean_ep_length": 4.8, "mean_collisions": 0.03, "mean_intent_rate": 0.228, "policy_loss": -0.026158, "value_loss": 4.251343, "entropy": 1.179673, "total_loss": 2.087717, "clip_frac": 0.206299, "sps": 379.0, "elapsed_s": 44.9}
+{"update": 9, "steps_done": 18432, "mean_step_reward": 0.2512, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 330, "mean_ep_return": 1.545, "mean_ep_length": 6.2, "mean_collisions": 0.05, "mean_intent_rate": 0.261, "policy_loss": -0.017344, "value_loss": 5.061886, "entropy": 1.136107, "total_loss": 2.502238, "clip_frac": 0.107788, "sps": 387.3, "elapsed_s": 50.2}
+{"update": 10, "steps_done": 20480, "mean_step_reward": 0.3336, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 314, "mean_ep_return": 2.173, "mean_ep_length": 6.5, "mean_collisions": 0.04, "mean_intent_rate": 0.273, "policy_loss": -0.015101, "value_loss": 5.112308, "entropy": 1.041114, "total_loss": 2.530642, "clip_frac": 0.088257, "sps": 388.3, "elapsed_s": 55.5}
+{"update": 11, "steps_done": 22528, "mean_step_reward": 0.3972, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 305, "mean_ep_return": 2.69, "mean_ep_length": 6.7, "mean_collisions": 0.05, "mean_intent_rate": 0.309, "policy_loss": -0.014369, "value_loss": 4.646124, "entropy": 1.007504, "total_loss": 2.298618, "clip_frac": 0.135498, "sps": 380.1, "elapsed_s": 60.9}
+{"update": 12, "steps_done": 24576, "mean_step_reward": 0.4515, "min_step_reward": -10.5816, "max_step_reward": 2.5, "n_episodes_finished": 266, "mean_ep_return": 3.443, "mean_ep_length": 7.7, "mean_collisions": 0.03, "mean_intent_rate": 0.328, "policy_loss": -0.014674, "value_loss": 5.178558, "entropy": 1.011849, "total_loss": 2.564487, "clip_frac": 0.187866, "sps": 390.9, "elapsed_s": 66.1}
+{"update": 13, "steps_done": 26624, "mean_step_reward": 0.4629, "min_step_reward": -10.5398, "max_step_reward": 2.5, "n_episodes_finished": 208, "mean_ep_return": 4.582, "mean_ep_length": 9.8, "mean_collisions": 0.02, "mean_intent_rate": 0.348, "policy_loss": -0.016327, "value_loss": 4.595904, "entropy": 0.939062, "total_loss": 2.272234, "clip_frac": 0.115723, "sps": 396.9, "elapsed_s": 71.3}
+{"update": 14, "steps_done": 28672, "mean_step_reward": 0.4419, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 122, "mean_ep_return": 7.42, "mean_ep_length": 16.8, "mean_collisions": 0.04, "mean_intent_rate": 0.36, "policy_loss": -0.006554, "value_loss": 7.377189, "entropy": 0.890054, "total_loss": 3.67314, "clip_frac": 0.153442, "sps": 405.6, "elapsed_s": 76.3}
+{"update": 15, "steps_done": 30720, "mean_step_reward": 0.3313, "min_step_reward": -7.5867, "max_step_reward": 2.5, "n_episodes_finished": 86, "mean_ep_return": 7.78, "mean_ep_length": 23.6, "mean_collisions": 0.05, "mean_intent_rate": 0.301, "policy_loss": -0.00636, "value_loss": 7.827714, "entropy": 0.886759, "total_loss": 3.898629, "clip_frac": 0.110107, "sps": 408.9, "elapsed_s": 81.3}
+{"update": 16, "steps_done": 32768, "mean_step_reward": 0.3632, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 60, "mean_ep_return": 12.48, "mean_ep_length": 34.4, "mean_collisions": 0.2, "mean_intent_rate": 0.329, "policy_loss": -0.000237, "value_loss": 11.452, "entropy": 0.827669, "total_loss": 5.717486, "clip_frac": 0.158203, "sps": 408.9, "elapsed_s": 86.3}
+{"update": 17, "steps_done": 34816, "mean_step_reward": 0.3432, "min_step_reward": -10.5881, "max_step_reward": 2.5, "n_episodes_finished": 66, "mean_ep_return": 10.63, "mean_ep_length": 30.8, "mean_collisions": 0.23, "mean_intent_rate": 0.343, "policy_loss": -0.001029, "value_loss": 10.844101, "entropy": 0.749451, "total_loss": 5.413528, "clip_frac": 0.144409, "sps": 413.1, "elapsed_s": 91.3}
+{"update": 18, "steps_done": 36864, "mean_step_reward": 0.3011, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 42, "mean_ep_return": 14.815, "mean_ep_length": 49.0, "mean_collisions": 0.31, "mean_intent_rate": 0.303, "policy_loss": 0.000763, "value_loss": 10.393293, "entropy": 0.714917, "total_loss": 5.190261, "clip_frac": 0.104126, "sps": 413.1, "elapsed_s": 96.3}
+{"update": 19, "steps_done": 38912, "mean_step_reward": 0.3851, "min_step_reward": -10.5347, "max_step_reward": 2.5, "n_episodes_finished": 38, "mean_ep_return": 20.354, "mean_ep_length": 53.6, "mean_collisions": 0.29, "mean_intent_rate": 0.329, "policy_loss": -0.003396, "value_loss": 13.677117, "entropy": 0.647151, "total_loss": 6.828691, "clip_frac": 0.08374, "sps": 413.5, "elapsed_s": 101.2}
+{"update": 20, "steps_done": 40960, "mean_step_reward": 0.3824, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 38, "mean_ep_return": 20.72, "mean_ep_length": 53.7, "mean_collisions": 0.18, "mean_intent_rate": 0.328, "policy_loss": -0.00324, "value_loss": 10.67055, "entropy": 0.658629, "total_loss": 5.325449, "clip_frac": 0.067139, "sps": 401.8, "elapsed_s": 106.3}
+{"update": 21, "steps_done": 43008, "mean_step_reward": 0.3362, "min_step_reward": -10.5614, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 19.457, "mean_ep_length": 57.5, "mean_collisions": 0.2, "mean_intent_rate": 0.304, "policy_loss": 0.003857, "value_loss": 13.721792, "entropy": 0.670832, "total_loss": 6.858045, "clip_frac": 0.130127, "sps": 415.6, "elapsed_s": 111.2}
+{"update": 22, "steps_done": 45056, "mean_step_reward": 0.3749, "min_step_reward": -10.5835, "max_step_reward": 2.5, "n_episodes_finished": 38, "mean_ep_return": 20.242, "mean_ep_length": 54.6, "mean_collisions": 0.21, "mean_intent_rate": 0.318, "policy_loss": -0.001937, "value_loss": 13.654996, "entropy": 0.675121, "total_loss": 6.81881, "clip_frac": 0.148926, "sps": 413.5, "elapsed_s": 116.2}
+{"update": 23, "steps_done": 47104, "mean_step_reward": 0.3955, "min_step_reward": -7.5826, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 22.209, "mean_ep_length": 56.1, "mean_collisions": 0.14, "mean_intent_rate": 0.321, "policy_loss": -0.003687, "value_loss": 13.792976, "entropy": 0.673168, "total_loss": 6.88607, "clip_frac": 0.084961, "sps": 413.4, "elapsed_s": 121.2}
+{"update": 24, "steps_done": 49152, "mean_step_reward": 0.3571, "min_step_reward": -10.5831, "max_step_reward": 2.5, "n_episodes_finished": 39, "mean_ep_return": 18.364, "mean_ep_length": 51.4, "mean_collisions": 0.26, "mean_intent_rate": 0.32, "policy_loss": 0.000668, "value_loss": 15.257973, "entropy": 0.67543, "total_loss": 7.6229, "clip_frac": 0.028931, "sps": 418.1, "elapsed_s": 126.1}
+{"update": 25, "steps_done": 51200, "mean_step_reward": 0.3544, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 20.576, "mean_ep_length": 55.9, "mean_collisions": 0.05, "mean_intent_rate": 0.297, "policy_loss": -0.000769, "value_loss": 16.811141, "entropy": 0.673909, "total_loss": 8.398063, "clip_frac": 0.03479, "sps": 412.7, "elapsed_s": 131.0}
+{"update": 26, "steps_done": 53248, "mean_step_reward": 0.4181, "min_step_reward": -10.5701, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 24.478, "mean_ep_length": 60.2, "mean_collisions": 0.29, "mean_intent_rate": 0.329, "policy_loss": -0.000674, "value_loss": 14.820069, "entropy": 0.664578, "total_loss": 7.402715, "clip_frac": 0.016113, "sps": 412.3, "elapsed_s": 136.0}
+{"update": 27, "steps_done": 55296, "mean_step_reward": 0.3986, "min_step_reward": -10.5786, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 22.591, "mean_ep_length": 57.1, "mean_collisions": 0.25, "mean_intent_rate": 0.315, "policy_loss": 0.00107, "value_loss": 17.958569, "entropy": 0.686126, "total_loss": 8.973494, "clip_frac": 0.065186, "sps": 416.2, "elapsed_s": 140.9}
+{"update": 28, "steps_done": 57344, "mean_step_reward": 0.4156, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 25.431, "mean_ep_length": 59.4, "mean_collisions": 0.35, "mean_intent_rate": 0.34, "policy_loss": -0.001626, "value_loss": 15.668549, "entropy": 0.689446, "total_loss": 7.825754, "clip_frac": 0.057251, "sps": 411.3, "elapsed_s": 145.9}
+{"update": 29, "steps_done": 59392, "mean_step_reward": 0.3432, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 20.116, "mean_ep_length": 59.9, "mean_collisions": 0.77, "mean_intent_rate": 0.33, "policy_loss": -0.00111, "value_loss": 22.003734, "entropy": 0.686022, "total_loss": 10.993897, "clip_frac": 0.039673, "sps": 413.6, "elapsed_s": 150.8}
+{"update": 30, "steps_done": 61440, "mean_step_reward": 0.4203, "min_step_reward": -10.5878, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 27.513, "mean_ep_length": 65.6, "mean_collisions": 0.61, "mean_intent_rate": 0.346, "policy_loss": 0.005387, "value_loss": 22.624646, "entropy": 0.646853, "total_loss": 11.311241, "clip_frac": 0.120605, "sps": 410.3, "elapsed_s": 155.8}
+{"update": 31, "steps_done": 63488, "mean_step_reward": 0.397, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 25.533, "mean_ep_length": 64.5, "mean_collisions": 0.41, "mean_intent_rate": 0.325, "policy_loss": 0.000327, "value_loss": 20.989354, "entropy": 0.617524, "total_loss": 10.488828, "clip_frac": 0.044189, "sps": 411.6, "elapsed_s": 160.8}
+{"update": 32, "steps_done": 65536, "mean_step_reward": 0.3878, "min_step_reward": -10.5832, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 23.062, "mean_ep_length": 59.6, "mean_collisions": 0.5, "mean_intent_rate": 0.33, "policy_loss": 0.001373, "value_loss": 19.037031, "entropy": 0.608681, "total_loss": 9.513802, "clip_frac": 0.046143, "sps": 410.3, "elapsed_s": 165.8}
+{"update": 33, "steps_done": 67584, "mean_step_reward": 0.4384, "min_step_reward": -10.549, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 29.352, "mean_ep_length": 67.4, "mean_collisions": 0.43, "mean_intent_rate": 0.346, "policy_loss": -0.000894, "value_loss": 20.511685, "entropy": 0.548177, "total_loss": 10.249466, "clip_frac": 0.024902, "sps": 416.2, "elapsed_s": 170.7}
+{"update": 34, "steps_done": 69632, "mean_step_reward": 0.3984, "min_step_reward": -10.5842, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 23.769, "mean_ep_length": 60.4, "mean_collisions": 0.44, "mean_intent_rate": 0.33, "policy_loss": -0.001126, "value_loss": 16.919359, "entropy": 0.581683, "total_loss": 8.452737, "clip_frac": 0.0448, "sps": 407.9, "elapsed_s": 175.7}
+{"update": 35, "steps_done": 71680, "mean_step_reward": 0.3676, "min_step_reward": -7.5883, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 21.398, "mean_ep_length": 56.6, "mean_collisions": 0.11, "mean_intent_rate": 0.323, "policy_loss": 0.002786, "value_loss": 16.158855, "entropy": 0.575765, "total_loss": 8.076456, "clip_frac": 0.105957, "sps": 409.4, "elapsed_s": 180.7}
+{"update": 36, "steps_done": 73728, "mean_step_reward": 0.3873, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 23.677, "mean_ep_length": 62.1, "mean_collisions": 0.27, "mean_intent_rate": 0.315, "policy_loss": -0.000427, "value_loss": 14.975034, "entropy": 0.631386, "total_loss": 7.480776, "clip_frac": 0.03833, "sps": 406.4, "elapsed_s": 185.8}
+{"update": 37, "steps_done": 75776, "mean_step_reward": 0.4485, "min_step_reward": -7.5847, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 26.963, "mean_ep_length": 59.7, "mean_collisions": 0.2, "mean_intent_rate": 0.339, "policy_loss": -0.002501, "value_loss": 17.315308, "entropy": 0.641384, "total_loss": 8.648739, "clip_frac": 0.04248, "sps": 401.5, "elapsed_s": 190.9}
+{"update": 38, "steps_done": 77824, "mean_step_reward": 0.415, "min_step_reward": -10.5128, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 22.7, "mean_ep_length": 53.9, "mean_collisions": 0.46, "mean_intent_rate": 0.34, "policy_loss": 0.000433, "value_loss": 20.781456, "entropy": 0.64283, "total_loss": 10.384732, "clip_frac": 0.052246, "sps": 411.2, "elapsed_s": 195.9}
+{"update": 39, "steps_done": 79872, "mean_step_reward": 0.4663, "min_step_reward": -10.5794, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 27.09, "mean_ep_length": 59.9, "mean_collisions": 0.23, "mean_intent_rate": 0.333, "policy_loss": -0.001752, "value_loss": 20.738587, "entropy": 0.698733, "total_loss": 10.360555, "clip_frac": 0.042603, "sps": 421.3, "elapsed_s": 200.7}
+{"update": 40, "steps_done": 81920, "mean_step_reward": 0.4742, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 28.679, "mean_ep_length": 60.1, "mean_collisions": 0.56, "mean_intent_rate": 0.37, "policy_loss": -0.000395, "value_loss": 21.812208, "entropy": 0.697686, "total_loss": 10.898732, "clip_frac": 0.049316, "sps": 414.2, "elapsed_s": 205.7}
+{"update": 41, "steps_done": 83968, "mean_step_reward": 0.3525, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 20.109, "mean_ep_length": 57.2, "mean_collisions": 1.14, "mean_intent_rate": 0.358, "policy_loss": 0.000117, "value_loss": 25.182746, "entropy": 0.694234, "total_loss": 12.584547, "clip_frac": 0.040283, "sps": 409.5, "elapsed_s": 210.7}
+{"update": 42, "steps_done": 86016, "mean_step_reward": 0.3582, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 21.569, "mean_ep_length": 59.8, "mean_collisions": 0.68, "mean_intent_rate": 0.336, "policy_loss": 0.002189, "value_loss": 18.52566, "entropy": 0.693856, "total_loss": 9.258081, "clip_frac": 0.06897, "sps": 414.2, "elapsed_s": 215.6}
+{"update": 43, "steps_done": 88064, "mean_step_reward": 0.3851, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 22.991, "mean_ep_length": 59.0, "mean_collisions": 1.17, "mean_intent_rate": 0.373, "policy_loss": 0.000741, "value_loss": 21.84424, "entropy": 0.70766, "total_loss": 10.915785, "clip_frac": 0.045776, "sps": 420.3, "elapsed_s": 220.5}
+{"update": 44, "steps_done": 90112, "mean_step_reward": 0.4302, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 23.794, "mean_ep_length": 55.4, "mean_collisions": 0.83, "mean_intent_rate": 0.369, "policy_loss": 0.00033, "value_loss": 19.338179, "entropy": 0.70787, "total_loss": 9.662341, "clip_frac": 0.029907, "sps": 416.6, "elapsed_s": 225.4}
+{"update": 45, "steps_done": 92160, "mean_step_reward": 0.4854, "min_step_reward": -10.5858, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 28.662, "mean_ep_length": 59.4, "mean_collisions": 0.37, "mean_intent_rate": 0.355, "policy_loss": -0.000925, "value_loss": 21.513402, "entropy": 0.744407, "total_loss": 10.748332, "clip_frac": 0.073975, "sps": 406.4, "elapsed_s": 230.5}
+{"update": 46, "steps_done": 94208, "mean_step_reward": 0.4716, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 28.061, "mean_ep_length": 59.5, "mean_collisions": 0.77, "mean_intent_rate": 0.38, "policy_loss": -0.00212, "value_loss": 30.143577, "entropy": 0.726648, "total_loss": 15.062402, "clip_frac": 0.026855, "sps": 411.6, "elapsed_s": 235.4}
+{"update": 47, "steps_done": 96256, "mean_step_reward": 0.4374, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 25.965, "mean_ep_length": 60.7, "mean_collisions": 1.12, "mean_intent_rate": 0.384, "policy_loss": -0.003325, "value_loss": 28.213755, "entropy": 0.785435, "total_loss": 14.095699, "clip_frac": 0.039551, "sps": 412.2, "elapsed_s": 240.4}
+{"update": 1, "steps_done": 2048, "mean_step_reward": -0.4067, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 380, "mean_ep_return": -2.196, "mean_ep_length": 5.4, "mean_collisions": 0.26, "mean_intent_rate": 0.142, "policy_loss": -0.01167, "value_loss": 19.715643, "entropy": 1.60582, "total_loss": 9.830093, "clip_frac": 0.054199, "sps": 345.4, "elapsed_s": 5.9}
+{"update": 2, "steps_done": 4096, "mean_step_reward": -0.2433, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 473, "mean_ep_return": -1.051, "mean_ep_length": 4.3, "mean_collisions": 0.12, "mean_intent_rate": 0.15, "policy_loss": -0.020157, "value_loss": 8.248289, "entropy": 1.554816, "total_loss": 4.08844, "clip_frac": 0.326416, "sps": 383.5, "elapsed_s": 11.3}
+{"update": 3, "steps_done": 6144, "mean_step_reward": -0.2488, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 638, "mean_ep_return": -0.798, "mean_ep_length": 3.2, "mean_collisions": 0.08, "mean_intent_rate": 0.122, "policy_loss": -0.019128, "value_loss": 5.589981, "entropy": 1.489499, "total_loss": 2.760968, "clip_frac": 0.257202, "sps": 370.3, "elapsed_s": 16.8}
+{"update": 4, "steps_done": 8192, "mean_step_reward": -0.2169, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 710, "mean_ep_return": -0.626, "mean_ep_length": 2.9, "mean_collisions": 0.07, "mean_intent_rate": 0.128, "policy_loss": -0.022311, "value_loss": 4.769057, "entropy": 1.421401, "total_loss": 2.348004, "clip_frac": 0.174438, "sps": 359.5, "elapsed_s": 22.5}
+{"update": 5, "steps_done": 10240, "mean_step_reward": -0.0809, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 762, "mean_ep_return": -0.217, "mean_ep_length": 2.7, "mean_collisions": 0.03, "mean_intent_rate": 0.132, "policy_loss": -0.023377, "value_loss": 3.373912, "entropy": 1.416993, "total_loss": 1.649409, "clip_frac": 0.168823, "sps": 361.9, "elapsed_s": 28.2}
+{"update": 6, "steps_done": 12288, "mean_step_reward": 0.0102, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 608, "mean_ep_return": 0.031, "mean_ep_length": 3.4, "mean_collisions": 0.05, "mean_intent_rate": 0.172, "policy_loss": -0.023403, "value_loss": 4.467767, "entropy": 1.339986, "total_loss": 2.197081, "clip_frac": 0.213013, "sps": 370.6, "elapsed_s": 33.7}
+{"update": 7, "steps_done": 14336, "mean_step_reward": 0.0859, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 603, "mean_ep_return": 0.298, "mean_ep_length": 3.4, "mean_collisions": 0.03, "mean_intent_rate": 0.178, "policy_loss": -0.033986, "value_loss": 3.133823, "entropy": 1.3039, "total_loss": 1.519887, "clip_frac": 0.330566, "sps": 371.5, "elapsed_s": 39.2}
+{"update": 8, "steps_done": 16384, "mean_step_reward": 0.2357, "min_step_reward": -10.589, "max_step_reward": 2.5, "n_episodes_finished": 468, "mean_ep_return": 1.027, "mean_ep_length": 4.4, "mean_collisions": 0.03, "mean_intent_rate": 0.226, "policy_loss": -0.028106, "value_loss": 4.194826, "entropy": 1.245821, "total_loss": 2.056849, "clip_frac": 0.221558, "sps": 380.2, "elapsed_s": 44.6}
+{"update": 9, "steps_done": 18432, "mean_step_reward": 0.2333, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 340, "mean_ep_return": 1.384, "mean_ep_length": 6.0, "mean_collisions": 0.03, "mean_intent_rate": 0.251, "policy_loss": -0.020351, "value_loss": 3.979134, "entropy": 1.189862, "total_loss": 1.957317, "clip_frac": 0.11731, "sps": 389.0, "elapsed_s": 49.9}
+{"update": 10, "steps_done": 20480, "mean_step_reward": 0.3722, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 269, "mean_ep_return": 2.851, "mean_ep_length": 7.6, "mean_collisions": 0.04, "mean_intent_rate": 0.308, "policy_loss": -0.010454, "value_loss": 5.667221, "entropy": 1.133304, "total_loss": 2.811824, "clip_frac": 0.087036, "sps": 393.8, "elapsed_s": 55.1}
+{"update": 11, "steps_done": 22528, "mean_step_reward": 0.4247, "min_step_reward": -10.5902, "max_step_reward": 2.5, "n_episodes_finished": 261, "mean_ep_return": 3.327, "mean_ep_length": 7.8, "mean_collisions": 0.02, "mean_intent_rate": 0.307, "policy_loss": -0.014945, "value_loss": 3.978551, "entropy": 1.056373, "total_loss": 1.963767, "clip_frac": 0.137573, "sps": 384.8, "elapsed_s": 60.4}
+{"update": 12, "steps_done": 24576, "mean_step_reward": 0.4042, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 204, "mean_ep_return": 4.067, "mean_ep_length": 10.0, "mean_collisions": 0.07, "mean_intent_rate": 0.351, "policy_loss": -0.007418, "value_loss": 6.124565, "entropy": 1.048312, "total_loss": 3.044381, "clip_frac": 0.124023, "sps": 396.9, "elapsed_s": 65.5}
+{"update": 13, "steps_done": 26624, "mean_step_reward": 0.3855, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 153, "mean_ep_return": 5.106, "mean_ep_length": 13.6, "mean_collisions": 0.13, "mean_intent_rate": 0.354, "policy_loss": -5.1e-05, "value_loss": 9.597437, "entropy": 1.015955, "total_loss": 4.788508, "clip_frac": 0.257446, "sps": 404.6, "elapsed_s": 70.6}
+{"update": 14, "steps_done": 28672, "mean_step_reward": 0.3836, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 117, "mean_ep_return": 6.766, "mean_ep_length": 17.6, "mean_collisions": 0.14, "mean_intent_rate": 0.361, "policy_loss": -0.008587, "value_loss": 7.013787, "entropy": 0.919003, "total_loss": 3.489117, "clip_frac": 0.188354, "sps": 401.3, "elapsed_s": 75.7}
+{"update": 15, "steps_done": 30720, "mean_step_reward": 0.2961, "min_step_reward": -10.5468, "max_step_reward": 2.5, "n_episodes_finished": 71, "mean_ep_return": 8.537, "mean_ep_length": 28.2, "mean_collisions": 0.2, "mean_intent_rate": 0.327, "policy_loss": 0.005326, "value_loss": 8.886476, "entropy": 0.905661, "total_loss": 4.439507, "clip_frac": 0.239502, "sps": 411.9, "elapsed_s": 80.7}
+{"update": 16, "steps_done": 32768, "mean_step_reward": 0.3292, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 51, "mean_ep_return": 13.2, "mean_ep_length": 41.0, "mean_collisions": 0.31, "mean_intent_rate": 0.316, "policy_loss": -0.001746, "value_loss": 10.881723, "entropy": 0.842538, "total_loss": 5.43069, "clip_frac": 0.103882, "sps": 411.9, "elapsed_s": 85.7}
+{"update": 17, "steps_done": 34816, "mean_step_reward": 0.3517, "min_step_reward": -10.5728, "max_step_reward": 2.5, "n_episodes_finished": 62, "mean_ep_return": 11.574, "mean_ep_length": 32.8, "mean_collisions": 0.29, "mean_intent_rate": 0.339, "policy_loss": 0.004671, "value_loss": 14.042823, "entropy": 0.801522, "total_loss": 7.018067, "clip_frac": 0.247681, "sps": 413.4, "elapsed_s": 90.6}
+{"update": 18, "steps_done": 36864, "mean_step_reward": 0.2971, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 42, "mean_ep_return": 14.533, "mean_ep_length": 49.0, "mean_collisions": 0.45, "mean_intent_rate": 0.299, "policy_loss": 0.004976, "value_loss": 12.521394, "entropy": 0.784038, "total_loss": 6.257832, "clip_frac": 0.148804, "sps": 412.5, "elapsed_s": 95.6}
+{"update": 19, "steps_done": 38912, "mean_step_reward": 0.401, "min_step_reward": -10.525, "max_step_reward": 2.5, "n_episodes_finished": 45, "mean_ep_return": 18.346, "mean_ep_length": 45.7, "mean_collisions": 0.31, "mean_intent_rate": 0.355, "policy_loss": -0.002646, "value_loss": 12.216675, "entropy": 0.722279, "total_loss": 6.098469, "clip_frac": 0.093262, "sps": 416.5, "elapsed_s": 100.5}
+{"update": 20, "steps_done": 40960, "mean_step_reward": 0.2555, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 13.912, "mean_ep_length": 54.5, "mean_collisions": 0.57, "mean_intent_rate": 0.29, "policy_loss": -0.001867, "value_loss": 14.678293, "entropy": 0.723899, "total_loss": 7.330041, "clip_frac": 0.059082, "sps": 403.6, "elapsed_s": 105.6}
+{"update": 21, "steps_done": 43008, "mean_step_reward": 0.3761, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 40, "mean_ep_return": 19.586, "mean_ep_length": 51.9, "mean_collisions": 0.28, "mean_intent_rate": 0.321, "policy_loss": 0.001468, "value_loss": 14.818502, "entropy": 0.664827, "total_loss": 7.404071, "clip_frac": 0.096924, "sps": 412.2, "elapsed_s": 110.5}
+{"update": 22, "steps_done": 45056, "mean_step_reward": 0.3989, "min_step_reward": -10.5479, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 24.379, "mean_ep_length": 62.2, "mean_collisions": 0.19, "mean_intent_rate": 0.32, "policy_loss": -0.001123, "value_loss": 12.6571, "entropy": 0.65361, "total_loss": 6.320891, "clip_frac": 0.071533, "sps": 415.4, "elapsed_s": 115.5}
+{"update": 23, "steps_done": 47104, "mean_step_reward": 0.4373, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 26.15, "mean_ep_length": 59.7, "mean_collisions": 0.31, "mean_intent_rate": 0.352, "policy_loss": 0.000928, "value_loss": 16.871116, "entropy": 0.624131, "total_loss": 8.430244, "clip_frac": 0.032104, "sps": 415.3, "elapsed_s": 120.4}
+{"update": 24, "steps_done": 49152, "mean_step_reward": 0.4259, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 26.596, "mean_ep_length": 61.8, "mean_collisions": 0.3, "mean_intent_rate": 0.328, "policy_loss": 2.4e-05, "value_loss": 17.843995, "entropy": 0.664311, "total_loss": 8.915378, "clip_frac": 0.044067, "sps": 409.3, "elapsed_s": 125.4}
+{"update": 25, "steps_done": 51200, "mean_step_reward": 0.4157, "min_step_reward": -10.5818, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 26.044, "mean_ep_length": 62.6, "mean_collisions": 0.36, "mean_intent_rate": 0.329, "policy_loss": 0.002534, "value_loss": 18.15192, "entropy": 0.674321, "total_loss": 9.071751, "clip_frac": 0.098267, "sps": 416.2, "elapsed_s": 130.3}
+{"update": 26, "steps_done": 53248, "mean_step_reward": 0.3946, "min_step_reward": -10.5592, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 22.575, "mean_ep_length": 57.9, "mean_collisions": 0.54, "mean_intent_rate": 0.336, "policy_loss": 0.001435, "value_loss": 22.069994, "entropy": 0.732637, "total_loss": 11.029106, "clip_frac": 0.069824, "sps": 411.3, "elapsed_s": 135.3}
+{"update": 27, "steps_done": 55296, "mean_step_reward": 0.4384, "min_step_reward": -10.588, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 25.682, "mean_ep_length": 58.5, "mean_collisions": 0.6, "mean_intent_rate": 0.359, "policy_loss": -0.000905, "value_loss": 19.803536, "entropy": 0.734127, "total_loss": 9.893522, "clip_frac": 0.072266, "sps": 414.4, "elapsed_s": 140.2}
+{"update": 28, "steps_done": 57344, "mean_step_reward": 0.3775, "min_step_reward": -10.5896, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 21.988, "mean_ep_length": 58.5, "mean_collisions": 0.69, "mean_intent_rate": 0.343, "policy_loss": 5.4e-05, "value_loss": 16.351582, "entropy": 0.719813, "total_loss": 8.168647, "clip_frac": 0.049805, "sps": 407.9, "elapsed_s": 145.3}
+{"update": 29, "steps_done": 59392, "mean_step_reward": 0.3691, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 21.837, "mean_ep_length": 59.1, "mean_collisions": 0.4, "mean_intent_rate": 0.317, "policy_loss": -0.002232, "value_loss": 19.472904, "entropy": 0.739717, "total_loss": 9.726823, "clip_frac": 0.064575, "sps": 415.3, "elapsed_s": 150.2}
+{"update": 30, "steps_done": 61440, "mean_step_reward": 0.3437, "min_step_reward": -10.5831, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 21.185, "mean_ep_length": 60.3, "mean_collisions": 0.71, "mean_intent_rate": 0.328, "policy_loss": -0.002338, "value_loss": 20.649461, "entropy": 0.692904, "total_loss": 10.315464, "clip_frac": 0.030029, "sps": 414.8, "elapsed_s": 155.1}
+{"update": 31, "steps_done": 63488, "mean_step_reward": 0.3922, "min_step_reward": -10.5799, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 23.755, "mean_ep_length": 61.0, "mean_collisions": 0.64, "mean_intent_rate": 0.334, "policy_loss": 0.003796, "value_loss": 18.011389, "entropy": 0.613184, "total_loss": 9.003359, "clip_frac": 0.143311, "sps": 416.1, "elapsed_s": 160.1}
+{"update": 32, "steps_done": 65536, "mean_step_reward": 0.4262, "min_step_reward": -10.513, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 23.91, "mean_ep_length": 56.6, "mean_collisions": 0.38, "mean_intent_rate": 0.345, "policy_loss": 0.006023, "value_loss": 18.939232, "entropy": 0.544623, "total_loss": 9.470193, "clip_frac": 0.098267, "sps": 417.5, "elapsed_s": 165.0}
+{"update": 33, "steps_done": 67584, "mean_step_reward": 0.3055, "min_step_reward": -10.5673, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 17.611, "mean_ep_length": 59.1, "mean_collisions": 0.65, "mean_intent_rate": 0.311, "policy_loss": -0.002183, "value_loss": 17.28346, "entropy": 0.434571, "total_loss": 8.635201, "clip_frac": 0.041504, "sps": 421.0, "elapsed_s": 169.8}
+{"update": 34, "steps_done": 69632, "mean_step_reward": 0.4065, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 26.732, "mean_ep_length": 63.6, "mean_collisions": 0.34, "mean_intent_rate": 0.33, "policy_loss": -0.000809, "value_loss": 19.08998, "entropy": 0.441352, "total_loss": 9.539767, "clip_frac": 0.025513, "sps": 417.6, "elapsed_s": 174.7}
+{"update": 35, "steps_done": 71680, "mean_step_reward": 0.422, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 27.059, "mean_ep_length": 65.5, "mean_collisions": 0.28, "mean_intent_rate": 0.328, "policy_loss": -0.003858, "value_loss": 18.487082, "entropy": 0.447887, "total_loss": 9.235204, "clip_frac": 0.040283, "sps": 414.0, "elapsed_s": 179.7}
+{"update": 36, "steps_done": 73728, "mean_step_reward": 0.4461, "min_step_reward": -10.5652, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 28.713, "mean_ep_length": 65.3, "mean_collisions": 0.23, "mean_intent_rate": 0.333, "policy_loss": -0.004011, "value_loss": 16.415451, "entropy": 0.46508, "total_loss": 8.199064, "clip_frac": 0.060059, "sps": 413.9, "elapsed_s": 184.6}
+{"update": 37, "steps_done": 75776, "mean_step_reward": 0.3929, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 24.799, "mean_ep_length": 62.7, "mean_collisions": 0.55, "mean_intent_rate": 0.339, "policy_loss": 0.002789, "value_loss": 18.250496, "entropy": 0.444698, "total_loss": 9.123591, "clip_frac": 0.059937, "sps": 406.4, "elapsed_s": 189.7}
+{"update": 38, "steps_done": 77824, "mean_step_reward": 0.4959, "min_step_reward": -10.5761, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 33.407, "mean_ep_length": 67.0, "mean_collisions": 0.27, "mean_intent_rate": 0.361, "policy_loss": 0.002981, "value_loss": 19.364769, "entropy": 0.455103, "total_loss": 9.680815, "clip_frac": 0.061523, "sps": 417.4, "elapsed_s": 194.6}
+{"update": 39, "steps_done": 79872, "mean_step_reward": 0.4062, "min_step_reward": -10.5843, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 24.677, "mean_ep_length": 59.9, "mean_collisions": 0.51, "mean_intent_rate": 0.342, "policy_loss": -0.000418, "value_loss": 17.317227, "entropy": 0.489526, "total_loss": 8.653301, "clip_frac": 0.024414, "sps": 411.9, "elapsed_s": 199.6}
+{"update": 40, "steps_done": 81920, "mean_step_reward": 0.3512, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 20.568, "mean_ep_length": 59.9, "mean_collisions": 0.64, "mean_intent_rate": 0.328, "policy_loss": 0.001868, "value_loss": 20.060968, "entropy": 0.444351, "total_loss": 10.027908, "clip_frac": 0.05835, "sps": 413.7, "elapsed_s": 204.5}
+{"update": 41, "steps_done": 83968, "mean_step_reward": 0.3954, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 25.731, "mean_ep_length": 64.2, "mean_collisions": 0.39, "mean_intent_rate": 0.327, "policy_loss": -0.000569, "value_loss": 17.261312, "entropy": 0.457312, "total_loss": 8.625514, "clip_frac": 0.045166, "sps": 411.3, "elapsed_s": 209.5}
+{"update": 42, "steps_done": 86016, "mean_step_reward": 0.4349, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 26.509, "mean_ep_length": 60.4, "mean_collisions": 0.42, "mean_intent_rate": 0.341, "policy_loss": 0.002362, "value_loss": 18.952302, "entropy": 0.457957, "total_loss": 9.473933, "clip_frac": 0.040405, "sps": 413.4, "elapsed_s": 214.4}
+{"update": 43, "steps_done": 88064, "mean_step_reward": 0.4613, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 26.079, "mean_ep_length": 56.2, "mean_collisions": 0.39, "mean_intent_rate": 0.352, "policy_loss": -0.00036, "value_loss": 18.564019, "entropy": 0.490465, "total_loss": 9.276745, "clip_frac": 0.025879, "sps": 414.6, "elapsed_s": 219.4}
+{"update": 44, "steps_done": 90112, "mean_step_reward": 0.4596, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 28.485, "mean_ep_length": 62.5, "mean_collisions": 0.67, "mean_intent_rate": 0.358, "policy_loss": 0.002443, "value_loss": 26.110486, "entropy": 0.55067, "total_loss": 13.05218, "clip_frac": 0.06665, "sps": 416.6, "elapsed_s": 224.3}
+{"update": 45, "steps_done": 92160, "mean_step_reward": 0.5123, "min_step_reward": -10.5889, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 31.455, "mean_ep_length": 61.9, "mean_collisions": 0.68, "mean_intent_rate": 0.385, "policy_loss": -0.000137, "value_loss": 22.865882, "entropy": 0.570873, "total_loss": 11.427095, "clip_frac": 0.046143, "sps": 405.7, "elapsed_s": 229.3}
+{"update": 46, "steps_done": 94208, "mean_step_reward": 0.5148, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 30.823, "mean_ep_length": 59.9, "mean_collisions": 0.65, "mean_intent_rate": 0.382, "policy_loss": -0.000562, "value_loss": 27.935357, "entropy": 0.637067, "total_loss": 13.960746, "clip_frac": 0.038818, "sps": 413.8, "elapsed_s": 234.3}
+{"update": 47, "steps_done": 96256, "mean_step_reward": 0.4203, "min_step_reward": -10.5433, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 24.596, "mean_ep_length": 59.3, "mean_collisions": 1.29, "mean_intent_rate": 0.385, "policy_loss": 0.005042, "value_loss": 27.879968, "entropy": 0.575499, "total_loss": 13.939271, "clip_frac": 0.104614, "sps": 414.4, "elapsed_s": 239.2}
+{"update": 48, "steps_done": 98304, "mean_step_reward": 0.4976, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 30.355, "mean_ep_length": 62.0, "mean_collisions": 0.42, "mean_intent_rate": 0.361, "policy_loss": -0.001534, "value_loss": 24.102213, "entropy": 0.523606, "total_loss": 12.044337, "clip_frac": 0.058105, "sps": 420.1, "elapsed_s": 244.1}
+{"update": 49, "steps_done": 100352, "mean_step_reward": 0.5195, "min_step_reward": -10.5288, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 29.68, "mean_ep_length": 55.5, "mean_collisions": 0.68, "mean_intent_rate": 0.392, "policy_loss": 0.002033, "value_loss": 23.907336, "entropy": 0.582679, "total_loss": 11.949874, "clip_frac": 0.065186, "sps": 415.8, "elapsed_s": 249.0}
+{"update": 50, "steps_done": 102400, "mean_step_reward": 0.4912, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 30.067, "mean_ep_length": 60.4, "mean_collisions": 0.82, "mean_intent_rate": 0.39, "policy_loss": -7e-05, "value_loss": 29.856751, "entropy": 0.564935, "total_loss": 14.922656, "clip_frac": 0.034546, "sps": 414.4, "elapsed_s": 254.0}
+{"update": 51, "steps_done": 104448, "mean_step_reward": 0.587, "min_step_reward": -10.5831, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 35.115, "mean_ep_length": 61.1, "mean_collisions": 0.94, "mean_intent_rate": 0.415, "policy_loss": -0.000546, "value_loss": 32.759215, "entropy": 0.604026, "total_loss": 16.373022, "clip_frac": 0.026978, "sps": 414.6, "elapsed_s": 258.9}
+{"update": 52, "steps_done": 106496, "mean_step_reward": 0.4274, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 26.019, "mean_ep_length": 61.7, "mean_collisions": 1.36, "mean_intent_rate": 0.387, "policy_loss": -8.8e-05, "value_loss": 30.881102, "entropy": 0.636704, "total_loss": 15.434096, "clip_frac": 0.014648, "sps": 411.8, "elapsed_s": 263.9}
+{"update": 53, "steps_done": 108544, "mean_step_reward": 0.4441, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 27.093, "mean_ep_length": 60.2, "mean_collisions": 1.56, "mean_intent_rate": 0.415, "policy_loss": 0.008394, "value_loss": 32.830533, "entropy": 0.610024, "total_loss": 16.417561, "clip_frac": 0.098633, "sps": 414.6, "elapsed_s": 268.9}
+{"update": 54, "steps_done": 110592, "mean_step_reward": 0.5066, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 31.982, "mean_ep_length": 62.7, "mean_collisions": 1.09, "mean_intent_rate": 0.406, "policy_loss": -0.002773, "value_loss": 24.871068, "entropy": 0.645776, "total_loss": 12.426303, "clip_frac": 0.048462, "sps": 403.5, "elapsed_s": 273.9}
+{"update": 55, "steps_done": 112640, "mean_step_reward": 0.5599, "min_step_reward": -10.589, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 38.049, "mean_ep_length": 69.4, "mean_collisions": 1.47, "mean_intent_rate": 0.43, "policy_loss": -0.000988, "value_loss": 38.278041, "entropy": 0.605333, "total_loss": 19.131979, "clip_frac": 0.042114, "sps": 400.3, "elapsed_s": 279.0}
+{"update": 56, "steps_done": 114688, "mean_step_reward": 0.5351, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 35.546, "mean_ep_length": 64.2, "mean_collisions": 0.59, "mean_intent_rate": 0.388, "policy_loss": 0.001586, "value_loss": 26.483618, "entropy": 0.613102, "total_loss": 13.237264, "clip_frac": 0.061035, "sps": 409.2, "elapsed_s": 284.1}
+{"update": 57, "steps_done": 116736, "mean_step_reward": 0.5973, "min_step_reward": -10.5724, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 33.767, "mean_ep_length": 58.7, "mean_collisions": 0.85, "mean_intent_rate": 0.41, "policy_loss": -0.000228, "value_loss": 32.305146, "entropy": 0.621852, "total_loss": 16.146126, "clip_frac": 0.039673, "sps": 414.4, "elapsed_s": 289.0}
+{"update": 58, "steps_done": 118784, "mean_step_reward": 0.4884, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 30.031, "mean_ep_length": 60.4, "mean_collisions": 0.97, "mean_intent_rate": 0.4, "policy_loss": -0.00091, "value_loss": 31.88257, "entropy": 0.614756, "total_loss": 15.934228, "clip_frac": 0.049683, "sps": 400.3, "elapsed_s": 294.1}
+{"update": 59, "steps_done": 120832, "mean_step_reward": 0.5648, "min_step_reward": -10.5674, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 33.541, "mean_ep_length": 59.3, "mean_collisions": 1.09, "mean_intent_rate": 0.43, "policy_loss": 9.5e-05, "value_loss": 31.356182, "entropy": 0.518473, "total_loss": 15.673001, "clip_frac": 0.018921, "sps": 408.6, "elapsed_s": 299.1}
+{"update": 60, "steps_done": 122880, "mean_step_reward": 0.5877, "min_step_reward": -10.589, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 34.268, "mean_ep_length": 58.2, "mean_collisions": 0.86, "mean_intent_rate": 0.42, "policy_loss": 0.000118, "value_loss": 34.560794, "entropy": 0.591297, "total_loss": 17.274602, "clip_frac": 0.043945, "sps": 413.6, "elapsed_s": 304.1}
+{"update": 61, "steps_done": 124928, "mean_step_reward": 0.6005, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 35.0, "mean_ep_length": 58.4, "mean_collisions": 0.85, "mean_intent_rate": 0.425, "policy_loss": -0.000525, "value_loss": 32.638276, "entropy": 0.584424, "total_loss": 16.312769, "clip_frac": 0.038696, "sps": 415.3, "elapsed_s": 309.0}
+{"update": 62, "steps_done": 126976, "mean_step_reward": 0.3875, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 23.657, "mean_ep_length": 60.1, "mean_collisions": 1.89, "mean_intent_rate": 0.403, "policy_loss": 0.002736, "value_loss": 37.866268, "entropy": 0.563102, "total_loss": 18.930239, "clip_frac": 0.050781, "sps": 408.1, "elapsed_s": 314.0}
+{"update": 63, "steps_done": 129024, "mean_step_reward": 0.505, "min_step_reward": -10.588, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 29.575, "mean_ep_length": 58.3, "mean_collisions": 1.71, "mean_intent_rate": 0.444, "policy_loss": -0.000473, "value_loss": 40.032136, "entropy": 0.538297, "total_loss": 20.010212, "clip_frac": 0.027344, "sps": 420.3, "elapsed_s": 318.9}
+{"update": 64, "steps_done": 131072, "mean_step_reward": 0.5756, "min_step_reward": -10.5848, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 37.363, "mean_ep_length": 65.6, "mean_collisions": 1.1, "mean_intent_rate": 0.419, "policy_loss": -0.000812, "value_loss": 29.308134, "entropy": 0.549473, "total_loss": 14.64776, "clip_frac": 0.041016, "sps": 413.4, "elapsed_s": 323.9}
+{"update": 65, "steps_done": 133120, "mean_step_reward": 0.6118, "min_step_reward": -10.5616, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 36.781, "mean_ep_length": 60.9, "mean_collisions": 0.65, "mean_intent_rate": 0.406, "policy_loss": 8.3e-05, "value_loss": 29.407993, "entropy": 0.567352, "total_loss": 14.698406, "clip_frac": 0.052979, "sps": 415.1, "elapsed_s": 328.8}
+{"update": 66, "steps_done": 135168, "mean_step_reward": 0.6404, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 41.115, "mean_ep_length": 62.9, "mean_collisions": 0.82, "mean_intent_rate": 0.429, "policy_loss": -0.001614, "value_loss": 33.002676, "entropy": 0.574048, "total_loss": 16.493983, "clip_frac": 0.03894, "sps": 416.9, "elapsed_s": 333.7}
+{"update": 67, "steps_done": 137216, "mean_step_reward": 0.5549, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 34.679, "mean_ep_length": 62.9, "mean_collisions": 2.0, "mean_intent_rate": 0.464, "policy_loss": -0.002901, "value_loss": 43.318844, "entropy": 0.578404, "total_loss": 21.650736, "clip_frac": 0.059448, "sps": 416.0, "elapsed_s": 338.6}
+{"update": 68, "steps_done": 139264, "mean_step_reward": 0.619, "min_step_reward": -10.5866, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 40.75, "mean_ep_length": 65.0, "mean_collisions": 1.03, "mean_intent_rate": 0.439, "policy_loss": 0.001113, "value_loss": 31.771315, "entropy": 0.550233, "total_loss": 15.881268, "clip_frac": 0.014404, "sps": 414.7, "elapsed_s": 343.6}
+{"update": 69, "steps_done": 141312, "mean_step_reward": 0.5761, "min_step_reward": -10.5789, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 33.648, "mean_ep_length": 58.8, "mean_collisions": 1.43, "mean_intent_rate": 0.445, "policy_loss": 0.003871, "value_loss": 38.229028, "entropy": 0.592952, "total_loss": 19.112455, "clip_frac": 0.069458, "sps": 427.7, "elapsed_s": 348.4}
+{"update": 70, "steps_done": 143360, "mean_step_reward": 0.5254, "min_step_reward": -10.5551, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 32.975, "mean_ep_length": 62.9, "mean_collisions": 1.21, "mean_intent_rate": 0.413, "policy_loss": 0.007388, "value_loss": 38.42157, "entropy": 0.544781, "total_loss": 19.212725, "clip_frac": 0.060913, "sps": 419.9, "elapsed_s": 353.2}
+{"update": 71, "steps_done": 145408, "mean_step_reward": 0.619, "min_step_reward": -10.5876, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 37.424, "mean_ep_length": 60.8, "mean_collisions": 1.33, "mean_intent_rate": 0.454, "policy_loss": 0.000547, "value_loss": 39.188851, "entropy": 0.607636, "total_loss": 19.588896, "clip_frac": 0.087769, "sps": 407.7, "elapsed_s": 358.3}
+{"update": 72, "steps_done": 147456, "mean_step_reward": 0.6651, "min_step_reward": -10.5863, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.031, "mean_ep_length": 60.3, "mean_collisions": 2.21, "mean_intent_rate": 0.514, "policy_loss": 1e-05, "value_loss": 49.294181, "entropy": 0.598223, "total_loss": 24.641119, "clip_frac": 0.032227, "sps": 415.3, "elapsed_s": 363.2}
+{"update": 73, "steps_done": 149504, "mean_step_reward": 0.6056, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 36.719, "mean_ep_length": 58.7, "mean_collisions": 0.97, "mean_intent_rate": 0.44, "policy_loss": -0.000481, "value_loss": 42.715315, "entropy": 0.576614, "total_loss": 21.35141, "clip_frac": 0.073242, "sps": 415.2, "elapsed_s": 368.1}
+{"update": 74, "steps_done": 151552, "mean_step_reward": 0.5624, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 34.027, "mean_ep_length": 62.3, "mean_collisions": 1.55, "mean_intent_rate": 0.444, "policy_loss": 0.001672, "value_loss": 40.378847, "entropy": 0.582278, "total_loss": 20.185273, "clip_frac": 0.060181, "sps": 415.4, "elapsed_s": 373.1}
+{"update": 75, "steps_done": 153600, "mean_step_reward": 0.5586, "min_step_reward": -10.5832, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 35.78, "mean_ep_length": 62.0, "mean_collisions": 1.24, "mean_intent_rate": 0.436, "policy_loss": -8.9e-05, "value_loss": 40.359126, "entropy": 0.515083, "total_loss": 20.174323, "clip_frac": 0.055542, "sps": 411.3, "elapsed_s": 378.0}
+{"update": 76, "steps_done": 155648, "mean_step_reward": 0.5616, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 32.509, "mean_ep_length": 58.5, "mean_collisions": 1.15, "mean_intent_rate": 0.43, "policy_loss": 0.002169, "value_loss": 39.722772, "entropy": 0.498199, "total_loss": 19.858574, "clip_frac": 0.030518, "sps": 419.0, "elapsed_s": 382.9}
+{"update": 77, "steps_done": 157696, "mean_step_reward": 0.4994, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 30.663, "mean_ep_length": 62.1, "mean_collisions": 1.06, "mean_intent_rate": 0.398, "policy_loss": -0.002267, "value_loss": 31.369032, "entropy": 0.494297, "total_loss": 15.677306, "clip_frac": 0.02771, "sps": 414.3, "elapsed_s": 387.9}
+{"update": 78, "steps_done": 159744, "mean_step_reward": 0.6052, "min_step_reward": -10.5624, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 34.776, "mean_ep_length": 57.3, "mean_collisions": 0.75, "mean_intent_rate": 0.424, "policy_loss": -0.000298, "value_loss": 35.116498, "entropy": 0.494246, "total_loss": 17.553009, "clip_frac": 0.02478, "sps": 412.9, "elapsed_s": 392.8}
+{"update": 79, "steps_done": 161792, "mean_step_reward": 0.6149, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 36.716, "mean_ep_length": 59.8, "mean_collisions": 0.97, "mean_intent_rate": 0.435, "policy_loss": -0.000428, "value_loss": 38.71159, "entropy": 0.545723, "total_loss": 19.34991, "clip_frac": 0.057739, "sps": 407.4, "elapsed_s": 397.9}
+{"update": 80, "steps_done": 163840, "mean_step_reward": 0.6546, "min_step_reward": -10.5734, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 39.822, "mean_ep_length": 60.7, "mean_collisions": 1.03, "mean_intent_rate": 0.451, "policy_loss": 0.003063, "value_loss": 41.489977, "entropy": 0.548425, "total_loss": 20.742567, "clip_frac": 0.064819, "sps": 413.9, "elapsed_s": 402.8}
+{"update": 81, "steps_done": 165888, "mean_step_reward": 0.7814, "min_step_reward": -10.5884, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 47.21, "mean_ep_length": 61.2, "mean_collisions": 1.21, "mean_intent_rate": 0.498, "policy_loss": 0.000877, "value_loss": 53.403998, "entropy": 0.636092, "total_loss": 26.696515, "clip_frac": 0.062134, "sps": 418.8, "elapsed_s": 407.7}
+{"update": 82, "steps_done": 167936, "mean_step_reward": 0.5617, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 35.476, "mean_ep_length": 62.2, "mean_collisions": 2.64, "mean_intent_rate": 0.509, "policy_loss": 0.00829, "value_loss": 55.013009, "entropy": 0.662744, "total_loss": 27.508167, "clip_frac": 0.133667, "sps": 414.2, "elapsed_s": 412.6}
+{"update": 83, "steps_done": 169984, "mean_step_reward": 0.7278, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 43.995, "mean_ep_length": 61.7, "mean_collisions": 2.09, "mean_intent_rate": 0.525, "policy_loss": 0.000647, "value_loss": 56.306124, "entropy": 0.668285, "total_loss": 28.147027, "clip_frac": 0.050171, "sps": 422.1, "elapsed_s": 417.5}
+{"update": 84, "steps_done": 172032, "mean_step_reward": 0.5488, "min_step_reward": -10.5603, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 35.254, "mean_ep_length": 63.0, "mean_collisions": 2.45, "mean_intent_rate": 0.494, "policy_loss": -0.001206, "value_loss": 57.585649, "entropy": 0.671661, "total_loss": 28.784903, "clip_frac": 0.024536, "sps": 418.5, "elapsed_s": 422.4}
+{"update": 85, "steps_done": 174080, "mean_step_reward": 0.6736, "min_step_reward": -10.5705, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 41.672, "mean_ep_length": 62.2, "mean_collisions": 2.06, "mean_intent_rate": 0.508, "policy_loss": 0.002197, "value_loss": 51.352004, "entropy": 0.667983, "total_loss": 25.671519, "clip_frac": 0.092529, "sps": 413.5, "elapsed_s": 427.3}
+{"update": 86, "steps_done": 176128, "mean_step_reward": 0.5473, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 33.631, "mean_ep_length": 60.7, "mean_collisions": 3.06, "mean_intent_rate": 0.523, "policy_loss": 0.000459, "value_loss": 68.536802, "entropy": 0.663899, "total_loss": 34.262221, "clip_frac": 0.168579, "sps": 419.2, "elapsed_s": 432.2}
+{"update": 87, "steps_done": 178176, "mean_step_reward": 0.7129, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 47.561, "mean_ep_length": 66.1, "mean_collisions": 2.13, "mean_intent_rate": 0.528, "policy_loss": -0.000849, "value_loss": 54.885573, "entropy": 0.6648, "total_loss": 27.43529, "clip_frac": 0.029541, "sps": 410.7, "elapsed_s": 437.2}
+{"update": 88, "steps_done": 180224, "mean_step_reward": 0.7697, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 49.415, "mean_ep_length": 64.6, "mean_collisions": 2.16, "mean_intent_rate": 0.544, "policy_loss": -0.000306, "value_loss": 56.648234, "entropy": 0.663211, "total_loss": 28.31718, "clip_frac": 0.036743, "sps": 402.9, "elapsed_s": 442.3}
+{"update": 89, "steps_done": 182272, "mean_step_reward": 0.7528, "min_step_reward": -10.5912, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 44.142, "mean_ep_length": 59.2, "mean_collisions": 1.49, "mean_intent_rate": 0.508, "policy_loss": 0.00059, "value_loss": 52.210516, "entropy": 0.686808, "total_loss": 26.09898, "clip_frac": 0.116821, "sps": 406.9, "elapsed_s": 447.3}
+{"update": 90, "steps_done": 184320, "mean_step_reward": 0.8048, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 48.424, "mean_ep_length": 60.7, "mean_collisions": 1.79, "mean_intent_rate": 0.543, "policy_loss": 0.002104, "value_loss": 59.899416, "entropy": 0.678342, "total_loss": 29.945028, "clip_frac": 0.085083, "sps": 406.0, "elapsed_s": 452.4}
+{"update": 91, "steps_done": 186368, "mean_step_reward": 0.7744, "min_step_reward": -10.5901, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 45.928, "mean_ep_length": 57.9, "mean_collisions": 2.06, "mean_intent_rate": 0.558, "policy_loss": 0.006656, "value_loss": 68.650673, "entropy": 0.695953, "total_loss": 34.325033, "clip_frac": 0.115479, "sps": 412.5, "elapsed_s": 457.3}
+{"update": 92, "steps_done": 188416, "mean_step_reward": 0.5567, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 34.766, "mean_ep_length": 61.5, "mean_collisions": 3.24, "mean_intent_rate": 0.544, "policy_loss": -0.002376, "value_loss": 60.967323, "entropy": 0.656995, "total_loss": 30.474715, "clip_frac": 0.032715, "sps": 420.5, "elapsed_s": 462.2}
+{"update": 93, "steps_done": 190464, "mean_step_reward": 0.6818, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 41.83, "mean_ep_length": 64.8, "mean_collisions": 2.94, "mean_intent_rate": 0.538, "policy_loss": -0.001106, "value_loss": 57.8066, "entropy": 0.613998, "total_loss": 28.896054, "clip_frac": 0.018921, "sps": 414.5, "elapsed_s": 467.2}
+{"update": 94, "steps_done": 192512, "mean_step_reward": 0.6602, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 41.607, "mean_ep_length": 61.8, "mean_collisions": 2.29, "mean_intent_rate": 0.521, "policy_loss": 0.001611, "value_loss": 59.709005, "entropy": 0.600798, "total_loss": 29.850106, "clip_frac": 0.111206, "sps": 416.5, "elapsed_s": 472.1}
+{"update": 95, "steps_done": 194560, "mean_step_reward": 0.6783, "min_step_reward": -10.5784, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 44.977, "mean_ep_length": 66.0, "mean_collisions": 1.16, "mean_intent_rate": 0.464, "policy_loss": 0.001046, "value_loss": 45.620886, "entropy": 0.522038, "total_loss": 22.806269, "clip_frac": 0.034546, "sps": 411.9, "elapsed_s": 477.0}
+{"update": 96, "steps_done": 196608, "mean_step_reward": 0.6685, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 44.437, "mean_ep_length": 65.8, "mean_collisions": 1.52, "mean_intent_rate": 0.473, "policy_loss": 0.001912, "value_loss": 44.404638, "entropy": 0.523382, "total_loss": 22.198997, "clip_frac": 0.070801, "sps": 403.3, "elapsed_s": 482.1}
+{"update": 97, "steps_done": 198656, "mean_step_reward": 0.6374, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 37.495, "mean_ep_length": 59.3, "mean_collisions": 1.77, "mean_intent_rate": 0.494, "policy_loss": -0.001195, "value_loss": 59.19711, "entropy": 0.590932, "total_loss": 29.59145, "clip_frac": 0.075928, "sps": 404.0, "elapsed_s": 487.2}
+{"update": 98, "steps_done": 200704, "mean_step_reward": 0.8417, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 52.016, "mean_ep_length": 62.1, "mean_collisions": 0.59, "mean_intent_rate": 0.488, "policy_loss": -0.001459, "value_loss": 46.814317, "entropy": 0.574817, "total_loss": 23.399951, "clip_frac": 0.068726, "sps": 416.0, "elapsed_s": 492.1}
+{"update": 99, "steps_done": 202752, "mean_step_reward": 0.8164, "min_step_reward": -10.5857, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.043, "mean_ep_length": 62.8, "mean_collisions": 1.24, "mean_intent_rate": 0.51, "policy_loss": 0.000964, "value_loss": 55.830724, "entropy": 0.559714, "total_loss": 27.910729, "clip_frac": 0.057495, "sps": 414.6, "elapsed_s": 497.1}
+{"update": 100, "steps_done": 204800, "mean_step_reward": 0.8159, "min_step_reward": -10.5824, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 53.793, "mean_ep_length": 65.2, "mean_collisions": 0.94, "mean_intent_rate": 0.5, "policy_loss": 0.001016, "value_loss": 56.950975, "entropy": 0.596439, "total_loss": 28.470539, "clip_frac": 0.132812, "sps": 413.2, "elapsed_s": 502.0}
+{"update": 101, "steps_done": 206848, "mean_step_reward": 0.8111, "min_step_reward": -10.5827, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 50.353, "mean_ep_length": 62.3, "mean_collisions": 0.91, "mean_intent_rate": 0.498, "policy_loss": -0.000144, "value_loss": 60.86435, "entropy": 0.53886, "total_loss": 30.426642, "clip_frac": 0.046631, "sps": 409.5, "elapsed_s": 507.0}
+{"update": 102, "steps_done": 208896, "mean_step_reward": 0.8051, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 54.911, "mean_ep_length": 67.9, "mean_collisions": 1.19, "mean_intent_rate": 0.499, "policy_loss": 0.001965, "value_loss": 60.486964, "entropy": 0.571931, "total_loss": 30.239728, "clip_frac": 0.106079, "sps": 412.7, "elapsed_s": 512.0}
+{"update": 103, "steps_done": 210944, "mean_step_reward": 0.6145, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 36.616, "mean_ep_length": 59.8, "mean_collisions": 1.06, "mean_intent_rate": 0.443, "policy_loss": 0.003825, "value_loss": 49.519928, "entropy": 0.488633, "total_loss": 24.758902, "clip_frac": 0.055176, "sps": 410.6, "elapsed_s": 517.0}
+{"update": 104, "steps_done": 212992, "mean_step_reward": 0.7961, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 49.977, "mean_ep_length": 63.4, "mean_collisions": 0.75, "mean_intent_rate": 0.475, "policy_loss": -0.000283, "value_loss": 51.844605, "entropy": 0.588867, "total_loss": 25.916131, "clip_frac": 0.087036, "sps": 415.5, "elapsed_s": 521.9}
+{"update": 105, "steps_done": 215040, "mean_step_reward": 0.6423, "min_step_reward": -10.5306, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 39.76, "mean_ep_length": 60.7, "mean_collisions": 0.88, "mean_intent_rate": 0.445, "policy_loss": 0.000683, "value_loss": 52.12537, "entropy": 0.454714, "total_loss": 26.058821, "clip_frac": 0.055542, "sps": 401.6, "elapsed_s": 527.0}
+{"update": 106, "steps_done": 217088, "mean_step_reward": 0.7085, "min_step_reward": -10.5416, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 48.936, "mean_ep_length": 70.0, "mean_collisions": 1.0, "mean_intent_rate": 0.457, "policy_loss": 0.001063, "value_loss": 47.602888, "entropy": 0.496588, "total_loss": 23.797541, "clip_frac": 0.036011, "sps": 415.1, "elapsed_s": 532.0}
+{"update": 107, "steps_done": 219136, "mean_step_reward": 0.7886, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 45.778, "mean_ep_length": 57.6, "mean_collisions": 0.86, "mean_intent_rate": 0.484, "policy_loss": 0.005155, "value_loss": 67.739528, "entropy": 0.543459, "total_loss": 33.869484, "clip_frac": 0.05957, "sps": 412.0, "elapsed_s": 536.9}
+{"update": 108, "steps_done": 221184, "mean_step_reward": 0.8362, "min_step_reward": -10.5867, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.576, "mean_ep_length": 61.8, "mean_collisions": 0.85, "mean_intent_rate": 0.496, "policy_loss": -0.001639, "value_loss": 57.02123, "entropy": 0.572973, "total_loss": 28.503246, "clip_frac": 0.062622, "sps": 413.7, "elapsed_s": 541.9}
+{"update": 109, "steps_done": 223232, "mean_step_reward": 0.8156, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 50.191, "mean_ep_length": 61.3, "mean_collisions": 1.48, "mean_intent_rate": 0.528, "policy_loss": 0.000154, "value_loss": 59.960654, "entropy": 0.587618, "total_loss": 29.974604, "clip_frac": 0.049194, "sps": 416.3, "elapsed_s": 546.8}
+{"update": 110, "steps_done": 225280, "mean_step_reward": 0.7869, "min_step_reward": -10.5707, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 44.656, "mean_ep_length": 57.4, "mean_collisions": 0.75, "mean_intent_rate": 0.472, "policy_loss": 0.003122, "value_loss": 55.972008, "entropy": 0.569105, "total_loss": 27.983435, "clip_frac": 0.072632, "sps": 415.1, "elapsed_s": 551.7}
+{"update": 111, "steps_done": 227328, "mean_step_reward": 0.7588, "min_step_reward": -10.549, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 46.076, "mean_ep_length": 60.4, "mean_collisions": 1.38, "mean_intent_rate": 0.509, "policy_loss": 0.002236, "value_loss": 72.073726, "entropy": 0.554126, "total_loss": 36.033558, "clip_frac": 0.05896, "sps": 416.8, "elapsed_s": 556.6}
+{"update": 112, "steps_done": 229376, "mean_step_reward": 0.6661, "min_step_reward": -10.5795, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 41.598, "mean_ep_length": 62.3, "mean_collisions": 2.27, "mean_intent_rate": 0.526, "policy_loss": 0.001941, "value_loss": 70.469727, "entropy": 0.583124, "total_loss": 35.230973, "clip_frac": 0.12085, "sps": 413.7, "elapsed_s": 561.6}
+{"update": 113, "steps_done": 231424, "mean_step_reward": 1.0067, "min_step_reward": -10.5734, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 64.809, "mean_ep_length": 64.3, "mean_collisions": 0.81, "mean_intent_rate": 0.56, "policy_loss": 0.003459, "value_loss": 67.676957, "entropy": 0.581576, "total_loss": 33.836122, "clip_frac": 0.101807, "sps": 403.5, "elapsed_s": 566.7}
+{"update": 114, "steps_done": 233472, "mean_step_reward": 0.8798, "min_step_reward": -10.5498, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.687, "mean_ep_length": 62.7, "mean_collisions": 2.09, "mean_intent_rate": 0.583, "policy_loss": 0.008431, "value_loss": 72.942695, "entropy": 0.623444, "total_loss": 36.473545, "clip_frac": 0.132568, "sps": 416.7, "elapsed_s": 571.6}
+{"update": 115, "steps_done": 235520, "mean_step_reward": 0.7732, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 48.271, "mean_ep_length": 59.7, "mean_collisions": 2.35, "mean_intent_rate": 0.58, "policy_loss": 0.000547, "value_loss": 91.937474, "entropy": 0.603061, "total_loss": 45.963253, "clip_frac": 0.129395, "sps": 410.8, "elapsed_s": 576.6}
+{"update": 116, "steps_done": 237568, "mean_step_reward": 0.79, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 44.57, "mean_ep_length": 57.4, "mean_collisions": 2.31, "mean_intent_rate": 0.569, "policy_loss": 0.006642, "value_loss": 75.872967, "entropy": 0.588448, "total_loss": 37.937241, "clip_frac": 0.15686, "sps": 415.3, "elapsed_s": 581.5}
+{"update": 117, "steps_done": 239616, "mean_step_reward": 0.641, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 39.328, "mean_ep_length": 62.9, "mean_collisions": 2.36, "mean_intent_rate": 0.508, "policy_loss": 0.008492, "value_loss": 63.201426, "entropy": 0.557364, "total_loss": 31.603631, "clip_frac": 0.11145, "sps": 411.8, "elapsed_s": 586.5}
+{"update": 118, "steps_done": 241664, "mean_step_reward": 0.7698, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 46.53, "mean_ep_length": 60.3, "mean_collisions": 1.56, "mean_intent_rate": 0.515, "policy_loss": 0.011044, "value_loss": 57.643222, "entropy": 0.539337, "total_loss": 28.827262, "clip_frac": 0.160767, "sps": 411.1, "elapsed_s": 591.5}
+{"update": 119, "steps_done": 243712, "mean_step_reward": 0.7597, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 43.267, "mean_ep_length": 56.9, "mean_collisions": 1.03, "mean_intent_rate": 0.488, "policy_loss": 0.002275, "value_loss": 50.067278, "entropy": 0.549319, "total_loss": 25.030421, "clip_frac": 0.085571, "sps": 413.0, "elapsed_s": 596.4}
+{"update": 120, "steps_done": 245760, "mean_step_reward": 0.6807, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 44.781, "mean_ep_length": 65.8, "mean_collisions": 1.59, "mean_intent_rate": 0.486, "policy_loss": 0.011688, "value_loss": 67.186672, "entropy": 0.511843, "total_loss": 33.599906, "clip_frac": 0.178101, "sps": 415.5, "elapsed_s": 601.4}
+{"update": 121, "steps_done": 247808, "mean_step_reward": 0.8169, "min_step_reward": -10.5679, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 50.148, "mean_ep_length": 61.4, "mean_collisions": 0.85, "mean_intent_rate": 0.492, "policy_loss": 0.002876, "value_loss": 50.254332, "entropy": 0.495519, "total_loss": 25.125087, "clip_frac": 0.092529, "sps": 410.0, "elapsed_s": 606.3}
+{"update": 122, "steps_done": 249856, "mean_step_reward": 0.6265, "min_step_reward": -10.552, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 38.31, "mean_ep_length": 60.6, "mean_collisions": 0.65, "mean_intent_rate": 0.423, "policy_loss": 0.004584, "value_loss": 46.706901, "entropy": 0.457775, "total_loss": 23.353457, "clip_frac": 0.064819, "sps": 406.7, "elapsed_s": 611.4}
+{"update": 123, "steps_done": 251904, "mean_step_reward": 0.6084, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 35.58, "mean_ep_length": 58.7, "mean_collisions": 1.46, "mean_intent_rate": 0.462, "policy_loss": 0.00348, "value_loss": 49.412022, "entropy": 0.488316, "total_loss": 24.704608, "clip_frac": 0.119263, "sps": 414.1, "elapsed_s": 616.3}
+{"update": 124, "steps_done": 253952, "mean_step_reward": 0.6327, "min_step_reward": -10.538, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 40.294, "mean_ep_length": 63.2, "mean_collisions": 0.78, "mean_intent_rate": 0.43, "policy_loss": 0.00591, "value_loss": 41.340517, "entropy": 0.450175, "total_loss": 20.671667, "clip_frac": 0.101074, "sps": 414.5, "elapsed_s": 621.3}
+{"update": 125, "steps_done": 256000, "mean_step_reward": 0.7639, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 45.975, "mean_ep_length": 60.2, "mean_collisions": 0.74, "mean_intent_rate": 0.468, "policy_loss": 0.005106, "value_loss": 44.899613, "entropy": 0.49475, "total_loss": 22.449966, "clip_frac": 0.101685, "sps": 411.6, "elapsed_s": 626.2}
+{"update": 126, "steps_done": 258048, "mean_step_reward": 0.6614, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 41.968, "mean_ep_length": 62.8, "mean_collisions": 1.36, "mean_intent_rate": 0.47, "policy_loss": 0.000474, "value_loss": 50.424968, "entropy": 0.466728, "total_loss": 25.208291, "clip_frac": 0.069336, "sps": 411.0, "elapsed_s": 631.2}
+{"update": 127, "steps_done": 260096, "mean_step_reward": 0.6144, "min_step_reward": -10.5896, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 36.043, "mean_ep_length": 60.2, "mean_collisions": 1.79, "mean_intent_rate": 0.47, "policy_loss": 0.015289, "value_loss": 59.102692, "entropy": 0.434966, "total_loss": 29.562285, "clip_frac": 0.189697, "sps": 411.9, "elapsed_s": 636.2}
+{"update": 128, "steps_done": 262144, "mean_step_reward": 0.6869, "min_step_reward": -10.5443, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 40.902, "mean_ep_length": 60.1, "mean_collisions": 0.94, "mean_intent_rate": 0.445, "policy_loss": 0.005179, "value_loss": 51.246785, "entropy": 0.472549, "total_loss": 25.623846, "clip_frac": 0.105713, "sps": 412.7, "elapsed_s": 641.2}
+{"update": 129, "steps_done": 264192, "mean_step_reward": 0.5484, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 35.75, "mean_ep_length": 64.2, "mean_collisions": 1.75, "mean_intent_rate": 0.454, "policy_loss": 0.001271, "value_loss": 62.648638, "entropy": 0.42542, "total_loss": 31.321337, "clip_frac": 0.036621, "sps": 416.6, "elapsed_s": 646.1}
+{"update": 130, "steps_done": 266240, "mean_step_reward": 0.7343, "min_step_reward": -10.5841, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 42.223, "mean_ep_length": 57.1, "mean_collisions": 0.89, "mean_intent_rate": 0.476, "policy_loss": -0.002319, "value_loss": 47.973247, "entropy": 0.535363, "total_loss": 23.97895, "clip_frac": 0.108398, "sps": 404.4, "elapsed_s": 651.1}
+{"update": 131, "steps_done": 268288, "mean_step_reward": 0.673, "min_step_reward": -10.5843, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 39.755, "mean_ep_length": 59.6, "mean_collisions": 0.74, "mean_intent_rate": 0.436, "policy_loss": -0.001271, "value_loss": 41.262875, "entropy": 0.51572, "total_loss": 20.625009, "clip_frac": 0.099487, "sps": 415.1, "elapsed_s": 656.1}
+{"update": 132, "steps_done": 270336, "mean_step_reward": 0.8403, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.178, "mean_ep_length": 61.1, "mean_collisions": 1.3, "mean_intent_rate": 0.523, "policy_loss": 0.008058, "value_loss": 47.979014, "entropy": 0.597549, "total_loss": 23.991589, "clip_frac": 0.201294, "sps": 410.5, "elapsed_s": 661.1}
+{"update": 133, "steps_done": 272384, "mean_step_reward": 0.9019, "min_step_reward": -10.5651, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 54.895, "mean_ep_length": 60.9, "mean_collisions": 1.12, "mean_intent_rate": 0.536, "policy_loss": 0.003589, "value_loss": 59.448529, "entropy": 0.537615, "total_loss": 29.722478, "clip_frac": 0.098999, "sps": 414.3, "elapsed_s": 666.0}
+{"update": 134, "steps_done": 274432, "mean_step_reward": 0.9706, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 56.802, "mean_ep_length": 58.5, "mean_collisions": 1.29, "mean_intent_rate": 0.571, "policy_loss": 0.009654, "value_loss": 76.865008, "entropy": 0.533338, "total_loss": 38.436825, "clip_frac": 0.151245, "sps": 413.2, "elapsed_s": 671.0}
+{"update": 135, "steps_done": 276480, "mean_step_reward": 0.7485, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 50.71, "mean_ep_length": 67.0, "mean_collisions": 1.71, "mean_intent_rate": 0.51, "policy_loss": 0.000814, "value_loss": 63.349555, "entropy": 0.497779, "total_loss": 31.670613, "clip_frac": 0.072144, "sps": 415.7, "elapsed_s": 675.9}
+{"update": 136, "steps_done": 278528, "mean_step_reward": 0.7011, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 42.589, "mean_ep_length": 61.0, "mean_collisions": 1.06, "mean_intent_rate": 0.466, "policy_loss": 0.000607, "value_loss": 55.00869, "entropy": 0.473597, "total_loss": 27.500216, "clip_frac": 0.05896, "sps": 413.5, "elapsed_s": 680.9}
+{"update": 137, "steps_done": 280576, "mean_step_reward": 0.7692, "min_step_reward": -10.5557, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 49.919, "mean_ep_length": 65.0, "mean_collisions": 1.03, "mean_intent_rate": 0.491, "policy_loss": 0.002033, "value_loss": 49.322874, "entropy": 0.453483, "total_loss": 24.658935, "clip_frac": 0.061157, "sps": 409.5, "elapsed_s": 685.9}
+{"update": 138, "steps_done": 282624, "mean_step_reward": 0.6817, "min_step_reward": -10.583, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 45.183, "mean_ep_length": 66.4, "mean_collisions": 0.74, "mean_intent_rate": 0.443, "policy_loss": -0.001612, "value_loss": 45.038534, "entropy": 0.432845, "total_loss": 22.513327, "clip_frac": 0.049316, "sps": 415.3, "elapsed_s": 690.8}
+{"update": 139, "steps_done": 284672, "mean_step_reward": 0.9386, "min_step_reward": -10.5439, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 55.988, "mean_ep_length": 59.2, "mean_collisions": 0.62, "mean_intent_rate": 0.522, "policy_loss": 0.006912, "value_loss": 57.408047, "entropy": 0.510083, "total_loss": 28.705835, "clip_frac": 0.203369, "sps": 407.9, "elapsed_s": 695.8}
+{"update": 140, "steps_done": 286720, "mean_step_reward": 0.878, "min_step_reward": -10.5623, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 53.709, "mean_ep_length": 61.4, "mean_collisions": 1.26, "mean_intent_rate": 0.534, "policy_loss": 0.000209, "value_loss": 68.118142, "entropy": 0.495455, "total_loss": 34.054325, "clip_frac": 0.060181, "sps": 412.1, "elapsed_s": 700.8}
+{"update": 141, "steps_done": 288768, "mean_step_reward": 0.9143, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 54.894, "mean_ep_length": 59.3, "mean_collisions": 0.91, "mean_intent_rate": 0.534, "policy_loss": 0.002024, "value_loss": 74.018429, "entropy": 0.477657, "total_loss": 37.006462, "clip_frac": 0.057983, "sps": 415.9, "elapsed_s": 705.7}
+{"update": 142, "steps_done": 290816, "mean_step_reward": 0.9297, "min_step_reward": -10.5844, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 53.933, "mean_ep_length": 59.1, "mean_collisions": 1.14, "mean_intent_rate": 0.536, "policy_loss": 0.002734, "value_loss": 73.881081, "entropy": 0.48402, "total_loss": 36.938434, "clip_frac": 0.04541, "sps": 415.5, "elapsed_s": 710.6}
+{"update": 143, "steps_done": 292864, "mean_step_reward": 0.8526, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.208, "mean_ep_length": 63.6, "mean_collisions": 1.03, "mean_intent_rate": 0.512, "policy_loss": -0.002475, "value_loss": 68.283923, "entropy": 0.496598, "total_loss": 34.13452, "clip_frac": 0.075439, "sps": 416.5, "elapsed_s": 715.6}
+{"update": 144, "steps_done": 294912, "mean_step_reward": 0.943, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 60.771, "mean_ep_length": 64.5, "mean_collisions": 1.38, "mean_intent_rate": 0.56, "policy_loss": 0.008425, "value_loss": 82.61834, "entropy": 0.524619, "total_loss": 41.312348, "clip_frac": 0.223022, "sps": 415.1, "elapsed_s": 720.5}
+{"update": 145, "steps_done": 296960, "mean_step_reward": 1.0205, "min_step_reward": -10.5481, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 63.78, "mean_ep_length": 63.2, "mean_collisions": 2.03, "mean_intent_rate": 0.618, "policy_loss": 0.00841, "value_loss": 123.248631, "entropy": 0.531299, "total_loss": 61.627413, "clip_frac": 0.190552, "sps": 416.4, "elapsed_s": 725.4}
+{"update": 146, "steps_done": 299008, "mean_step_reward": 0.7856, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 52.553, "mean_ep_length": 65.6, "mean_collisions": 4.32, "mean_intent_rate": 0.667, "policy_loss": 0.011101, "value_loss": 141.412431, "entropy": 0.541301, "total_loss": 70.711903, "clip_frac": 0.213745, "sps": 415.4, "elapsed_s": 730.3}
+{"update": 147, "steps_done": 301056, "mean_step_reward": 0.8665, "min_step_reward": -10.58, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 53.488, "mean_ep_length": 60.9, "mean_collisions": 3.12, "mean_intent_rate": 0.642, "policy_loss": 0.0051, "value_loss": 96.615767, "entropy": 0.522564, "total_loss": 48.307757, "clip_frac": 0.136963, "sps": 407.4, "elapsed_s": 735.4}
+{"update": 148, "steps_done": 303104, "mean_step_reward": 0.8301, "min_step_reward": -10.5893, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 54.6, "mean_ep_length": 67.6, "mean_collisions": 3.27, "mean_intent_rate": 0.613, "policy_loss": 0.004461, "value_loss": 85.078667, "entropy": 0.539492, "total_loss": 42.5384, "clip_frac": 0.132446, "sps": 411.6, "elapsed_s": 740.3}
+{"update": 149, "steps_done": 305152, "mean_step_reward": 1.0099, "min_step_reward": -10.5724, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 64.833, "mean_ep_length": 64.2, "mean_collisions": 2.59, "mean_intent_rate": 0.649, "policy_loss": 0.000398, "value_loss": 92.985374, "entropy": 0.546817, "total_loss": 46.487616, "clip_frac": 0.053101, "sps": 410.9, "elapsed_s": 745.3}
+{"update": 150, "steps_done": 307200, "mean_step_reward": 1.0663, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 66.401, "mean_ep_length": 61.7, "mean_collisions": 1.24, "mean_intent_rate": 0.595, "policy_loss": 0.004926, "value_loss": 93.865206, "entropy": 0.534454, "total_loss": 46.932185, "clip_frac": 0.141113, "sps": 414.3, "elapsed_s": 750.3}
+{"update": 151, "steps_done": 309248, "mean_step_reward": 0.8513, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 53.677, "mean_ep_length": 62.8, "mean_collisions": 2.88, "mean_intent_rate": 0.621, "policy_loss": 0.008441, "value_loss": 93.901079, "entropy": 0.538645, "total_loss": 46.953594, "clip_frac": 0.130005, "sps": 412.7, "elapsed_s": 755.2}
+{"update": 152, "steps_done": 311296, "mean_step_reward": 0.9687, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 60.591, "mean_ep_length": 63.0, "mean_collisions": 3.33, "mean_intent_rate": 0.675, "policy_loss": 0.008063, "value_loss": 167.302428, "entropy": 0.543784, "total_loss": 83.653839, "clip_frac": 0.155396, "sps": 408.4, "elapsed_s": 760.3}
+{"update": 153, "steps_done": 313344, "mean_step_reward": 0.9233, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 53.862, "mean_ep_length": 58.3, "mean_collisions": 3.43, "mean_intent_rate": 0.68, "policy_loss": -0.000965, "value_loss": 126.399575, "entropy": 0.525069, "total_loss": 63.193571, "clip_frac": 0.086182, "sps": 412.5, "elapsed_s": 765.2}
+{"update": 154, "steps_done": 315392, "mean_step_reward": 0.9268, "min_step_reward": -10.5857, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 58.243, "mean_ep_length": 61.4, "mean_collisions": 3.59, "mean_intent_rate": 0.689, "policy_loss": -0.001707, "value_loss": 101.649659, "entropy": 0.521331, "total_loss": 50.817909, "clip_frac": 0.078857, "sps": 411.9, "elapsed_s": 770.2}
+{"update": 155, "steps_done": 317440, "mean_step_reward": 0.8753, "min_step_reward": -10.5843, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 51.836, "mean_ep_length": 61.3, "mean_collisions": 3.32, "mean_intent_rate": 0.639, "policy_loss": 0.000964, "value_loss": 117.673482, "entropy": 0.516139, "total_loss": 58.832543, "clip_frac": 0.088013, "sps": 415.4, "elapsed_s": 775.1}
+{"update": 156, "steps_done": 319488, "mean_step_reward": 0.8079, "min_step_reward": -10.543, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 52.863, "mean_ep_length": 62.9, "mean_collisions": 2.73, "mean_intent_rate": 0.6, "policy_loss": 0.003386, "value_loss": 75.79191, "entropy": 0.512176, "total_loss": 37.894219, "clip_frac": 0.081543, "sps": 406.6, "elapsed_s": 780.2}
+{"update": 157, "steps_done": 321536, "mean_step_reward": 0.9305, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 55.432, "mean_ep_length": 61.5, "mean_collisions": 2.36, "mean_intent_rate": 0.608, "policy_loss": 0.002816, "value_loss": 74.039216, "entropy": 0.529396, "total_loss": 37.01713, "clip_frac": 0.119995, "sps": 411.0, "elapsed_s": 785.2}
+{"update": 158, "steps_done": 323584, "mean_step_reward": 0.9496, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 60.993, "mean_ep_length": 63.5, "mean_collisions": 2.84, "mean_intent_rate": 0.647, "policy_loss": -0.000681, "value_loss": 90.788582, "entropy": 0.556476, "total_loss": 45.388045, "clip_frac": 0.06311, "sps": 409.9, "elapsed_s": 790.2}
+{"update": 159, "steps_done": 325632, "mean_step_reward": 0.8244, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 50.601, "mean_ep_length": 61.8, "mean_collisions": 3.7, "mean_intent_rate": 0.647, "policy_loss": 0.002571, "value_loss": 112.501733, "entropy": 0.534329, "total_loss": 56.248095, "clip_frac": 0.092896, "sps": 418.1, "elapsed_s": 795.1}
+{"update": 160, "steps_done": 327680, "mean_step_reward": 0.781, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 45.217, "mean_ep_length": 57.7, "mean_collisions": 3.67, "mean_intent_rate": 0.651, "policy_loss": 0.003897, "value_loss": 137.898416, "entropy": 0.537349, "total_loss": 68.947732, "clip_frac": 0.159302, "sps": 413.8, "elapsed_s": 800.0}
+{"update": 161, "steps_done": 329728, "mean_step_reward": 0.9249, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 53.863, "mean_ep_length": 58.0, "mean_collisions": 2.56, "mean_intent_rate": 0.633, "policy_loss": 0.003376, "value_loss": 98.478081, "entropy": 0.520979, "total_loss": 49.237207, "clip_frac": 0.107544, "sps": 412.7, "elapsed_s": 805.0}
+{"update": 162, "steps_done": 331776, "mean_step_reward": 0.9806, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 63.013, "mean_ep_length": 62.5, "mean_collisions": 2.47, "mean_intent_rate": 0.641, "policy_loss": -5e-06, "value_loss": 93.950574, "entropy": 0.532224, "total_loss": 46.96996, "clip_frac": 0.0802, "sps": 418.4, "elapsed_s": 809.9}
+{"update": 163, "steps_done": 333824, "mean_step_reward": 0.8453, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 46.632, "mean_ep_length": 57.4, "mean_collisions": 3.97, "mean_intent_rate": 0.687, "policy_loss": 0.000605, "value_loss": 121.049595, "entropy": 0.520583, "total_loss": 60.520197, "clip_frac": 0.095581, "sps": 413.6, "elapsed_s": 814.8}
+{"update": 164, "steps_done": 335872, "mean_step_reward": 0.9852, "min_step_reward": -10.5651, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 59.73, "mean_ep_length": 60.7, "mean_collisions": 3.41, "mean_intent_rate": 0.695, "policy_loss": 0.015931, "value_loss": 121.276868, "entropy": 0.523723, "total_loss": 60.649128, "clip_frac": 0.20459, "sps": 405.6, "elapsed_s": 819.9}
+{"update": 165, "steps_done": 337920, "mean_step_reward": 0.5638, "min_step_reward": -10.5853, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 34.568, "mean_ep_length": 61.1, "mean_collisions": 6.27, "mean_intent_rate": 0.707, "policy_loss": 0.020994, "value_loss": 166.491983, "entropy": 0.53662, "total_loss": 83.261618, "clip_frac": 0.261108, "sps": 414.8, "elapsed_s": 824.8}
+{"update": 166, "steps_done": 339968, "mean_step_reward": 0.7926, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 49.409, "mean_ep_length": 62.2, "mean_collisions": 4.76, "mean_intent_rate": 0.696, "policy_loss": -0.000572, "value_loss": 111.408563, "entropy": 0.562116, "total_loss": 55.698089, "clip_frac": 0.074463, "sps": 411.2, "elapsed_s": 829.8}
+{"update": 167, "steps_done": 342016, "mean_step_reward": 0.4609, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 24.752, "mean_ep_length": 57.6, "mean_collisions": 6.49, "mean_intent_rate": 0.694, "policy_loss": 0.019525, "value_loss": 255.1007, "entropy": 0.573681, "total_loss": 127.564138, "clip_frac": 0.313965, "sps": 413.5, "elapsed_s": 834.7}
+{"update": 168, "steps_done": 344064, "mean_step_reward": 0.543, "min_step_reward": -10.5863, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 35.796, "mean_ep_length": 61.6, "mean_collisions": 5.03, "mean_intent_rate": 0.649, "policy_loss": 0.033762, "value_loss": 144.334733, "entropy": 0.582645, "total_loss": 72.195302, "clip_frac": 0.33252, "sps": 413.8, "elapsed_s": 839.7}
+{"update": 169, "steps_done": 346112, "mean_step_reward": 0.4203, "min_step_reward": -10.5786, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 24.718, "mean_ep_length": 62.9, "mean_collisions": 7.31, "mean_intent_rate": 0.702, "policy_loss": 0.008676, "value_loss": 185.563345, "entropy": 0.590889, "total_loss": 92.78444, "clip_frac": 0.205322, "sps": 414.2, "elapsed_s": 844.6}
+{"update": 170, "steps_done": 348160, "mean_step_reward": 0.7762, "min_step_reward": -10.5833, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 48.975, "mean_ep_length": 62.0, "mean_collisions": 4.06, "mean_intent_rate": 0.664, "policy_loss": 0.006297, "value_loss": 83.912932, "entropy": 0.610637, "total_loss": 41.956657, "clip_frac": 0.115845, "sps": 414.7, "elapsed_s": 849.6}
+{"update": 171, "steps_done": 350208, "mean_step_reward": 0.7075, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 41.094, "mean_ep_length": 57.3, "mean_collisions": 4.23, "mean_intent_rate": 0.661, "policy_loss": 0.003846, "value_loss": 112.100288, "entropy": 0.606857, "total_loss": 56.047921, "clip_frac": 0.129639, "sps": 407.9, "elapsed_s": 854.6}
+{"update": 172, "steps_done": 352256, "mean_step_reward": 0.9439, "min_step_reward": -10.5866, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 58.545, "mean_ep_length": 63.0, "mean_collisions": 3.3, "mean_intent_rate": 0.665, "policy_loss": 0.025541, "value_loss": 85.330107, "entropy": 0.546302, "total_loss": 42.685131, "clip_frac": 0.283936, "sps": 414.4, "elapsed_s": 859.5}
+{"update": 173, "steps_done": 354304, "mean_step_reward": 0.8883, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 58.014, "mean_ep_length": 65.2, "mean_collisions": 3.58, "mean_intent_rate": 0.659, "policy_loss": 0.014958, "value_loss": 83.121318, "entropy": 0.578165, "total_loss": 41.569835, "clip_frac": 0.252075, "sps": 406.3, "elapsed_s": 864.6}
+{"update": 174, "steps_done": 356352, "mean_step_reward": 1.0004, "min_step_reward": -10.5817, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 59.872, "mean_ep_length": 60.9, "mean_collisions": 1.7, "mean_intent_rate": 0.595, "policy_loss": 0.013429, "value_loss": 75.461135, "entropy": 0.587891, "total_loss": 37.738117, "clip_frac": 0.251099, "sps": 412.2, "elapsed_s": 869.5}
+{"update": 175, "steps_done": 358400, "mean_step_reward": 0.9706, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 56.224, "mean_ep_length": 57.7, "mean_collisions": 2.33, "mean_intent_rate": 0.635, "policy_loss": 0.015456, "value_loss": 94.172741, "entropy": 0.553423, "total_loss": 47.096293, "clip_frac": 0.186401, "sps": 409.3, "elapsed_s": 874.5}
+{"update": 176, "steps_done": 360448, "mean_step_reward": 1.0889, "min_step_reward": -10.5465, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 71.991, "mean_ep_length": 65.5, "mean_collisions": 3.22, "mean_intent_rate": 0.709, "policy_loss": 0.001923, "value_loss": 107.815981, "entropy": 0.553619, "total_loss": 53.904378, "clip_frac": 0.048584, "sps": 413.3, "elapsed_s": 879.5}
+{"update": 177, "steps_done": 362496, "mean_step_reward": 0.8763, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.847, "mean_ep_length": 63.2, "mean_collisions": 4.28, "mean_intent_rate": 0.697, "policy_loss": 0.007633, "value_loss": 120.927992, "entropy": 0.572708, "total_loss": 60.465902, "clip_frac": 0.137451, "sps": 411.6, "elapsed_s": 884.5}
+{"update": 178, "steps_done": 364544, "mean_step_reward": 1.0085, "min_step_reward": -10.5673, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 62.105, "mean_ep_length": 61.8, "mean_collisions": 2.79, "mean_intent_rate": 0.667, "policy_loss": 0.017898, "value_loss": 101.869164, "entropy": 0.570596, "total_loss": 50.946774, "clip_frac": 0.176392, "sps": 409.7, "elapsed_s": 889.5}
+{"update": 179, "steps_done": 366592, "mean_step_reward": 0.8453, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 53.416, "mean_ep_length": 62.7, "mean_collisions": 3.64, "mean_intent_rate": 0.648, "policy_loss": 0.008603, "value_loss": 88.318874, "entropy": 0.543321, "total_loss": 44.162607, "clip_frac": 0.187134, "sps": 413.2, "elapsed_s": 894.4}
+{"update": 180, "steps_done": 368640, "mean_step_reward": 0.9381, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 62.117, "mean_ep_length": 63.2, "mean_collisions": 3.12, "mean_intent_rate": 0.674, "policy_loss": 0.001528, "value_loss": 117.20766, "entropy": 0.571541, "total_loss": 58.599643, "clip_frac": 0.139282, "sps": 413.3, "elapsed_s": 899.4}
+{"update": 181, "steps_done": 370688, "mean_step_reward": 1.0167, "min_step_reward": -10.5897, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 58.42, "mean_ep_length": 59.7, "mean_collisions": 3.37, "mean_intent_rate": 0.694, "policy_loss": 0.0072, "value_loss": 92.069441, "entropy": 0.530817, "total_loss": 46.036613, "clip_frac": 0.167969, "sps": 414.0, "elapsed_s": 904.3}
+{"update": 182, "steps_done": 372736, "mean_step_reward": 0.9842, "min_step_reward": -10.5837, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 62.723, "mean_ep_length": 64.7, "mean_collisions": 3.61, "mean_intent_rate": 0.685, "policy_loss": 0.015778, "value_loss": 121.302688, "entropy": 0.524296, "total_loss": 60.66188, "clip_frac": 0.160034, "sps": 402.7, "elapsed_s": 909.4}
+{"update": 183, "steps_done": 374784, "mean_step_reward": 0.8197, "min_step_reward": -10.5808, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.362, "mean_ep_length": 62.1, "mean_collisions": 4.42, "mean_intent_rate": 0.691, "policy_loss": 0.009903, "value_loss": 180.208762, "entropy": 0.502833, "total_loss": 90.109256, "clip_frac": 0.260498, "sps": 414.8, "elapsed_s": 914.4}
+{"update": 184, "steps_done": 376832, "mean_step_reward": 0.8421, "min_step_reward": -10.5532, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 51.046, "mean_ep_length": 60.6, "mean_collisions": 3.12, "mean_intent_rate": 0.637, "policy_loss": 0.013366, "value_loss": 104.011934, "entropy": 0.528335, "total_loss": 52.014049, "clip_frac": 0.175049, "sps": 417.4, "elapsed_s": 919.3}
+{"update": 185, "steps_done": 378880, "mean_step_reward": 0.7774, "min_step_reward": -10.5816, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 46.355, "mean_ep_length": 58.6, "mean_collisions": 4.46, "mean_intent_rate": 0.702, "policy_loss": 0.048724, "value_loss": 130.771991, "entropy": 0.56127, "total_loss": 65.429106, "clip_frac": 0.370117, "sps": 410.4, "elapsed_s": 924.3}
+{"update": 186, "steps_done": 380928, "mean_step_reward": -0.5054, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": -29.947, "mean_ep_length": 58.7, "mean_collisions": 13.17, "mean_intent_rate": 0.76, "policy_loss": 0.011297, "value_loss": 241.974168, "entropy": 0.548228, "total_loss": 120.992899, "clip_frac": 0.123047, "sps": 414.2, "elapsed_s": 929.2}
+{"update": 187, "steps_done": 382976, "mean_step_reward": -0.1484, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": -6.06, "mean_ep_length": 59.6, "mean_collisions": 11.41, "mean_intent_rate": 0.782, "policy_loss": -0.000593, "value_loss": 214.421228, "entropy": 0.545939, "total_loss": 107.204562, "clip_frac": 0.170898, "sps": 413.7, "elapsed_s": 934.2}
+{"update": 188, "steps_done": 385024, "mean_step_reward": 0.192, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 10.322, "mean_ep_length": 61.9, "mean_collisions": 10.06, "mean_intent_rate": 0.773, "policy_loss": -0.002068, "value_loss": 212.812533, "entropy": 0.544202, "total_loss": 106.398758, "clip_frac": 0.084351, "sps": 417.9, "elapsed_s": 939.1}
+{"update": 189, "steps_done": 387072, "mean_step_reward": -0.0632, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": -3.934, "mean_ep_length": 57.7, "mean_collisions": 9.94, "mean_intent_rate": 0.732, "policy_loss": -0.000627, "value_loss": 222.033523, "entropy": 0.585031, "total_loss": 111.010285, "clip_frac": 0.163452, "sps": 415.4, "elapsed_s": 944.0}
+{"update": 190, "steps_done": 389120, "mean_step_reward": 0.1418, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 8.772, "mean_ep_length": 64.8, "mean_collisions": 9.94, "mean_intent_rate": 0.737, "policy_loss": 0.002452, "value_loss": 207.962427, "entropy": 0.572645, "total_loss": 103.977939, "clip_frac": 0.078613, "sps": 409.2, "elapsed_s": 949.0}
+{"update": 191, "steps_done": 391168, "mean_step_reward": 0.4816, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 30.511, "mean_ep_length": 64.7, "mean_collisions": 6.94, "mean_intent_rate": 0.697, "policy_loss": 0.012265, "value_loss": 132.356225, "entropy": 0.564731, "total_loss": 66.18473, "clip_frac": 0.108765, "sps": 415.4, "elapsed_s": 953.9}
+{"update": 192, "steps_done": 393216, "mean_step_reward": 0.3644, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 21.118, "mean_ep_length": 62.9, "mean_collisions": 8.38, "mean_intent_rate": 0.738, "policy_loss": -0.000892, "value_loss": 152.759934, "entropy": 0.543001, "total_loss": 76.373645, "clip_frac": 0.10144, "sps": 412.7, "elapsed_s": 958.9}
+{"update": 193, "steps_done": 395264, "mean_step_reward": 0.8504, "min_step_reward": -10.5774, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 51.876, "mean_ep_length": 60.4, "mean_collisions": 4.62, "mean_intent_rate": 0.724, "policy_loss": 0.022172, "value_loss": 104.121435, "entropy": 0.541025, "total_loss": 52.077479, "clip_frac": 0.224976, "sps": 411.7, "elapsed_s": 963.9}
+{"update": 194, "steps_done": 397312, "mean_step_reward": 0.5528, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 34.594, "mean_ep_length": 61.5, "mean_collisions": 6.76, "mean_intent_rate": 0.738, "policy_loss": 0.004258, "value_loss": 131.186365, "entropy": 0.55318, "total_loss": 65.591909, "clip_frac": 0.115479, "sps": 409.7, "elapsed_s": 968.9}
+{"update": 195, "steps_done": 399360, "mean_step_reward": 0.4287, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 27.14, "mean_ep_length": 63.2, "mean_collisions": 7.53, "mean_intent_rate": 0.712, "policy_loss": -0.002152, "value_loss": 214.87842, "entropy": 0.532155, "total_loss": 107.431738, "clip_frac": 0.177612, "sps": 413.6, "elapsed_s": 973.8}
+{"update": 196, "steps_done": 401408, "mean_step_reward": 0.7973, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 47.254, "mean_ep_length": 60.1, "mean_collisions": 4.79, "mean_intent_rate": 0.698, "policy_loss": 0.005323, "value_loss": 130.342989, "entropy": 0.510674, "total_loss": 65.171711, "clip_frac": 0.104614, "sps": 417.8, "elapsed_s": 978.7}
+{"update": 197, "steps_done": 403456, "mean_step_reward": 0.8145, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 53.987, "mean_ep_length": 65.5, "mean_collisions": 4.81, "mean_intent_rate": 0.697, "policy_loss": 0.004417, "value_loss": 107.03486, "entropy": 0.509909, "total_loss": 53.516748, "clip_frac": 0.121948, "sps": 416.0, "elapsed_s": 983.6}
+{"update": 198, "steps_done": 405504, "mean_step_reward": 0.7852, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 45.474, "mean_ep_length": 58.3, "mean_collisions": 5.03, "mean_intent_rate": 0.717, "policy_loss": 0.007509, "value_loss": 173.516929, "entropy": 0.502048, "total_loss": 86.760953, "clip_frac": 0.143188, "sps": 413.6, "elapsed_s": 988.6}
+{"update": 199, "steps_done": 407552, "mean_step_reward": 0.8258, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 53.046, "mean_ep_length": 63.3, "mean_collisions": 5.16, "mean_intent_rate": 0.726, "policy_loss": 0.003196, "value_loss": 130.561013, "entropy": 0.470804, "total_loss": 65.278995, "clip_frac": 0.086792, "sps": 406.3, "elapsed_s": 993.6}
+{"update": 200, "steps_done": 409600, "mean_step_reward": 1.1572, "min_step_reward": -10.5786, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 70.491, "mean_ep_length": 61.3, "mean_collisions": 3.06, "mean_intent_rate": 0.729, "policy_loss": 0.011883, "value_loss": 100.59446, "entropy": 0.478737, "total_loss": 50.304326, "clip_frac": 0.151245, "sps": 412.6, "elapsed_s": 998.6}
+{"update": 201, "steps_done": 411648, "mean_step_reward": 0.6692, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.614, "mean_ep_length": 60.9, "mean_collisions": 5.53, "mean_intent_rate": 0.701, "policy_loss": 0.008565, "value_loss": 188.131862, "entropy": 0.507363, "total_loss": 94.069422, "clip_frac": 0.146484, "sps": 414.5, "elapsed_s": 1003.6}
+{"update": 202, "steps_done": 413696, "mean_step_reward": 1.1776, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 69.454, "mean_ep_length": 59.1, "mean_collisions": 1.77, "mean_intent_rate": 0.667, "policy_loss": 0.028347, "value_loss": 90.911636, "entropy": 0.501734, "total_loss": 45.479148, "clip_frac": 0.221313, "sps": 410.7, "elapsed_s": 1008.5}
+{"update": 203, "steps_done": 415744, "mean_step_reward": 0.895, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 52.755, "mean_ep_length": 57.5, "mean_collisions": 3.69, "mean_intent_rate": 0.694, "policy_loss": 0.006747, "value_loss": 109.674822, "entropy": 0.516628, "total_loss": 54.838991, "clip_frac": 0.147827, "sps": 413.2, "elapsed_s": 1013.5}
+{"update": 204, "steps_done": 417792, "mean_step_reward": 0.8774, "min_step_reward": -10.5865, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.521, "mean_ep_length": 61.3, "mean_collisions": 4.41, "mean_intent_rate": 0.706, "policy_loss": 0.010472, "value_loss": 113.63434, "entropy": 0.505556, "total_loss": 56.822586, "clip_frac": 0.168701, "sps": 417.4, "elapsed_s": 1018.4}
+{"update": 205, "steps_done": 419840, "mean_step_reward": 0.7889, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 48.364, "mean_ep_length": 61.8, "mean_collisions": 4.42, "mean_intent_rate": 0.679, "policy_loss": -0.003622, "value_loss": 123.227998, "entropy": 0.513324, "total_loss": 61.605244, "clip_frac": 0.105957, "sps": 417.8, "elapsed_s": 1023.3}
+{"update": 206, "steps_done": 421888, "mean_step_reward": 0.977, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 59.37, "mean_ep_length": 60.4, "mean_collisions": 2.62, "mean_intent_rate": 0.652, "policy_loss": 0.029479, "value_loss": 141.183634, "entropy": 0.5315, "total_loss": 70.615981, "clip_frac": 0.275269, "sps": 419.6, "elapsed_s": 1028.2}
+{"update": 207, "steps_done": 423936, "mean_step_reward": 1.0091, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 61.395, "mean_ep_length": 60.0, "mean_collisions": 2.88, "mean_intent_rate": 0.689, "policy_loss": -0.000464, "value_loss": 97.69312, "entropy": 0.488201, "total_loss": 48.841214, "clip_frac": 0.120605, "sps": 405.5, "elapsed_s": 1033.2}
+{"update": 208, "steps_done": 425984, "mean_step_reward": 0.8112, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 48.744, "mean_ep_length": 60.8, "mean_collisions": 4.15, "mean_intent_rate": 0.667, "policy_loss": 0.012273, "value_loss": 103.893926, "entropy": 0.459511, "total_loss": 51.954641, "clip_frac": 0.119995, "sps": 412.5, "elapsed_s": 1038.2}
+{"update": 209, "steps_done": 428032, "mean_step_reward": 0.949, "min_step_reward": -10.5372, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 56.73, "mean_ep_length": 59.9, "mean_collisions": 2.97, "mean_intent_rate": 0.661, "policy_loss": 0.001803, "value_loss": 106.268231, "entropy": 0.486163, "total_loss": 53.131056, "clip_frac": 0.087158, "sps": 409.6, "elapsed_s": 1043.2}
+{"update": 210, "steps_done": 430080, "mean_step_reward": 0.8533, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 53.658, "mean_ep_length": 63.5, "mean_collisions": 4.16, "mean_intent_rate": 0.673, "policy_loss": 0.024509, "value_loss": 148.21394, "entropy": 0.510696, "total_loss": 74.126371, "clip_frac": 0.255493, "sps": 414.4, "elapsed_s": 1048.2}
+{"update": 211, "steps_done": 432128, "mean_step_reward": 0.7932, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 53.008, "mean_ep_length": 66.1, "mean_collisions": 5.9, "mean_intent_rate": 0.743, "policy_loss": -0.001274, "value_loss": 149.196663, "entropy": 0.483815, "total_loss": 74.59222, "clip_frac": 0.119751, "sps": 411.1, "elapsed_s": 1053.1}
+{"update": 212, "steps_done": 434176, "mean_step_reward": 0.8434, "min_step_reward": -10.5562, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 54.822, "mean_ep_length": 65.6, "mean_collisions": 4.26, "mean_intent_rate": 0.677, "policy_loss": 0.002085, "value_loss": 99.047057, "entropy": 0.484064, "total_loss": 49.520773, "clip_frac": 0.078003, "sps": 409.5, "elapsed_s": 1058.1}
+{"update": 213, "steps_done": 436224, "mean_step_reward": 0.8635, "min_step_reward": -10.5768, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 53.329, "mean_ep_length": 60.9, "mean_collisions": 3.74, "mean_intent_rate": 0.673, "policy_loss": -5.9e-05, "value_loss": 106.111497, "entropy": 0.498623, "total_loss": 53.050703, "clip_frac": 0.10022, "sps": 416.4, "elapsed_s": 1063.1}
+{"update": 214, "steps_done": 438272, "mean_step_reward": 1.0829, "min_step_reward": -10.5356, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 68.899, "mean_ep_length": 63.7, "mean_collisions": 2.06, "mean_intent_rate": 0.65, "policy_loss": 0.014932, "value_loss": 98.260576, "entropy": 0.520966, "total_loss": 49.140011, "clip_frac": 0.177246, "sps": 414.0, "elapsed_s": 1068.0}
+{"update": 215, "steps_done": 440320, "mean_step_reward": 0.811, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 51.095, "mean_ep_length": 63.6, "mean_collisions": 4.16, "mean_intent_rate": 0.664, "policy_loss": 0.0141, "value_loss": 105.505203, "entropy": 0.519386, "total_loss": 52.761508, "clip_frac": 0.200806, "sps": 416.5, "elapsed_s": 1072.9}
+{"update": 216, "steps_done": 442368, "mean_step_reward": 0.7574, "min_step_reward": -10.5873, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 43.491, "mean_ep_length": 56.7, "mean_collisions": 5.54, "mean_intent_rate": 0.765, "policy_loss": 0.005451, "value_loss": 148.668702, "entropy": 0.477069, "total_loss": 74.335032, "clip_frac": 0.102539, "sps": 405.7, "elapsed_s": 1078.0}
+{"update": 217, "steps_done": 444416, "mean_step_reward": 0.8309, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 48.055, "mean_ep_length": 57.7, "mean_collisions": 4.34, "mean_intent_rate": 0.707, "policy_loss": 0.016302, "value_loss": 131.488201, "entropy": 0.464081, "total_loss": 65.755762, "clip_frac": 0.210693, "sps": 417.8, "elapsed_s": 1082.9}
+{"update": 218, "steps_done": 446464, "mean_step_reward": 1.1596, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 74.682, "mean_ep_length": 64.9, "mean_collisions": 1.88, "mean_intent_rate": 0.659, "policy_loss": 0.014793, "value_loss": 98.66362, "entropy": 0.491559, "total_loss": 49.341688, "clip_frac": 0.182617, "sps": 413.3, "elapsed_s": 1087.8}
+{"update": 219, "steps_done": 448512, "mean_step_reward": 1.2518, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 81.463, "mean_ep_length": 65.3, "mean_collisions": 1.71, "mean_intent_rate": 0.68, "policy_loss": 0.016296, "value_loss": 110.126376, "entropy": 0.495721, "total_loss": 55.074527, "clip_frac": 0.187622, "sps": 407.5, "elapsed_s": 1092.9}
+{"update": 220, "steps_done": 450560, "mean_step_reward": 1.1403, "min_step_reward": -10.5006, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 74.975, "mean_ep_length": 64.3, "mean_collisions": 1.75, "mean_intent_rate": 0.659, "policy_loss": 0.034382, "value_loss": 110.386158, "entropy": 0.49026, "total_loss": 55.222559, "clip_frac": 0.275879, "sps": 409.8, "elapsed_s": 1097.9}
+{"update": 221, "steps_done": 452608, "mean_step_reward": 0.8578, "min_step_reward": -10.5833, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 50.609, "mean_ep_length": 60.0, "mean_collisions": 3.85, "mean_intent_rate": 0.664, "policy_loss": 0.010387, "value_loss": 125.502386, "entropy": 0.470771, "total_loss": 62.756873, "clip_frac": 0.197632, "sps": 414.9, "elapsed_s": 1102.8}
+{"update": 222, "steps_done": 454656, "mean_step_reward": 0.7922, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 48.467, "mean_ep_length": 63.2, "mean_collisions": 4.44, "mean_intent_rate": 0.679, "policy_loss": 0.021342, "value_loss": 150.999315, "entropy": 0.470922, "total_loss": 75.51629, "clip_frac": 0.160278, "sps": 410.7, "elapsed_s": 1107.8}
+{"update": 223, "steps_done": 456704, "mean_step_reward": 1.0306, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 70.089, "mean_ep_length": 64.5, "mean_collisions": 3.38, "mean_intent_rate": 0.717, "policy_loss": 0.001701, "value_loss": 113.803046, "entropy": 0.453717, "total_loss": 56.898688, "clip_frac": 0.098022, "sps": 414.6, "elapsed_s": 1112.7}
+{"update": 224, "steps_done": 458752, "mean_step_reward": 1.0938, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 63.152, "mean_ep_length": 61.3, "mean_collisions": 2.85, "mean_intent_rate": 0.678, "policy_loss": 0.008384, "value_loss": 107.058441, "entropy": 0.445091, "total_loss": 53.533153, "clip_frac": 0.121826, "sps": 402.8, "elapsed_s": 1117.8}
+{"update": 225, "steps_done": 460800, "mean_step_reward": 0.7988, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 51.378, "mean_ep_length": 61.9, "mean_collisions": 4.32, "mean_intent_rate": 0.691, "policy_loss": 0.007698, "value_loss": 150.085973, "entropy": 0.451051, "total_loss": 75.046175, "clip_frac": 0.163696, "sps": 413.8, "elapsed_s": 1122.8}
+{"update": 226, "steps_done": 462848, "mean_step_reward": 0.7837, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 48.574, "mean_ep_length": 61.8, "mean_collisions": 4.18, "mean_intent_rate": 0.658, "policy_loss": 0.000639, "value_loss": 137.149853, "entropy": 0.464492, "total_loss": 68.57092, "clip_frac": 0.117432, "sps": 413.3, "elapsed_s": 1127.7}
+{"update": 227, "steps_done": 464896, "mean_step_reward": 1.0008, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 60.003, "mean_ep_length": 58.6, "mean_collisions": 3.18, "mean_intent_rate": 0.701, "policy_loss": 0.046339, "value_loss": 140.977299, "entropy": 0.480888, "total_loss": 70.53018, "clip_frac": 0.313477, "sps": 411.5, "elapsed_s": 1132.7}
+{"update": 228, "steps_done": 466944, "mean_step_reward": 1.0076, "min_step_reward": -10.5876, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 64.696, "mean_ep_length": 65.8, "mean_collisions": 4.16, "mean_intent_rate": 0.712, "policy_loss": 0.007314, "value_loss": 142.690899, "entropy": 0.485755, "total_loss": 71.347906, "clip_frac": 0.155151, "sps": 415.3, "elapsed_s": 1137.6}
+{"update": 229, "steps_done": 468992, "mean_step_reward": 0.7763, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 48.559, "mean_ep_length": 59.0, "mean_collisions": 4.88, "mean_intent_rate": 0.735, "policy_loss": 0.001342, "value_loss": 136.065462, "entropy": 0.462287, "total_loss": 68.029451, "clip_frac": 0.11731, "sps": 415.2, "elapsed_s": 1142.6}
+{"update": 230, "steps_done": 471040, "mean_step_reward": 0.9365, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 51.256, "mean_ep_length": 57.9, "mean_collisions": 4.36, "mean_intent_rate": 0.727, "policy_loss": 0.002151, "value_loss": 112.19657, "entropy": 0.460723, "total_loss": 56.095829, "clip_frac": 0.09436, "sps": 409.9, "elapsed_s": 1147.6}
+{"update": 231, "steps_done": 473088, "mean_step_reward": 0.8709, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.114, "mean_ep_length": 61.9, "mean_collisions": 4.44, "mean_intent_rate": 0.704, "policy_loss": 0.008418, "value_loss": 131.515776, "entropy": 0.44678, "total_loss": 65.761838, "clip_frac": 0.118652, "sps": 409.3, "elapsed_s": 1152.6}
+{"update": 232, "steps_done": 475136, "mean_step_reward": 0.7587, "min_step_reward": -10.5807, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 46.248, "mean_ep_length": 59.7, "mean_collisions": 5.26, "mean_intent_rate": 0.725, "policy_loss": 0.011081, "value_loss": 191.308009, "entropy": 0.423363, "total_loss": 95.660852, "clip_frac": 0.192627, "sps": 418.6, "elapsed_s": 1157.5}
+{"update": 233, "steps_done": 477184, "mean_step_reward": 0.9051, "min_step_reward": -10.5519, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 51.702, "mean_ep_length": 58.6, "mean_collisions": 3.24, "mean_intent_rate": 0.663, "policy_loss": 0.008003, "value_loss": 134.055641, "entropy": 0.438398, "total_loss": 67.03144, "clip_frac": 0.120361, "sps": 407.1, "elapsed_s": 1162.5}
+{"update": 234, "steps_done": 479232, "mean_step_reward": 1.02, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 59.087, "mean_ep_length": 57.6, "mean_collisions": 1.84, "mean_intent_rate": 0.619, "policy_loss": 0.038252, "value_loss": 96.154051, "entropy": 0.481019, "total_loss": 48.110468, "clip_frac": 0.299316, "sps": 419.6, "elapsed_s": 1167.4}
+{"update": 235, "steps_done": 481280, "mean_step_reward": 1.0769, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 62.241, "mean_ep_length": 57.9, "mean_collisions": 2.51, "mean_intent_rate": 0.684, "policy_loss": 0.030083, "value_loss": 91.303319, "entropy": 0.494004, "total_loss": 45.676803, "clip_frac": 0.282227, "sps": 415.4, "elapsed_s": 1172.3}
+{"update": 236, "steps_done": 483328, "mean_step_reward": 1.0709, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 64.406, "mean_ep_length": 59.9, "mean_collisions": 2.03, "mean_intent_rate": 0.644, "policy_loss": 0.025828, "value_loss": 89.390974, "entropy": 0.469524, "total_loss": 44.71662, "clip_frac": 0.333496, "sps": 416.3, "elapsed_s": 1177.2}
+{"update": 237, "steps_done": 485376, "mean_step_reward": 1.1467, "min_step_reward": -10.5468, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 66.051, "mean_ep_length": 58.4, "mean_collisions": 0.86, "mean_intent_rate": 0.595, "policy_loss": 0.009606, "value_loss": 88.888576, "entropy": 0.478681, "total_loss": 44.449107, "clip_frac": 0.159668, "sps": 412.6, "elapsed_s": 1182.2}
+{"update": 238, "steps_done": 487424, "mean_step_reward": 0.8967, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 59.45, "mean_ep_length": 66.1, "mean_collisions": 2.81, "mean_intent_rate": 0.629, "policy_loss": 0.008345, "value_loss": 126.556864, "entropy": 0.44536, "total_loss": 63.282324, "clip_frac": 0.149048, "sps": 414.5, "elapsed_s": 1187.1}
+{"update": 239, "steps_done": 489472, "mean_step_reward": 1.1519, "min_step_reward": -10.5707, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 75.146, "mean_ep_length": 64.8, "mean_collisions": 1.31, "mean_intent_rate": 0.622, "policy_loss": 0.007813, "value_loss": 108.373301, "entropy": 0.446162, "total_loss": 54.190002, "clip_frac": 0.125244, "sps": 415.6, "elapsed_s": 1192.1}
+{"update": 240, "steps_done": 491520, "mean_step_reward": 1.1715, "min_step_reward": -10.5853, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 78.798, "mean_ep_length": 67.0, "mean_collisions": 2.1, "mean_intent_rate": 0.672, "policy_loss": 0.001427, "value_loss": 121.70353, "entropy": 0.439472, "total_loss": 60.848798, "clip_frac": 0.073608, "sps": 416.3, "elapsed_s": 1197.0}
+{"update": 241, "steps_done": 493568, "mean_step_reward": 0.604, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 34.549, "mean_ep_length": 59.4, "mean_collisions": 4.82, "mean_intent_rate": 0.638, "policy_loss": 0.024023, "value_loss": 171.488569, "entropy": 0.454125, "total_loss": 85.763766, "clip_frac": 0.26062, "sps": 403.8, "elapsed_s": 1202.0}
+{"update": 242, "steps_done": 495616, "mean_step_reward": 1.0534, "min_step_reward": -10.5239, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 66.558, "mean_ep_length": 62.5, "mean_collisions": 2.88, "mean_intent_rate": 0.686, "policy_loss": 0.003215, "value_loss": 137.463912, "entropy": 0.45798, "total_loss": 68.730592, "clip_frac": 0.070923, "sps": 411.5, "elapsed_s": 1207.0}
+{"update": 243, "steps_done": 497664, "mean_step_reward": 1.2314, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 73.686, "mean_ep_length": 58.9, "mean_collisions": 1.76, "mean_intent_rate": 0.692, "policy_loss": 0.008846, "value_loss": 102.766987, "entropy": 0.460351, "total_loss": 51.387735, "clip_frac": 0.092407, "sps": 418.1, "elapsed_s": 1211.9}
+{"update": 244, "steps_done": 499712, "mean_step_reward": 0.9974, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 62.669, "mean_ep_length": 64.3, "mean_collisions": 2.5, "mean_intent_rate": 0.63, "policy_loss": 0.018776, "value_loss": 99.612359, "entropy": 0.425934, "total_loss": 49.820695, "clip_frac": 0.195679, "sps": 412.8, "elapsed_s": 1216.9}
+{"update": 245, "steps_done": 501760, "mean_step_reward": 1.1735, "min_step_reward": -10.5834, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 75.445, "mean_ep_length": 64.0, "mean_collisions": 1.94, "mean_intent_rate": 0.667, "policy_loss": 0.019758, "value_loss": 115.652787, "entropy": 0.433187, "total_loss": 57.841819, "clip_frac": 0.166748, "sps": 411.6, "elapsed_s": 1221.9}
+{"update": 246, "steps_done": 503808, "mean_step_reward": 1.0734, "min_step_reward": -10.5594, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 63.748, "mean_ep_length": 59.2, "mean_collisions": 1.71, "mean_intent_rate": 0.639, "policy_loss": 0.004598, "value_loss": 117.063734, "entropy": 0.41159, "total_loss": 58.53235, "clip_frac": 0.102051, "sps": 411.1, "elapsed_s": 1226.8}
+{"update": 247, "steps_done": 505856, "mean_step_reward": 1.0189, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 65.453, "mean_ep_length": 64.2, "mean_collisions": 2.34, "mean_intent_rate": 0.634, "policy_loss": 0.003911, "value_loss": 114.73146, "entropy": 0.462766, "total_loss": 57.365013, "clip_frac": 0.14917, "sps": 414.3, "elapsed_s": 1231.8}
+{"update": 248, "steps_done": 507904, "mean_step_reward": 1.1538, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 76.153, "mean_ep_length": 65.8, "mean_collisions": 1.19, "mean_intent_rate": 0.621, "policy_loss": 0.026073, "value_loss": 93.944629, "entropy": 0.488243, "total_loss": 46.993505, "clip_frac": 0.271973, "sps": 406.3, "elapsed_s": 1236.8}
+{"update": 249, "steps_done": 509952, "mean_step_reward": 1.2323, "min_step_reward": -10.5799, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 80.534, "mean_ep_length": 65.5, "mean_collisions": 1.32, "mean_intent_rate": 0.651, "policy_loss": 0.007158, "value_loss": 104.027968, "entropy": 0.453746, "total_loss": 52.016604, "clip_frac": 0.159546, "sps": 415.1, "elapsed_s": 1241.8}
+{"update": 250, "steps_done": 512000, "mean_step_reward": 0.8801, "min_step_reward": -7.5731, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 56.989, "mean_ep_length": 63.2, "mean_collisions": 0.47, "mean_intent_rate": 0.496, "policy_loss": 0.00584, "value_loss": 84.431867, "entropy": 0.438118, "total_loss": 42.217393, "clip_frac": 0.117798, "sps": 400.0, "elapsed_s": 1246.9}
+{"update": 251, "steps_done": 514048, "mean_step_reward": 0.9046, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.593, "mean_ep_length": 60.3, "mean_collisions": 3.03, "mean_intent_rate": 0.636, "policy_loss": 0.005623, "value_loss": 137.301196, "entropy": 0.4718, "total_loss": 68.651504, "clip_frac": 0.183105, "sps": 409.6, "elapsed_s": 1251.9}
+{"update": 252, "steps_done": 516096, "mean_step_reward": 1.0194, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 64.983, "mean_ep_length": 62.8, "mean_collisions": 2.5, "mean_intent_rate": 0.649, "policy_loss": 0.002015, "value_loss": 130.299526, "entropy": 0.448146, "total_loss": 65.147297, "clip_frac": 0.061035, "sps": 411.0, "elapsed_s": 1256.9}
+{"update": 253, "steps_done": 518144, "mean_step_reward": 1.1351, "min_step_reward": -10.5846, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 72.398, "mean_ep_length": 64.5, "mean_collisions": 1.42, "mean_intent_rate": 0.622, "policy_loss": 0.023524, "value_loss": 112.135085, "entropy": 0.452725, "total_loss": 56.086539, "clip_frac": 0.227905, "sps": 417.1, "elapsed_s": 1261.8}
+{"update": 254, "steps_done": 520192, "mean_step_reward": 0.6663, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 42.282, "mean_ep_length": 61.6, "mean_collisions": 5.81, "mean_intent_rate": 0.717, "policy_loss": 0.051314, "value_loss": 174.172805, "entropy": 0.421174, "total_loss": 87.133506, "clip_frac": 0.303833, "sps": 418.1, "elapsed_s": 1266.7}
+{"update": 255, "steps_done": 522240, "mean_step_reward": 0.7566, "min_step_reward": -10.5798, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 46.942, "mean_ep_length": 65.1, "mean_collisions": 5.69, "mean_intent_rate": 0.713, "policy_loss": 0.054343, "value_loss": 161.226832, "entropy": 0.438187, "total_loss": 80.663378, "clip_frac": 0.335571, "sps": 416.0, "elapsed_s": 1271.6}
+{"update": 256, "steps_done": 524288, "mean_step_reward": 0.8435, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 53.217, "mean_ep_length": 62.9, "mean_collisions": 5.76, "mean_intent_rate": 0.772, "policy_loss": 0.009862, "value_loss": 138.860816, "entropy": 0.424169, "total_loss": 69.436028, "clip_frac": 0.192627, "sps": 411.7, "elapsed_s": 1276.6}
+{"update": 257, "steps_done": 526336, "mean_step_reward": 0.415, "min_step_reward": -10.5849, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 29.404, "mean_ep_length": 66.1, "mean_collisions": 9.1, "mean_intent_rate": 0.788, "policy_loss": 0.000783, "value_loss": 200.704134, "entropy": 0.430699, "total_loss": 100.348544, "clip_frac": 0.143799, "sps": 411.7, "elapsed_s": 1281.6}
+{"update": 258, "steps_done": 528384, "mean_step_reward": 0.8074, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 50.083, "mean_ep_length": 64.6, "mean_collisions": 5.97, "mean_intent_rate": 0.745, "policy_loss": 0.000495, "value_loss": 170.192139, "entropy": 0.426815, "total_loss": 85.092297, "clip_frac": 0.133057, "sps": 406.3, "elapsed_s": 1286.6}
+{"update": 259, "steps_done": 530432, "mean_step_reward": 0.7851, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 53.306, "mean_ep_length": 66.1, "mean_collisions": 5.62, "mean_intent_rate": 0.739, "policy_loss": 0.007938, "value_loss": 142.40986, "entropy": 0.443477, "total_loss": 71.208434, "clip_frac": 0.201416, "sps": 414.5, "elapsed_s": 1291.5}
+{"update": 260, "steps_done": 532480, "mean_step_reward": 1.0333, "min_step_reward": -10.5877, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 61.735, "mean_ep_length": 59.4, "mean_collisions": 2.88, "mean_intent_rate": 0.691, "policy_loss": 0.00772, "value_loss": 107.007582, "entropy": 0.465643, "total_loss": 53.506855, "clip_frac": 0.129517, "sps": 414.5, "elapsed_s": 1296.5}
+{"update": 261, "steps_done": 534528, "mean_step_reward": 0.9942, "min_step_reward": -10.5793, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 65.094, "mean_ep_length": 65.4, "mean_collisions": 3.61, "mean_intent_rate": 0.696, "policy_loss": 0.022413, "value_loss": 89.068646, "entropy": 0.459829, "total_loss": 44.552137, "clip_frac": 0.192749, "sps": 413.4, "elapsed_s": 1301.4}
+{"update": 262, "steps_done": 536576, "mean_step_reward": 0.6406, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 39.48, "mean_ep_length": 61.7, "mean_collisions": 7.12, "mean_intent_rate": 0.774, "policy_loss": 0.000415, "value_loss": 262.751716, "entropy": 0.429104, "total_loss": 131.371982, "clip_frac": 0.111572, "sps": 422.2, "elapsed_s": 1306.3}
+{"update": 263, "steps_done": 538624, "mean_step_reward": 0.661, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 40.206, "mean_ep_length": 63.0, "mean_collisions": 6.84, "mean_intent_rate": 0.753, "policy_loss": 0.050306, "value_loss": 148.514251, "entropy": 0.440341, "total_loss": 74.303029, "clip_frac": 0.282837, "sps": 417.1, "elapsed_s": 1311.2}
+{"update": 264, "steps_done": 540672, "mean_step_reward": 0.2545, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 16.404, "mean_ep_length": 59.3, "mean_collisions": 9.43, "mean_intent_rate": 0.804, "policy_loss": 0.003155, "value_loss": 208.359839, "entropy": 0.426021, "total_loss": 104.178814, "clip_frac": 0.247559, "sps": 417.2, "elapsed_s": 1316.1}
+{"update": 265, "steps_done": 542720, "mean_step_reward": 0.6772, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 45.608, "mean_ep_length": 66.8, "mean_collisions": 7.97, "mean_intent_rate": 0.801, "policy_loss": 0.004552, "value_loss": 148.082373, "entropy": 0.397392, "total_loss": 74.041765, "clip_frac": 0.123047, "sps": 411.8, "elapsed_s": 1321.1}
+{"update": 266, "steps_done": 544768, "mean_step_reward": 0.5681, "min_step_reward": -10.5821, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 33.677, "mean_ep_length": 60.8, "mean_collisions": 7.56, "mean_intent_rate": 0.772, "policy_loss": 0.008889, "value_loss": 235.197142, "entropy": 0.419422, "total_loss": 117.603265, "clip_frac": 0.178711, "sps": 407.2, "elapsed_s": 1326.1}
+{"update": 267, "steps_done": 546816, "mean_step_reward": 1.1822, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 78.959, "mean_ep_length": 66.5, "mean_collisions": 3.1, "mean_intent_rate": 0.728, "policy_loss": 0.017823, "value_loss": 126.197799, "entropy": 0.461954, "total_loss": 63.112103, "clip_frac": 0.197144, "sps": 403.7, "elapsed_s": 1331.2}
+{"update": 268, "steps_done": 548864, "mean_step_reward": 0.9277, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 58.605, "mean_ep_length": 62.7, "mean_collisions": 5.03, "mean_intent_rate": 0.76, "policy_loss": 0.013787, "value_loss": 154.983382, "entropy": 0.437239, "total_loss": 77.501107, "clip_frac": 0.170532, "sps": 409.7, "elapsed_s": 1336.2}
+{"update": 269, "steps_done": 550912, "mean_step_reward": 0.4388, "min_step_reward": -10.5761, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 32.367, "mean_ep_length": 64.1, "mean_collisions": 8.25, "mean_intent_rate": 0.771, "policy_loss": 0.006393, "value_loss": 244.366757, "entropy": 0.448344, "total_loss": 122.185288, "clip_frac": 0.275024, "sps": 414.5, "elapsed_s": 1341.1}
+{"update": 270, "steps_done": 552960, "mean_step_reward": 0.9024, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 48.67, "mean_ep_length": 58.2, "mean_collisions": 5.53, "mean_intent_rate": 0.774, "policy_loss": 0.012546, "value_loss": 162.419549, "entropy": 0.457993, "total_loss": 81.217741, "clip_frac": 0.14856, "sps": 412.3, "elapsed_s": 1346.1}
+{"update": 271, "steps_done": 555008, "mean_step_reward": 0.6825, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 42.25, "mean_ep_length": 62.9, "mean_collisions": 7.16, "mean_intent_rate": 0.782, "policy_loss": -0.002846, "value_loss": 136.524663, "entropy": 0.451283, "total_loss": 68.254974, "clip_frac": 0.090698, "sps": 415.2, "elapsed_s": 1351.0}
+{"update": 272, "steps_done": 557056, "mean_step_reward": 0.9982, "min_step_reward": -10.5846, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 60.406, "mean_ep_length": 58.7, "mean_collisions": 4.14, "mean_intent_rate": 0.754, "policy_loss": 0.003462, "value_loss": 137.866159, "entropy": 0.459392, "total_loss": 68.931947, "clip_frac": 0.11377, "sps": 415.5, "elapsed_s": 1356.0}
+{"update": 273, "steps_done": 559104, "mean_step_reward": 0.8321, "min_step_reward": -10.5562, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 50.595, "mean_ep_length": 62.8, "mean_collisions": 4.09, "mean_intent_rate": 0.672, "policy_loss": 0.003755, "value_loss": 111.648949, "entropy": 0.48227, "total_loss": 55.823406, "clip_frac": 0.082153, "sps": 415.6, "elapsed_s": 1360.9}
+{"update": 274, "steps_done": 561152, "mean_step_reward": 1.1392, "min_step_reward": -10.5352, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 70.555, "mean_ep_length": 61.6, "mean_collisions": 3.0, "mean_intent_rate": 0.722, "policy_loss": 0.033099, "value_loss": 103.484451, "entropy": 0.477951, "total_loss": 51.770545, "clip_frac": 0.270752, "sps": 411.0, "elapsed_s": 1365.9}
+{"update": 275, "steps_done": 563200, "mean_step_reward": 0.757, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 45.032, "mean_ep_length": 59.4, "mean_collisions": 5.03, "mean_intent_rate": 0.711, "policy_loss": 0.018603, "value_loss": 134.670711, "entropy": 0.528926, "total_loss": 67.34867, "clip_frac": 0.237061, "sps": 406.2, "elapsed_s": 1370.9}
+{"update": 276, "steps_done": 565248, "mean_step_reward": 0.5976, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 36.181, "mean_ep_length": 59.8, "mean_collisions": 6.5, "mean_intent_rate": 0.746, "policy_loss": 0.003344, "value_loss": 133.892966, "entropy": 0.446068, "total_loss": 66.945367, "clip_frac": 0.161377, "sps": 412.7, "elapsed_s": 1375.9}
+{"update": 277, "steps_done": 567296, "mean_step_reward": 1.1647, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 66.966, "mean_ep_length": 58.1, "mean_collisions": 3.0, "mean_intent_rate": 0.738, "policy_loss": 0.000204, "value_loss": 99.058425, "entropy": 0.42041, "total_loss": 49.525212, "clip_frac": 0.082397, "sps": 413.2, "elapsed_s": 1380.8}
+{"update": 278, "steps_done": 569344, "mean_step_reward": 1.268, "min_step_reward": -10.5842, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 79.331, "mean_ep_length": 62.6, "mean_collisions": 2.64, "mean_intent_rate": 0.74, "policy_loss": 0.021592, "value_loss": 136.243381, "entropy": 0.386379, "total_loss": 68.139418, "clip_frac": 0.201904, "sps": 409.4, "elapsed_s": 1385.8}
+{"update": 279, "steps_done": 571392, "mean_step_reward": 1.2269, "min_step_reward": -10.5567, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 73.415, "mean_ep_length": 58.7, "mean_collisions": 2.0, "mean_intent_rate": 0.707, "policy_loss": 0.047371, "value_loss": 101.028653, "entropy": 0.416311, "total_loss": 50.557534, "clip_frac": 0.324951, "sps": 415.7, "elapsed_s": 1390.8}
+{"update": 280, "steps_done": 573440, "mean_step_reward": 0.9904, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 62.482, "mean_ep_length": 64.0, "mean_collisions": 4.15, "mean_intent_rate": 0.723, "policy_loss": 0.000428, "value_loss": 171.918171, "entropy": 0.399901, "total_loss": 85.955514, "clip_frac": 0.125244, "sps": 416.1, "elapsed_s": 1395.7}
+{"update": 281, "steps_done": 575488, "mean_step_reward": 1.1298, "min_step_reward": -10.5492, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.812, "mean_ep_length": 61.9, "mean_collisions": 2.42, "mean_intent_rate": 0.676, "policy_loss": 0.036223, "value_loss": 132.510939, "entropy": 0.428929, "total_loss": 66.287403, "clip_frac": 0.308594, "sps": 411.6, "elapsed_s": 1400.7}
+{"update": 282, "steps_done": 577536, "mean_step_reward": 0.9828, "min_step_reward": -10.536, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 58.99, "mean_ep_length": 60.9, "mean_collisions": 3.79, "mean_intent_rate": 0.698, "policy_loss": 0.003049, "value_loss": 176.5078, "entropy": 0.407301, "total_loss": 88.252875, "clip_frac": 0.096436, "sps": 413.9, "elapsed_s": 1405.6}
+{"update": 283, "steps_done": 579584, "mean_step_reward": 1.1217, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 71.62, "mean_ep_length": 63.2, "mean_collisions": 3.03, "mean_intent_rate": 0.716, "policy_loss": 0.013628, "value_loss": 136.941726, "entropy": 0.431491, "total_loss": 68.480175, "clip_frac": 0.162842, "sps": 416.6, "elapsed_s": 1410.5}
+{"update": 284, "steps_done": 581632, "mean_step_reward": 1.0582, "min_step_reward": -10.5336, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 70.104, "mean_ep_length": 65.5, "mean_collisions": 4.52, "mean_intent_rate": 0.772, "policy_loss": 0.00434, "value_loss": 162.433454, "entropy": 0.43866, "total_loss": 81.216679, "clip_frac": 0.093872, "sps": 403.5, "elapsed_s": 1415.6}
+{"update": 285, "steps_done": 583680, "mean_step_reward": 1.065, "min_step_reward": -10.5843, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 62.611, "mean_ep_length": 60.1, "mean_collisions": 3.71, "mean_intent_rate": 0.736, "policy_loss": 0.007135, "value_loss": 135.675683, "entropy": 0.441253, "total_loss": 67.840564, "clip_frac": 0.115356, "sps": 412.5, "elapsed_s": 1420.6}
+{"update": 286, "steps_done": 585728, "mean_step_reward": 1.1156, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.202, "mean_ep_length": 61.6, "mean_collisions": 2.82, "mean_intent_rate": 0.701, "policy_loss": 0.015314, "value_loss": 123.99586, "entropy": 0.446623, "total_loss": 62.008777, "clip_frac": 0.194702, "sps": 412.2, "elapsed_s": 1425.6}
+{"update": 287, "steps_done": 587776, "mean_step_reward": 0.8854, "min_step_reward": -10.5752, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 57.173, "mean_ep_length": 62.8, "mean_collisions": 4.88, "mean_intent_rate": 0.737, "policy_loss": 0.015881, "value_loss": 151.229731, "entropy": 0.426438, "total_loss": 75.626482, "clip_frac": 0.230103, "sps": 410.8, "elapsed_s": 1430.5}
+{"update": 288, "steps_done": 589824, "mean_step_reward": 0.7725, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 47.595, "mean_ep_length": 62.5, "mean_collisions": 5.53, "mean_intent_rate": 0.733, "policy_loss": 0.010271, "value_loss": 150.804232, "entropy": 0.422783, "total_loss": 75.408159, "clip_frac": 0.167847, "sps": 413.5, "elapsed_s": 1435.5}
+{"update": 289, "steps_done": 591872, "mean_step_reward": 1.3293, "min_step_reward": -10.5839, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 79.139, "mean_ep_length": 59.7, "mean_collisions": 2.44, "mean_intent_rate": 0.761, "policy_loss": 0.044241, "value_loss": 142.12448, "entropy": 0.396371, "total_loss": 71.102517, "clip_frac": 0.2771, "sps": 412.1, "elapsed_s": 1440.5}
+{"update": 290, "steps_done": 593920, "mean_step_reward": 0.5224, "min_step_reward": -10.5766, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 34.269, "mean_ep_length": 66.1, "mean_collisions": 8.97, "mean_intent_rate": 0.81, "policy_loss": 0.01055, "value_loss": 271.969316, "entropy": 0.372893, "total_loss": 135.991478, "clip_frac": 0.233398, "sps": 412.5, "elapsed_s": 1445.4}
+{"update": 291, "steps_done": 595968, "mean_step_reward": 1.1457, "min_step_reward": -10.5703, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 73.78, "mean_ep_length": 64.2, "mean_collisions": 4.72, "mean_intent_rate": 0.803, "policy_loss": 0.014884, "value_loss": 186.785608, "entropy": 0.368829, "total_loss": 93.403999, "clip_frac": 0.146851, "sps": 407.5, "elapsed_s": 1450.5}
+{"update": 292, "steps_done": 598016, "mean_step_reward": 1.0072, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 61.326, "mean_ep_length": 60.5, "mean_collisions": 4.82, "mean_intent_rate": 0.786, "policy_loss": 0.012245, "value_loss": 142.492032, "entropy": 0.396683, "total_loss": 71.254294, "clip_frac": 0.146362, "sps": 405.1, "elapsed_s": 1455.5}
+{"update": 293, "steps_done": 600064, "mean_step_reward": 1.0322, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 60.483, "mean_ep_length": 59.3, "mean_collisions": 4.74, "mean_intent_rate": 0.789, "policy_loss": 0.00115, "value_loss": 163.198795, "entropy": 0.39933, "total_loss": 81.596554, "clip_frac": 0.109253, "sps": 412.8, "elapsed_s": 1460.5}
+{"update": 294, "steps_done": 602112, "mean_step_reward": 0.6788, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 42.704, "mean_ep_length": 62.5, "mean_collisions": 7.0, "mean_intent_rate": 0.787, "policy_loss": 0.000405, "value_loss": 174.271574, "entropy": 0.398432, "total_loss": 87.132207, "clip_frac": 0.158325, "sps": 416.4, "elapsed_s": 1465.4}
+{"update": 295, "steps_done": 604160, "mean_step_reward": 0.7348, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 45.274, "mean_ep_length": 62.1, "mean_collisions": 6.44, "mean_intent_rate": 0.775, "policy_loss": 0.008342, "value_loss": 218.15189, "entropy": 0.376777, "total_loss": 109.08052, "clip_frac": 0.096924, "sps": 413.1, "elapsed_s": 1470.4}
+{"update": 296, "steps_done": 606208, "mean_step_reward": 0.7747, "min_step_reward": -10.5748, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 52.826, "mean_ep_length": 65.0, "mean_collisions": 5.72, "mean_intent_rate": 0.743, "policy_loss": 0.00864, "value_loss": 191.972575, "entropy": 0.405606, "total_loss": 95.990872, "clip_frac": 0.140869, "sps": 412.2, "elapsed_s": 1475.3}
+{"update": 297, "steps_done": 608256, "mean_step_reward": 0.8905, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.333, "mean_ep_length": 64.1, "mean_collisions": 5.38, "mean_intent_rate": 0.745, "policy_loss": 0.006438, "value_loss": 183.477348, "entropy": 0.405815, "total_loss": 91.741054, "clip_frac": 0.155029, "sps": 409.9, "elapsed_s": 1480.3}
+{"update": 298, "steps_done": 610304, "mean_step_reward": 1.1369, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 70.812, "mean_ep_length": 61.2, "mean_collisions": 4.27, "mean_intent_rate": 0.804, "policy_loss": 0.007195, "value_loss": 154.191037, "entropy": 0.391616, "total_loss": 77.098797, "clip_frac": 0.110962, "sps": 410.1, "elapsed_s": 1485.3}
+{"update": 299, "steps_done": 612352, "mean_step_reward": 1.1155, "min_step_reward": -10.5737, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 66.801, "mean_ep_length": 59.4, "mean_collisions": 3.51, "mean_intent_rate": 0.753, "policy_loss": 0.004729, "value_loss": 134.84057, "entropy": 0.37814, "total_loss": 67.421234, "clip_frac": 0.137085, "sps": 410.2, "elapsed_s": 1490.3}
+{"update": 300, "steps_done": 614400, "mean_step_reward": 0.9576, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 56.215, "mean_ep_length": 59.2, "mean_collisions": 4.66, "mean_intent_rate": 0.763, "policy_loss": 0.016896, "value_loss": 148.257953, "entropy": 0.409872, "total_loss": 74.141773, "clip_frac": 0.163696, "sps": 411.5, "elapsed_s": 1495.3}
+{"update": 301, "steps_done": 616448, "mean_step_reward": 0.877, "min_step_reward": -10.5812, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.748, "mean_ep_length": 60.4, "mean_collisions": 5.71, "mean_intent_rate": 0.779, "policy_loss": 0.008278, "value_loss": 139.091229, "entropy": 0.38944, "total_loss": 69.549998, "clip_frac": 0.165039, "sps": 406.3, "elapsed_s": 1500.3}
+{"update": 302, "steps_done": 618496, "mean_step_reward": 0.7674, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 47.28, "mean_ep_length": 64.2, "mean_collisions": 6.45, "mean_intent_rate": 0.769, "policy_loss": -0.000392, "value_loss": 149.826308, "entropy": 0.394599, "total_loss": 74.908816, "clip_frac": 0.123169, "sps": 415.6, "elapsed_s": 1505.3}
+{"update": 303, "steps_done": 620544, "mean_step_reward": 1.2106, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 76.426, "mean_ep_length": 62.1, "mean_collisions": 3.12, "mean_intent_rate": 0.765, "policy_loss": 0.006896, "value_loss": 126.12014, "entropy": 0.369591, "total_loss": 63.063271, "clip_frac": 0.097778, "sps": 411.9, "elapsed_s": 1510.2}
+{"update": 304, "steps_done": 622592, "mean_step_reward": 0.7938, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 45.74, "mean_ep_length": 58.6, "mean_collisions": 6.56, "mean_intent_rate": 0.816, "policy_loss": 0.013287, "value_loss": 217.549009, "entropy": 0.384331, "total_loss": 108.783948, "clip_frac": 0.191406, "sps": 408.4, "elapsed_s": 1515.3}
+{"update": 305, "steps_done": 624640, "mean_step_reward": 0.9313, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 62.619, "mean_ep_length": 62.9, "mean_collisions": 5.61, "mean_intent_rate": 0.806, "policy_loss": 0.000534, "value_loss": 148.806209, "entropy": 0.354788, "total_loss": 74.400091, "clip_frac": 0.11853, "sps": 411.2, "elapsed_s": 1520.2}
+{"update": 306, "steps_done": 626688, "mean_step_reward": 0.97, "min_step_reward": -10.5753, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 59.691, "mean_ep_length": 65.9, "mean_collisions": 6.58, "mean_intent_rate": 0.815, "policy_loss": 0.013449, "value_loss": 173.244805, "entropy": 0.326668, "total_loss": 86.632585, "clip_frac": 0.124512, "sps": 414.2, "elapsed_s": 1525.2}
+{"update": 307, "steps_done": 628736, "mean_step_reward": 0.5762, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 38.405, "mean_ep_length": 66.6, "mean_collisions": 8.32, "mean_intent_rate": 0.789, "policy_loss": 0.00439, "value_loss": 233.208593, "entropy": 0.368117, "total_loss": 116.605006, "clip_frac": 0.161255, "sps": 410.6, "elapsed_s": 1530.2}
+{"update": 308, "steps_done": 630784, "mean_step_reward": 1.1273, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 72.104, "mean_ep_length": 63.4, "mean_collisions": 3.75, "mean_intent_rate": 0.759, "policy_loss": 0.013519, "value_loss": 157.204729, "entropy": 0.395986, "total_loss": 78.611925, "clip_frac": 0.173584, "sps": 413.4, "elapsed_s": 1535.1}
+{"update": 309, "steps_done": 632832, "mean_step_reward": 0.9414, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 60.045, "mean_ep_length": 64.2, "mean_collisions": 5.5, "mean_intent_rate": 0.774, "policy_loss": 0.003379, "value_loss": 197.761558, "entropy": 0.368954, "total_loss": 98.880468, "clip_frac": 0.100098, "sps": 405.7, "elapsed_s": 1540.2}
+{"update": 310, "steps_done": 634880, "mean_step_reward": 1.073, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 69.809, "mean_ep_length": 65.2, "mean_collisions": 4.16, "mean_intent_rate": 0.755, "policy_loss": 0.018149, "value_loss": 144.361884, "entropy": 0.41141, "total_loss": 72.194977, "clip_frac": 0.214233, "sps": 415.7, "elapsed_s": 1545.1}
+{"update": 311, "steps_done": 636928, "mean_step_reward": 1.3257, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 84.34, "mean_ep_length": 63.5, "mean_collisions": 3.73, "mean_intent_rate": 0.818, "policy_loss": 0.005301, "value_loss": 151.387983, "entropy": 0.370684, "total_loss": 75.695585, "clip_frac": 0.088623, "sps": 413.0, "elapsed_s": 1550.1}
+{"update": 312, "steps_done": 638976, "mean_step_reward": 1.0673, "min_step_reward": -10.5607, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 67.149, "mean_ep_length": 63.4, "mean_collisions": 4.03, "mean_intent_rate": 0.747, "policy_loss": 0.002546, "value_loss": 130.856604, "entropy": 0.388915, "total_loss": 65.426959, "clip_frac": 0.113892, "sps": 412.7, "elapsed_s": 1555.0}
+{"update": 313, "steps_done": 641024, "mean_step_reward": 0.8894, "min_step_reward": -10.558, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 55.089, "mean_ep_length": 61.0, "mean_collisions": 6.21, "mean_intent_rate": 0.822, "policy_loss": 0.002677, "value_loss": 234.410349, "entropy": 0.356721, "total_loss": 117.204285, "clip_frac": 0.108276, "sps": 413.3, "elapsed_s": 1560.0}
+{"update": 314, "steps_done": 643072, "mean_step_reward": 1.0442, "min_step_reward": -10.5855, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 67.339, "mean_ep_length": 65.2, "mean_collisions": 4.06, "mean_intent_rate": 0.743, "policy_loss": 0.052585, "value_loss": 132.412164, "entropy": 0.386922, "total_loss": 66.254798, "clip_frac": 0.315674, "sps": 421.7, "elapsed_s": 1564.8}
+{"update": 315, "steps_done": 645120, "mean_step_reward": 0.6612, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.674, "mean_ep_length": 60.8, "mean_collisions": 7.94, "mean_intent_rate": 0.838, "policy_loss": 0.00524, "value_loss": 225.010206, "entropy": 0.352404, "total_loss": 112.506818, "clip_frac": 0.202148, "sps": 410.0, "elapsed_s": 1569.8}
+{"update": 316, "steps_done": 647168, "mean_step_reward": 0.5473, "min_step_reward": -10.5835, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 29.688, "mean_ep_length": 57.5, "mean_collisions": 7.06, "mean_intent_rate": 0.772, "policy_loss": -0.000959, "value_loss": 185.793407, "entropy": 0.3821, "total_loss": 92.891923, "clip_frac": 0.086792, "sps": 380.8, "elapsed_s": 1575.2}
+{"update": 317, "steps_done": 649216, "mean_step_reward": 1.1528, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 67.944, "mean_ep_length": 57.9, "mean_collisions": 3.77, "mean_intent_rate": 0.787, "policy_loss": 0.016783, "value_loss": 125.29027, "entropy": 0.408599, "total_loss": 62.657832, "clip_frac": 0.133667, "sps": 374.6, "elapsed_s": 1580.7}
+{"update": 318, "steps_done": 651264, "mean_step_reward": 1.2291, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 80.904, "mean_ep_length": 66.6, "mean_collisions": 3.1, "mean_intent_rate": 0.743, "policy_loss": 0.007438, "value_loss": 97.092856, "entropy": 0.417815, "total_loss": 48.549688, "clip_frac": 0.130127, "sps": 381.1, "elapsed_s": 1586.1}
+{"update": 319, "steps_done": 653312, "mean_step_reward": 1.2048, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 82.714, "mean_ep_length": 67.4, "mean_collisions": 2.48, "mean_intent_rate": 0.711, "policy_loss": 0.021693, "value_loss": 99.431531, "entropy": 0.439491, "total_loss": 49.733064, "clip_frac": 0.202881, "sps": 381.3, "elapsed_s": 1591.4}
+{"update": 320, "steps_done": 655360, "mean_step_reward": 0.8997, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 60.664, "mean_ep_length": 67.9, "mean_collisions": 3.23, "mean_intent_rate": 0.639, "policy_loss": 0.018369, "value_loss": 109.269949, "entropy": 0.424602, "total_loss": 54.649098, "clip_frac": 0.254517, "sps": 383.2, "elapsed_s": 1596.8}
+{"update": 321, "steps_done": 657408, "mean_step_reward": 1.1654, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 71.118, "mean_ep_length": 60.4, "mean_collisions": 1.88, "mean_intent_rate": 0.68, "policy_loss": 0.017274, "value_loss": 102.06676, "entropy": 0.437574, "total_loss": 51.046278, "clip_frac": 0.168823, "sps": 372.1, "elapsed_s": 1602.3}
+{"update": 322, "steps_done": 659456, "mean_step_reward": 0.8426, "min_step_reward": -10.5762, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.307, "mean_ep_length": 61.1, "mean_collisions": 5.26, "mean_intent_rate": 0.751, "policy_loss": 0.013071, "value_loss": 231.91408, "entropy": 0.350178, "total_loss": 115.96661, "clip_frac": 0.182983, "sps": 374.7, "elapsed_s": 1607.7}
+{"update": 323, "steps_done": 661504, "mean_step_reward": 1.2848, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 70.962, "mean_ep_length": 57.0, "mean_collisions": 2.11, "mean_intent_rate": 0.717, "policy_loss": 0.040257, "value_loss": 126.056433, "entropy": 0.389972, "total_loss": 63.064574, "clip_frac": 0.295532, "sps": 379.1, "elapsed_s": 1613.1}
+{"update": 324, "steps_done": 663552, "mean_step_reward": 1.0579, "min_step_reward": -10.5833, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 69.429, "mean_ep_length": 65.1, "mean_collisions": 2.35, "mean_intent_rate": 0.658, "policy_loss": 0.013748, "value_loss": 110.099075, "entropy": 0.397498, "total_loss": 55.05931, "clip_frac": 0.19812, "sps": 370.1, "elapsed_s": 1618.7}
+{"update": 325, "steps_done": 665600, "mean_step_reward": 1.0434, "min_step_reward": -10.511, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 68.597, "mean_ep_length": 65.2, "mean_collisions": 3.41, "mean_intent_rate": 0.698, "policy_loss": 0.025767, "value_loss": 128.432048, "entropy": 0.420946, "total_loss": 64.237581, "clip_frac": 0.228149, "sps": 364.0, "elapsed_s": 1624.3}
+{"update": 326, "steps_done": 667648, "mean_step_reward": 1.0387, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 66.699, "mean_ep_length": 64.4, "mean_collisions": 2.19, "mean_intent_rate": 0.646, "policy_loss": 0.012322, "value_loss": 114.811227, "entropy": 0.384606, "total_loss": 57.414089, "clip_frac": 0.142456, "sps": 361.2, "elapsed_s": 1630.0}
+{"update": 327, "steps_done": 669696, "mean_step_reward": 1.0035, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 61.845, "mean_ep_length": 61.3, "mean_collisions": 1.45, "mean_intent_rate": 0.594, "policy_loss": 0.041018, "value_loss": 107.060581, "entropy": 0.376792, "total_loss": 53.56754, "clip_frac": 0.305542, "sps": 359.0, "elapsed_s": 1635.7}
+{"update": 328, "steps_done": 671744, "mean_step_reward": 1.2136, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 72.351, "mean_ep_length": 60.2, "mean_collisions": 0.68, "mean_intent_rate": 0.619, "policy_loss": 0.020023, "value_loss": 112.164037, "entropy": 0.383261, "total_loss": 56.098208, "clip_frac": 0.210449, "sps": 369.7, "elapsed_s": 1641.2}
+{"update": 329, "steps_done": 673792, "mean_step_reward": 1.0387, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 68.73, "mean_ep_length": 65.2, "mean_collisions": 1.84, "mean_intent_rate": 0.622, "policy_loss": 0.004871, "value_loss": 112.838795, "entropy": 0.381952, "total_loss": 56.420449, "clip_frac": 0.084961, "sps": 357.9, "elapsed_s": 1646.9}
+{"update": 330, "steps_done": 675840, "mean_step_reward": 1.1433, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 74.146, "mean_ep_length": 65.2, "mean_collisions": 2.16, "mean_intent_rate": 0.678, "policy_loss": 0.03055, "value_loss": 118.265485, "entropy": 0.370474, "total_loss": 59.159588, "clip_frac": 0.242676, "sps": 367.4, "elapsed_s": 1652.5}
+{"update": 331, "steps_done": 677888, "mean_step_reward": 1.242, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 75.0, "mean_ep_length": 60.2, "mean_collisions": 1.68, "mean_intent_rate": 0.69, "policy_loss": 0.01499, "value_loss": 136.050209, "entropy": 0.417963, "total_loss": 68.035916, "clip_frac": 0.164185, "sps": 368.9, "elapsed_s": 1658.1}
+{"update": 332, "steps_done": 679936, "mean_step_reward": 1.1157, "min_step_reward": -10.5668, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 67.853, "mean_ep_length": 60.9, "mean_collisions": 3.53, "mean_intent_rate": 0.744, "policy_loss": 0.006256, "value_loss": 152.906503, "entropy": 0.377327, "total_loss": 76.455734, "clip_frac": 0.147217, "sps": 359.7, "elapsed_s": 1663.8}
+{"update": 333, "steps_done": 681984, "mean_step_reward": 1.201, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 75.048, "mean_ep_length": 62.3, "mean_collisions": 3.36, "mean_intent_rate": 0.763, "policy_loss": 0.010733, "value_loss": 142.583789, "entropy": 0.35254, "total_loss": 71.299102, "clip_frac": 0.148193, "sps": 357.4, "elapsed_s": 1669.5}
+{"update": 334, "steps_done": 684032, "mean_step_reward": 1.1881, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 73.442, "mean_ep_length": 60.4, "mean_collisions": 2.79, "mean_intent_rate": 0.735, "policy_loss": 0.014802, "value_loss": 172.607832, "entropy": 0.364316, "total_loss": 86.315075, "clip_frac": 0.188477, "sps": 358.8, "elapsed_s": 1675.2}
+{"update": 335, "steps_done": 686080, "mean_step_reward": 1.3285, "min_step_reward": -10.5797, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 79.448, "mean_ep_length": 61.2, "mean_collisions": 3.12, "mean_intent_rate": 0.789, "policy_loss": 0.006524, "value_loss": 165.299754, "entropy": 0.321196, "total_loss": 82.653189, "clip_frac": 0.092773, "sps": 361.2, "elapsed_s": 1680.9}
+{"update": 336, "steps_done": 688128, "mean_step_reward": 1.0671, "min_step_reward": -10.5337, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 62.522, "mean_ep_length": 58.8, "mean_collisions": 4.2, "mean_intent_rate": 0.773, "policy_loss": 0.004867, "value_loss": 143.340959, "entropy": 0.374085, "total_loss": 71.671606, "clip_frac": 0.141357, "sps": 375.2, "elapsed_s": 1686.3}
+{"update": 337, "steps_done": 690176, "mean_step_reward": 0.9744, "min_step_reward": -10.5408, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 63.636, "mean_ep_length": 66.4, "mean_collisions": 5.57, "mean_intent_rate": 0.774, "policy_loss": 0.002467, "value_loss": 162.440536, "entropy": 0.347584, "total_loss": 81.219259, "clip_frac": 0.112183, "sps": 362.7, "elapsed_s": 1692.0}
+{"update": 338, "steps_done": 692224, "mean_step_reward": 1.2604, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 78.827, "mean_ep_length": 60.1, "mean_collisions": 2.88, "mean_intent_rate": 0.774, "policy_loss": 0.034368, "value_loss": 154.937284, "entropy": 0.338464, "total_loss": 77.499625, "clip_frac": 0.234375, "sps": 367.1, "elapsed_s": 1697.6}
+{"update": 339, "steps_done": 694272, "mean_step_reward": 1.2674, "min_step_reward": -10.5035, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 74.694, "mean_ep_length": 61.4, "mean_collisions": 3.03, "mean_intent_rate": 0.754, "policy_loss": 0.014972, "value_loss": 152.987544, "entropy": 0.35893, "total_loss": 76.505156, "clip_frac": 0.193237, "sps": 364.6, "elapsed_s": 1703.2}
+{"update": 340, "steps_done": 696320, "mean_step_reward": 1.0459, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 63.202, "mean_ep_length": 59.6, "mean_collisions": 3.49, "mean_intent_rate": 0.721, "policy_loss": 0.012117, "value_loss": 145.212939, "entropy": 0.329877, "total_loss": 72.615287, "clip_frac": 0.166016, "sps": 365.7, "elapsed_s": 1708.8}
+{"update": 341, "steps_done": 698368, "mean_step_reward": 1.305, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 79.651, "mean_ep_length": 61.5, "mean_collisions": 2.48, "mean_intent_rate": 0.749, "policy_loss": 0.004806, "value_loss": 143.445525, "entropy": 0.287686, "total_loss": 71.724692, "clip_frac": 0.075073, "sps": 364.7, "elapsed_s": 1714.4}
+{"update": 342, "steps_done": 700416, "mean_step_reward": 1.0979, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 67.094, "mean_ep_length": 60.6, "mean_collisions": 4.47, "mean_intent_rate": 0.799, "policy_loss": 0.017314, "value_loss": 205.586217, "entropy": 0.330787, "total_loss": 102.807116, "clip_frac": 0.233276, "sps": 365.9, "elapsed_s": 1720.0}
+{"update": 343, "steps_done": 702464, "mean_step_reward": 0.8965, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 56.849, "mean_ep_length": 63.7, "mean_collisions": 4.47, "mean_intent_rate": 0.712, "policy_loss": 0.004033, "value_loss": 189.668057, "entropy": 0.368598, "total_loss": 94.834375, "clip_frac": 0.130249, "sps": 360.8, "elapsed_s": 1725.7}
+{"update": 344, "steps_done": 704512, "mean_step_reward": 1.2375, "min_step_reward": -10.5416, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 81.82, "mean_ep_length": 65.7, "mean_collisions": 2.55, "mean_intent_rate": 0.721, "policy_loss": 0.015167, "value_loss": 149.44982, "entropy": 0.365195, "total_loss": 74.736425, "clip_frac": 0.19165, "sps": 368.2, "elapsed_s": 1731.2}
+{"update": 345, "steps_done": 706560, "mean_step_reward": 1.1385, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 77.356, "mean_ep_length": 65.7, "mean_collisions": 3.06, "mean_intent_rate": 0.725, "policy_loss": 0.013847, "value_loss": 191.947372, "entropy": 0.365543, "total_loss": 95.983878, "clip_frac": 0.177856, "sps": 359.5, "elapsed_s": 1736.9}
+{"update": 346, "steps_done": 708608, "mean_step_reward": 1.0533, "min_step_reward": -10.5716, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 64.913, "mean_ep_length": 64.2, "mean_collisions": 3.12, "mean_intent_rate": 0.673, "policy_loss": 0.004901, "value_loss": 178.045873, "entropy": 0.384998, "total_loss": 89.023988, "clip_frac": 0.085327, "sps": 360.9, "elapsed_s": 1742.6}
+{"update": 347, "steps_done": 710656, "mean_step_reward": 1.1234, "min_step_reward": -10.5277, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 66.707, "mean_ep_length": 58.8, "mean_collisions": 2.34, "mean_intent_rate": 0.693, "policy_loss": 0.020439, "value_loss": 142.751448, "entropy": 0.390427, "total_loss": 71.392259, "clip_frac": 0.20105, "sps": 359.8, "elapsed_s": 1748.3}
+{"update": 348, "steps_done": 712704, "mean_step_reward": 0.9055, "min_step_reward": -10.5879, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 56.58, "mean_ep_length": 62.2, "mean_collisions": 4.7, "mean_intent_rate": 0.735, "policy_loss": 0.03674, "value_loss": 173.877078, "entropy": 0.430573, "total_loss": 86.970974, "clip_frac": 0.294434, "sps": 374.1, "elapsed_s": 1753.8}
+{"update": 349, "steps_done": 714752, "mean_step_reward": 0.4711, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 28.657, "mean_ep_length": 62.4, "mean_collisions": 8.61, "mean_intent_rate": 0.789, "policy_loss": 0.003439, "value_loss": 241.077517, "entropy": 0.372802, "total_loss": 120.538469, "clip_frac": 0.155518, "sps": 364.6, "elapsed_s": 1759.4}
+{"update": 350, "steps_done": 716800, "mean_step_reward": 0.6313, "min_step_reward": -10.5518, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 38.961, "mean_ep_length": 63.2, "mean_collisions": 7.59, "mean_intent_rate": 0.788, "policy_loss": 0.013653, "value_loss": 199.098271, "entropy": 0.36026, "total_loss": 99.559186, "clip_frac": 0.133789, "sps": 364.9, "elapsed_s": 1765.0}
+{"update": 351, "steps_done": 718848, "mean_step_reward": 0.6211, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 39.411, "mean_ep_length": 62.6, "mean_collisions": 8.33, "mean_intent_rate": 0.831, "policy_loss": 0.003674, "value_loss": 228.574481, "entropy": 0.339311, "total_loss": 114.287522, "clip_frac": 0.186523, "sps": 365.7, "elapsed_s": 1770.6}
+{"update": 352, "steps_done": 720896, "mean_step_reward": 1.0667, "min_step_reward": -10.5542, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 61.97, "mean_ep_length": 58.1, "mean_collisions": 4.74, "mean_intent_rate": 0.807, "policy_loss": 0.009059, "value_loss": 163.692759, "entropy": 0.339962, "total_loss": 81.852039, "clip_frac": 0.105103, "sps": 357.9, "elapsed_s": 1776.3}
+{"update": 353, "steps_done": 722944, "mean_step_reward": 1.0649, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 62.542, "mean_ep_length": 58.6, "mean_collisions": 4.86, "mean_intent_rate": 0.809, "policy_loss": 0.002914, "value_loss": 133.033879, "entropy": 0.350652, "total_loss": 66.516347, "clip_frac": 0.144775, "sps": 363.4, "elapsed_s": 1782.0}
+{"update": 354, "steps_done": 724992, "mean_step_reward": 0.9265, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 60.38, "mean_ep_length": 64.7, "mean_collisions": 5.94, "mean_intent_rate": 0.791, "policy_loss": 0.01547, "value_loss": 173.227954, "entropy": 0.318612, "total_loss": 86.626261, "clip_frac": 0.148193, "sps": 371.5, "elapsed_s": 1787.5}
+{"update": 355, "steps_done": 727040, "mean_step_reward": 1.1667, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.409, "mean_ep_length": 60.4, "mean_collisions": 4.64, "mean_intent_rate": 0.816, "policy_loss": 0.017744, "value_loss": 170.624003, "entropy": 0.305713, "total_loss": 85.326689, "clip_frac": 0.141846, "sps": 372.7, "elapsed_s": 1793.0}
+{"update": 356, "steps_done": 729088, "mean_step_reward": 0.5784, "min_step_reward": -10.5821, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 37.528, "mean_ep_length": 61.7, "mean_collisions": 7.5, "mean_intent_rate": 0.795, "policy_loss": 0.027067, "value_loss": 185.391813, "entropy": 0.345965, "total_loss": 92.719514, "clip_frac": 0.175049, "sps": 366.2, "elapsed_s": 1798.6}
+{"update": 357, "steps_done": 731136, "mean_step_reward": 1.0205, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 61.358, "mean_ep_length": 60.9, "mean_collisions": 5.64, "mean_intent_rate": 0.823, "policy_loss": -0.002916, "value_loss": 187.724924, "entropy": 0.322871, "total_loss": 93.856317, "clip_frac": 0.144897, "sps": 367.4, "elapsed_s": 1804.1}
+{"update": 358, "steps_done": 733184, "mean_step_reward": 1.2877, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 81.968, "mean_ep_length": 63.2, "mean_collisions": 3.76, "mean_intent_rate": 0.817, "policy_loss": 0.007726, "value_loss": 171.756927, "entropy": 0.312819, "total_loss": 85.883061, "clip_frac": 0.081543, "sps": 363.3, "elapsed_s": 1809.8}
+{"update": 359, "steps_done": 735232, "mean_step_reward": 1.1319, "min_step_reward": -10.588, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 70.321, "mean_ep_length": 62.1, "mean_collisions": 4.73, "mean_intent_rate": 0.813, "policy_loss": 0.011168, "value_loss": 158.454051, "entropy": 0.310367, "total_loss": 79.23509, "clip_frac": 0.176025, "sps": 368.0, "elapsed_s": 1815.4}
+{"update": 360, "steps_done": 737280, "mean_step_reward": 1.0106, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 61.213, "mean_ep_length": 61.1, "mean_collisions": 4.88, "mean_intent_rate": 0.777, "policy_loss": 0.012736, "value_loss": 192.480675, "entropy": 0.314759, "total_loss": 96.249927, "clip_frac": 0.135254, "sps": 364.3, "elapsed_s": 1821.0}
+{"update": 361, "steps_done": 739328, "mean_step_reward": 1.2543, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 77.305, "mean_ep_length": 63.0, "mean_collisions": 2.97, "mean_intent_rate": 0.75, "policy_loss": 0.018411, "value_loss": 150.89898, "entropy": 0.349862, "total_loss": 75.464402, "clip_frac": 0.173096, "sps": 355.5, "elapsed_s": 1826.7}
+{"update": 362, "steps_done": 741376, "mean_step_reward": 1.333, "min_step_reward": -10.5308, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 87.032, "mean_ep_length": 64.0, "mean_collisions": 2.15, "mean_intent_rate": 0.742, "policy_loss": 0.015313, "value_loss": 127.416136, "entropy": 0.347976, "total_loss": 63.719901, "clip_frac": 0.181274, "sps": 358.3, "elapsed_s": 1832.5}
+{"update": 363, "steps_done": 743424, "mean_step_reward": 0.7709, "min_step_reward": -10.5659, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 48.355, "mean_ep_length": 61.3, "mean_collisions": 6.12, "mean_intent_rate": 0.771, "policy_loss": 0.005902, "value_loss": 229.10175, "entropy": 0.332312, "total_loss": 114.553454, "clip_frac": 0.145752, "sps": 361.0, "elapsed_s": 1838.1}
+{"update": 364, "steps_done": 745472, "mean_step_reward": 1.3623, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 78.977, "mean_ep_length": 59.1, "mean_collisions": 1.46, "mean_intent_rate": 0.706, "policy_loss": 0.019747, "value_loss": 136.109924, "entropy": 0.354994, "total_loss": 68.071159, "clip_frac": 0.17749, "sps": 363.6, "elapsed_s": 1843.8}
+{"update": 365, "steps_done": 747520, "mean_step_reward": 1.2834, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 79.156, "mean_ep_length": 62.1, "mean_collisions": 0.88, "mean_intent_rate": 0.65, "policy_loss": 0.009266, "value_loss": 120.673423, "entropy": 0.395617, "total_loss": 60.342021, "clip_frac": 0.138916, "sps": 358.7, "elapsed_s": 1849.5}
+{"update": 366, "steps_done": 749568, "mean_step_reward": 1.1491, "min_step_reward": -10.5797, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 68.065, "mean_ep_length": 59.3, "mean_collisions": 1.4, "mean_intent_rate": 0.645, "policy_loss": 0.040633, "value_loss": 128.341763, "entropy": 0.409549, "total_loss": 64.207419, "clip_frac": 0.349609, "sps": 351.4, "elapsed_s": 1855.3}
+{"update": 367, "steps_done": 751616, "mean_step_reward": 0.9204, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 62.37, "mean_ep_length": 66.4, "mean_collisions": 0.48, "mean_intent_rate": 0.515, "policy_loss": 0.009731, "value_loss": 83.178294, "entropy": 0.414854, "total_loss": 41.594729, "clip_frac": 0.147705, "sps": 362.9, "elapsed_s": 1860.9}
+{"update": 368, "steps_done": 753664, "mean_step_reward": 1.0514, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 69.158, "mean_ep_length": 64.8, "mean_collisions": 0.35, "mean_intent_rate": 0.551, "policy_loss": 0.018563, "value_loss": 77.587636, "entropy": 0.39811, "total_loss": 38.808401, "clip_frac": 0.203735, "sps": 375.0, "elapsed_s": 1866.4}
+{"update": 369, "steps_done": 755712, "mean_step_reward": 0.6668, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.22, "mean_ep_length": 60.8, "mean_collisions": 0.47, "mean_intent_rate": 0.421, "policy_loss": 0.00399, "value_loss": 68.36232, "entropy": 0.368277, "total_loss": 34.181467, "clip_frac": 0.074463, "sps": 390.1, "elapsed_s": 1871.7}
+{"update": 370, "steps_done": 757760, "mean_step_reward": 0.9128, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 58.883, "mean_ep_length": 64.7, "mean_collisions": 0.66, "mean_intent_rate": 0.511, "policy_loss": 0.011644, "value_loss": 61.320184, "entropy": 0.429673, "total_loss": 30.667439, "clip_frac": 0.242798, "sps": 390.1, "elapsed_s": 1876.9}
+{"update": 371, "steps_done": 759808, "mean_step_reward": 0.8912, "min_step_reward": -7.5827, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 55.356, "mean_ep_length": 62.5, "mean_collisions": 0.36, "mean_intent_rate": 0.487, "policy_loss": 0.004987, "value_loss": 65.375196, "entropy": 0.351923, "total_loss": 32.689065, "clip_frac": 0.117676, "sps": 366.4, "elapsed_s": 1882.5}
+{"update": 372, "steps_done": 761856, "mean_step_reward": 0.7896, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 52.827, "mean_ep_length": 66.6, "mean_collisions": 1.27, "mean_intent_rate": 0.499, "policy_loss": 0.042194, "value_loss": 85.830667, "entropy": 0.275685, "total_loss": 42.954771, "clip_frac": 0.283691, "sps": 366.5, "elapsed_s": 1888.1}
+{"update": 373, "steps_done": 763904, "mean_step_reward": 0.8085, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.379, "mean_ep_length": 62.9, "mean_collisions": 0.48, "mean_intent_rate": 0.473, "policy_loss": 0.006708, "value_loss": 71.712368, "entropy": 0.313919, "total_loss": 35.859753, "clip_frac": 0.116943, "sps": 360.0, "elapsed_s": 1893.8}
+{"update": 374, "steps_done": 765952, "mean_step_reward": 0.7506, "min_step_reward": -10.5301, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 45.848, "mean_ep_length": 61.9, "mean_collisions": 0.85, "mean_intent_rate": 0.46, "policy_loss": 0.04376, "value_loss": 62.342384, "entropy": 0.371058, "total_loss": 31.211241, "clip_frac": 0.271729, "sps": 355.4, "elapsed_s": 1899.5}
+{"update": 375, "steps_done": 768000, "mean_step_reward": 0.6329, "min_step_reward": -7.5777, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 42.476, "mean_ep_length": 66.3, "mean_collisions": 0.03, "mean_intent_rate": 0.386, "policy_loss": 0.013472, "value_loss": 43.236679, "entropy": 0.260494, "total_loss": 21.629206, "clip_frac": 0.101196, "sps": 361.3, "elapsed_s": 1905.2}
+{"update": 376, "steps_done": 770048, "mean_step_reward": 0.6661, "min_step_reward": -7.5118, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.289, "mean_ep_length": 60.9, "mean_collisions": 0.06, "mean_intent_rate": 0.403, "policy_loss": 0.007757, "value_loss": 42.754969, "entropy": 0.285029, "total_loss": 21.382391, "clip_frac": 0.079468, "sps": 364.8, "elapsed_s": 1910.8}
+{"update": 377, "steps_done": 772096, "mean_step_reward": 0.7059, "min_step_reward": -7.5849, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 45.526, "mean_ep_length": 63.5, "mean_collisions": 0.19, "mean_intent_rate": 0.426, "policy_loss": 0.009682, "value_loss": 40.070205, "entropy": 0.349673, "total_loss": 20.041288, "clip_frac": 0.119385, "sps": 349.5, "elapsed_s": 1916.7}
+{"update": 378, "steps_done": 774144, "mean_step_reward": 0.5713, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 33.465, "mean_ep_length": 59.5, "mean_collisions": 0.44, "mean_intent_rate": 0.39, "policy_loss": 0.003877, "value_loss": 29.860492, "entropy": 0.308974, "total_loss": 14.931033, "clip_frac": 0.045166, "sps": 360.4, "elapsed_s": 1922.4}
+{"update": 379, "steps_done": 776192, "mean_step_reward": 0.6444, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 40.439, "mean_ep_length": 62.9, "mean_collisions": 0.15, "mean_intent_rate": 0.392, "policy_loss": 0.006625, "value_loss": 26.493107, "entropy": 0.264173, "total_loss": 13.250537, "clip_frac": 0.061401, "sps": 360.2, "elapsed_s": 1928.1}
+{"update": 380, "steps_done": 778240, "mean_step_reward": 0.6754, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 44.038, "mean_ep_length": 65.1, "mean_collisions": 0.74, "mean_intent_rate": 0.436, "policy_loss": 0.01822, "value_loss": 46.791708, "entropy": 0.286557, "total_loss": 23.411209, "clip_frac": 0.18335, "sps": 352.6, "elapsed_s": 1933.9}
+{"update": 381, "steps_done": 780288, "mean_step_reward": 0.6612, "min_step_reward": -7.5872, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 39.732, "mean_ep_length": 60.1, "mean_collisions": 0.21, "mean_intent_rate": 0.411, "policy_loss": 0.054662, "value_loss": 36.419751, "entropy": 0.295707, "total_loss": 18.261581, "clip_frac": 0.234741, "sps": 359.2, "elapsed_s": 1939.6}
+{"update": 382, "steps_done": 782336, "mean_step_reward": 0.5092, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 30.606, "mean_ep_length": 60.2, "mean_collisions": 0.06, "mean_intent_rate": 0.345, "policy_loss": 0.007142, "value_loss": 28.717645, "entropy": 0.192036, "total_loss": 14.364044, "clip_frac": 0.066162, "sps": 355.2, "elapsed_s": 1945.3}
+{"update": 383, "steps_done": 784384, "mean_step_reward": 0.5303, "min_step_reward": -7.5865, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 35.304, "mean_ep_length": 65.8, "mean_collisions": 0.03, "mean_intent_rate": 0.352, "policy_loss": 0.011469, "value_loss": 25.362763, "entropy": 0.051901, "total_loss": 12.692332, "clip_frac": 0.031372, "sps": 359.1, "elapsed_s": 1951.0}
+{"update": 384, "steps_done": 786432, "mean_step_reward": 0.4467, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 26.796, "mean_ep_length": 60.7, "mean_collisions": 0.0, "mean_intent_rate": 0.321, "policy_loss": 0.000725, "value_loss": 22.233999, "entropy": 0.064647, "total_loss": 11.117078, "clip_frac": 0.007568, "sps": 357.8, "elapsed_s": 1956.8}
+{"update": 385, "steps_done": 788480, "mean_step_reward": 0.4274, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 27.566, "mean_ep_length": 61.8, "mean_collisions": 0.0, "mean_intent_rate": 0.321, "policy_loss": -0.00022, "value_loss": 20.872087, "entropy": 0.044807, "total_loss": 10.435376, "clip_frac": 0.007935, "sps": 355.5, "elapsed_s": 1962.5}
+{"update": 386, "steps_done": 790528, "mean_step_reward": 0.3889, "min_step_reward": -7.5851, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 21.596, "mean_ep_length": 57.9, "mean_collisions": 0.03, "mean_intent_rate": 0.302, "policy_loss": -2e-05, "value_loss": 19.062473, "entropy": 0.041743, "total_loss": 9.530799, "clip_frac": 0.002197, "sps": 350.4, "elapsed_s": 1968.4}
+{"update": 387, "steps_done": 792576, "mean_step_reward": 0.4523, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 30.17, "mean_ep_length": 65.4, "mean_collisions": 0.0, "mean_intent_rate": 0.33, "policy_loss": -7.4e-05, "value_loss": 18.364341, "entropy": 0.049627, "total_loss": 9.1816, "clip_frac": 0.00293, "sps": 350.6, "elapsed_s": 1974.2}
+{"update": 388, "steps_done": 794624, "mean_step_reward": 0.4722, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 29.389, "mean_ep_length": 63.6, "mean_collisions": 0.0, "mean_intent_rate": 0.33, "policy_loss": 0.000168, "value_loss": 21.961203, "entropy": 0.069181, "total_loss": 10.980078, "clip_frac": 0.005859, "sps": 353.3, "elapsed_s": 1980.0}
+{"update": 389, "steps_done": 796672, "mean_step_reward": 0.3981, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 24.952, "mean_ep_length": 62.2, "mean_collisions": 0.03, "mean_intent_rate": 0.313, "policy_loss": -9.5e-05, "value_loss": 19.189256, "entropy": 0.077975, "total_loss": 9.593754, "clip_frac": 0.008545, "sps": 354.6, "elapsed_s": 1985.8}
+{"update": 390, "steps_done": 798720, "mean_step_reward": 0.4504, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 30.602, "mean_ep_length": 67.9, "mean_collisions": 0.0, "mean_intent_rate": 0.327, "policy_loss": 2.2e-05, "value_loss": 19.345715, "entropy": 0.060902, "total_loss": 9.672271, "clip_frac": 0.006226, "sps": 354.8, "elapsed_s": 1991.6}
+{"update": 391, "steps_done": 800768, "mean_step_reward": 0.4187, "min_step_reward": -7.5834, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 27.719, "mean_ep_length": 66.2, "mean_collisions": 0.03, "mean_intent_rate": 0.315, "policy_loss": 0.000677, "value_loss": 17.567453, "entropy": 0.084986, "total_loss": 8.783553, "clip_frac": 0.009033, "sps": 356.2, "elapsed_s": 1997.3}
+{"update": 392, "steps_done": 802816, "mean_step_reward": 0.4503, "min_step_reward": -7.5247, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 29.169, "mean_ep_length": 64.7, "mean_collisions": 0.06, "mean_intent_rate": 0.325, "policy_loss": 0.000473, "value_loss": 19.246034, "entropy": 0.134087, "total_loss": 9.622149, "clip_frac": 0.013306, "sps": 357.5, "elapsed_s": 2003.0}
+{"update": 393, "steps_done": 804864, "mean_step_reward": 0.4854, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 29.138, "mean_ep_length": 60.6, "mean_collisions": 0.21, "mean_intent_rate": 0.344, "policy_loss": 0.001742, "value_loss": 20.958555, "entropy": 0.153008, "total_loss": 10.479489, "clip_frac": 0.017456, "sps": 359.7, "elapsed_s": 2008.7}
+{"update": 394, "steps_done": 806912, "mean_step_reward": 0.4632, "min_step_reward": -7.5892, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 28.686, "mean_ep_length": 62.2, "mean_collisions": 0.19, "mean_intent_rate": 0.331, "policy_loss": 0.000708, "value_loss": 22.166687, "entropy": 0.195432, "total_loss": 11.082097, "clip_frac": 0.014404, "sps": 360.2, "elapsed_s": 2014.4}
+{"update": 395, "steps_done": 808960, "mean_step_reward": 0.5031, "min_step_reward": -7.5868, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 31.393, "mean_ep_length": 61.4, "mean_collisions": 0.12, "mean_intent_rate": 0.355, "policy_loss": 0.001502, "value_loss": 19.071342, "entropy": 0.216339, "total_loss": 9.535009, "clip_frac": 0.030151, "sps": 390.8, "elapsed_s": 2019.7}
+{"update": 396, "steps_done": 811008, "mean_step_reward": 0.4871, "min_step_reward": -7.589, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 29.749, "mean_ep_length": 61.9, "mean_collisions": 0.39, "mean_intent_rate": 0.358, "policy_loss": 0.001967, "value_loss": 24.707192, "entropy": 0.263127, "total_loss": 12.352932, "clip_frac": 0.045776, "sps": 391.5, "elapsed_s": 2024.9}
+{"update": 397, "steps_done": 813056, "mean_step_reward": 0.6531, "min_step_reward": -10.5853, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 40.078, "mean_ep_length": 61.1, "mean_collisions": 0.47, "mean_intent_rate": 0.422, "policy_loss": 0.009818, "value_loss": 26.752166, "entropy": 0.290026, "total_loss": 13.383, "clip_frac": 0.095459, "sps": 377.2, "elapsed_s": 2030.3}
+{"update": 398, "steps_done": 815104, "mean_step_reward": 0.7418, "min_step_reward": -7.5782, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 46.335, "mean_ep_length": 61.6, "mean_collisions": 0.18, "mean_intent_rate": 0.436, "policy_loss": -0.003576, "value_loss": 37.84589, "entropy": 0.422413, "total_loss": 18.915145, "clip_frac": 0.104492, "sps": 354.6, "elapsed_s": 2036.1}
+{"update": 399, "steps_done": 817152, "mean_step_reward": 0.7127, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 42.668, "mean_ep_length": 61.0, "mean_collisions": 1.09, "mean_intent_rate": 0.469, "policy_loss": 0.016517, "value_loss": 43.556096, "entropy": 0.362612, "total_loss": 21.790939, "clip_frac": 0.194702, "sps": 370.0, "elapsed_s": 2041.6}
+{"update": 400, "steps_done": 819200, "mean_step_reward": 0.751, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 46.607, "mean_ep_length": 62.8, "mean_collisions": 0.75, "mean_intent_rate": 0.464, "policy_loss": 0.004737, "value_loss": 41.630923, "entropy": 0.343006, "total_loss": 20.816768, "clip_frac": 0.11792, "sps": 376.4, "elapsed_s": 2047.1}
+{"update": 401, "steps_done": 821248, "mean_step_reward": 0.8888, "min_step_reward": -7.5837, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 55.058, "mean_ep_length": 61.4, "mean_collisions": 0.76, "mean_intent_rate": 0.508, "policy_loss": 0.022373, "value_loss": 67.767581, "entropy": 0.355731, "total_loss": 33.902606, "clip_frac": 0.22522, "sps": 386.5, "elapsed_s": 2052.4}
+{"update": 402, "steps_done": 823296, "mean_step_reward": 0.8038, "min_step_reward": -10.5055, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 48.158, "mean_ep_length": 60.1, "mean_collisions": 2.12, "mean_intent_rate": 0.556, "policy_loss": 0.020552, "value_loss": 104.386519, "entropy": 0.385191, "total_loss": 52.20996, "clip_frac": 0.249512, "sps": 371.8, "elapsed_s": 2057.9}
+{"update": 403, "steps_done": 825344, "mean_step_reward": 0.9843, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 64.857, "mean_ep_length": 66.0, "mean_collisions": 0.65, "mean_intent_rate": 0.537, "policy_loss": 0.038666, "value_loss": 71.532779, "entropy": 0.380982, "total_loss": 35.801246, "clip_frac": 0.29895, "sps": 382.9, "elapsed_s": 2063.2}
+{"update": 404, "steps_done": 827392, "mean_step_reward": 0.8623, "min_step_reward": -10.5301, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 50.868, "mean_ep_length": 59.3, "mean_collisions": 2.32, "mean_intent_rate": 0.591, "policy_loss": 0.00386, "value_loss": 107.318092, "entropy": 0.405928, "total_loss": 53.658846, "clip_frac": 0.141602, "sps": 372.7, "elapsed_s": 2068.7}
+{"update": 405, "steps_done": 829440, "mean_step_reward": 0.9958, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 65.88, "mean_ep_length": 66.6, "mean_collisions": 1.9, "mean_intent_rate": 0.595, "policy_loss": 0.00591, "value_loss": 89.411613, "entropy": 0.412789, "total_loss": 44.707588, "clip_frac": 0.097168, "sps": 357.2, "elapsed_s": 2074.5}
+{"update": 406, "steps_done": 831488, "mean_step_reward": 1.1881, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 72.503, "mean_ep_length": 60.6, "mean_collisions": 0.18, "mean_intent_rate": 0.589, "policy_loss": 0.040125, "value_loss": 78.760254, "entropy": 0.400074, "total_loss": 39.416251, "clip_frac": 0.29834, "sps": 360.7, "elapsed_s": 2080.2}
+{"update": 407, "steps_done": 833536, "mean_step_reward": 0.8948, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.898, "mean_ep_length": 59.5, "mean_collisions": 1.94, "mean_intent_rate": 0.579, "policy_loss": 0.020492, "value_loss": 87.542453, "entropy": 0.389234, "total_loss": 43.787826, "clip_frac": 0.235107, "sps": 361.4, "elapsed_s": 2085.8}
+{"update": 408, "steps_done": 835584, "mean_step_reward": 1.1276, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 70.115, "mean_ep_length": 60.6, "mean_collisions": 0.94, "mean_intent_rate": 0.621, "policy_loss": 0.022398, "value_loss": 84.316359, "entropy": 0.396275, "total_loss": 42.176614, "clip_frac": 0.234619, "sps": 376.4, "elapsed_s": 2091.3}
+{"update": 409, "steps_done": 837632, "mean_step_reward": 1.17, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 67.68, "mean_ep_length": 59.1, "mean_collisions": 1.23, "mean_intent_rate": 0.629, "policy_loss": 0.005272, "value_loss": 115.035565, "entropy": 0.429234, "total_loss": 57.518761, "clip_frac": 0.177246, "sps": 360.9, "elapsed_s": 2096.9}
+{"update": 410, "steps_done": 839680, "mean_step_reward": 0.9203, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 55.961, "mean_ep_length": 62.2, "mean_collisions": 2.19, "mean_intent_rate": 0.603, "policy_loss": 0.009283, "value_loss": 115.762423, "entropy": 0.398724, "total_loss": 57.886507, "clip_frac": 0.161865, "sps": 353.2, "elapsed_s": 2102.7}
+{"update": 411, "steps_done": 841728, "mean_step_reward": 0.9485, "min_step_reward": -10.5823, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 63.904, "mean_ep_length": 66.5, "mean_collisions": 1.48, "mean_intent_rate": 0.575, "policy_loss": 0.011137, "value_loss": 86.780839, "entropy": 0.399343, "total_loss": 43.397563, "clip_frac": 0.174683, "sps": 370.7, "elapsed_s": 2108.3}
+{"update": 412, "steps_done": 843776, "mean_step_reward": 1.073, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 70.332, "mean_ep_length": 63.5, "mean_collisions": 1.66, "mean_intent_rate": 0.63, "policy_loss": 0.018926, "value_loss": 109.050061, "entropy": 0.396243, "total_loss": 54.539994, "clip_frac": 0.253662, "sps": 378.0, "elapsed_s": 2113.7}
+{"update": 413, "steps_done": 845824, "mean_step_reward": 1.2963, "min_step_reward": -10.5598, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 84.541, "mean_ep_length": 67.5, "mean_collisions": 1.39, "mean_intent_rate": 0.669, "policy_loss": 0.01642, "value_loss": 110.764346, "entropy": 0.387198, "total_loss": 55.394721, "clip_frac": 0.215698, "sps": 355.8, "elapsed_s": 2119.4}
+{"update": 414, "steps_done": 847872, "mean_step_reward": 1.1429, "min_step_reward": -10.5888, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.197, "mean_ep_length": 61.0, "mean_collisions": 2.21, "mean_intent_rate": 0.67, "policy_loss": 0.036891, "value_loss": 117.651487, "entropy": 0.382237, "total_loss": 58.858812, "clip_frac": 0.229126, "sps": 351.8, "elapsed_s": 2125.3}
+{"update": 415, "steps_done": 849920, "mean_step_reward": 1.1201, "min_step_reward": -10.5825, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 73.83, "mean_ep_length": 65.3, "mean_collisions": 3.59, "mean_intent_rate": 0.736, "policy_loss": 0.002814, "value_loss": 153.531881, "entropy": 0.363255, "total_loss": 76.765122, "clip_frac": 0.060303, "sps": 354.1, "elapsed_s": 2131.0}
+{"update": 416, "steps_done": 851968, "mean_step_reward": 0.848, "min_step_reward": -10.5479, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.169, "mean_ep_length": 59.5, "mean_collisions": 4.62, "mean_intent_rate": 0.727, "policy_loss": 0.012731, "value_loss": 145.505385, "entropy": 0.387651, "total_loss": 72.761547, "clip_frac": 0.19397, "sps": 356.8, "elapsed_s": 2136.8}
+{"update": 417, "steps_done": 854016, "mean_step_reward": 1.269, "min_step_reward": -10.5569, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 74.913, "mean_ep_length": 61.5, "mean_collisions": 3.52, "mean_intent_rate": 0.774, "policy_loss": 0.008169, "value_loss": 133.588059, "entropy": 0.381842, "total_loss": 66.79838, "clip_frac": 0.088135, "sps": 356.2, "elapsed_s": 2142.5}
+{"update": 418, "steps_done": 856064, "mean_step_reward": 1.0225, "min_step_reward": -10.5728, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 65.318, "mean_ep_length": 62.1, "mean_collisions": 3.48, "mean_intent_rate": 0.724, "policy_loss": 0.002486, "value_loss": 115.104257, "entropy": 0.410858, "total_loss": 57.550506, "clip_frac": 0.083374, "sps": 368.2, "elapsed_s": 2148.1}
+{"update": 419, "steps_done": 858112, "mean_step_reward": 0.8698, "min_step_reward": -10.5302, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 52.273, "mean_ep_length": 60.0, "mean_collisions": 4.34, "mean_intent_rate": 0.711, "policy_loss": 0.024014, "value_loss": 148.838613, "entropy": 0.427634, "total_loss": 74.439044, "clip_frac": 0.235718, "sps": 349.9, "elapsed_s": 2154.0}
+{"update": 420, "steps_done": 860160, "mean_step_reward": 1.0478, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 68.198, "mean_ep_length": 64.3, "mean_collisions": 3.94, "mean_intent_rate": 0.739, "policy_loss": 0.003601, "value_loss": 137.973037, "entropy": 0.406575, "total_loss": 68.986053, "clip_frac": 0.120728, "sps": 350.1, "elapsed_s": 2159.8}
+{"update": 421, "steps_done": 862208, "mean_step_reward": 1.1868, "min_step_reward": -10.5776, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 68.956, "mean_ep_length": 58.8, "mean_collisions": 3.51, "mean_intent_rate": 0.752, "policy_loss": -0.000259, "value_loss": 159.269416, "entropy": 0.353529, "total_loss": 79.630914, "clip_frac": 0.086182, "sps": 351.7, "elapsed_s": 2165.6}
+{"update": 422, "steps_done": 864256, "mean_step_reward": 1.0051, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 61.243, "mean_ep_length": 60.9, "mean_collisions": 4.26, "mean_intent_rate": 0.742, "policy_loss": 0.008748, "value_loss": 132.704853, "entropy": 0.394614, "total_loss": 66.357228, "clip_frac": 0.100098, "sps": 352.5, "elapsed_s": 2171.4}
+{"update": 423, "steps_done": 866304, "mean_step_reward": 1.0269, "min_step_reward": -10.5651, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 62.448, "mean_ep_length": 60.8, "mean_collisions": 4.38, "mean_intent_rate": 0.761, "policy_loss": -0.002978, "value_loss": 155.189332, "entropy": 0.388215, "total_loss": 77.587807, "clip_frac": 0.122681, "sps": 352.8, "elapsed_s": 2177.2}
+{"update": 424, "steps_done": 868352, "mean_step_reward": 1.0367, "min_step_reward": -10.579, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 66.453, "mean_ep_length": 60.8, "mean_collisions": 3.91, "mean_intent_rate": 0.759, "policy_loss": 0.018013, "value_loss": 141.487484, "entropy": 0.382717, "total_loss": 70.757928, "clip_frac": 0.192383, "sps": 357.8, "elapsed_s": 2183.0}
+{"update": 425, "steps_done": 870400, "mean_step_reward": 0.6635, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 36.298, "mean_ep_length": 59.1, "mean_collisions": 6.43, "mean_intent_rate": 0.748, "policy_loss": 0.009906, "value_loss": 167.549096, "entropy": 0.393051, "total_loss": 83.780525, "clip_frac": 0.165161, "sps": 391.8, "elapsed_s": 2188.2}
+{"update": 426, "steps_done": 872448, "mean_step_reward": 1.1991, "min_step_reward": -10.5297, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 79.37, "mean_ep_length": 66.1, "mean_collisions": 2.94, "mean_intent_rate": 0.735, "policy_loss": 0.066672, "value_loss": 135.054533, "entropy": 0.405932, "total_loss": 67.58988, "clip_frac": 0.264893, "sps": 367.6, "elapsed_s": 2193.8}
+{"update": 427, "steps_done": 874496, "mean_step_reward": 0.9526, "min_step_reward": -10.5596, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 56.155, "mean_ep_length": 59.0, "mean_collisions": 4.66, "mean_intent_rate": 0.759, "policy_loss": 0.003573, "value_loss": 145.324818, "entropy": 0.374993, "total_loss": 72.662233, "clip_frac": 0.088501, "sps": 349.4, "elapsed_s": 2199.6}
+{"update": 428, "steps_done": 876544, "mean_step_reward": 0.9914, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 61.286, "mean_ep_length": 62.0, "mean_collisions": 4.91, "mean_intent_rate": 0.773, "policy_loss": 0.025683, "value_loss": 139.052862, "entropy": 0.339841, "total_loss": 69.548715, "clip_frac": 0.218262, "sps": 339.2, "elapsed_s": 2205.7}
+{"update": 429, "steps_done": 878592, "mean_step_reward": 1.2961, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 73.878, "mean_ep_length": 56.4, "mean_collisions": 3.17, "mean_intent_rate": 0.805, "policy_loss": 0.031189, "value_loss": 160.344908, "entropy": 0.360547, "total_loss": 80.200037, "clip_frac": 0.235474, "sps": 346.5, "elapsed_s": 2211.6}
+{"update": 430, "steps_done": 880640, "mean_step_reward": 0.7171, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 44.119, "mean_ep_length": 62.5, "mean_collisions": 7.85, "mean_intent_rate": 0.835, "policy_loss": 0.004772, "value_loss": 237.942385, "entropy": 0.332991, "total_loss": 118.972634, "clip_frac": 0.085693, "sps": 354.2, "elapsed_s": 2217.4}
+{"update": 431, "steps_done": 882688, "mean_step_reward": 0.952, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 65.414, "mean_ep_length": 68.3, "mean_collisions": 6.03, "mean_intent_rate": 0.797, "policy_loss": 0.002891, "value_loss": 177.720981, "entropy": 0.363424, "total_loss": 88.859747, "clip_frac": 0.07373, "sps": 379.0, "elapsed_s": 2222.8}
+{"update": 432, "steps_done": 884736, "mean_step_reward": 0.9183, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 57.272, "mean_ep_length": 63.6, "mean_collisions": 6.0, "mean_intent_rate": 0.792, "policy_loss": -0.001187, "value_loss": 144.699474, "entropy": 0.369058, "total_loss": 72.344859, "clip_frac": 0.150146, "sps": 400.7, "elapsed_s": 2227.9}
+{"update": 433, "steps_done": 886784, "mean_step_reward": 1.1554, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 72.962, "mean_ep_length": 61.8, "mean_collisions": 3.64, "mean_intent_rate": 0.774, "policy_loss": 0.005635, "value_loss": 112.090473, "entropy": 0.377107, "total_loss": 56.0471, "clip_frac": 0.10022, "sps": 407.5, "elapsed_s": 2232.9}
+{"update": 434, "steps_done": 888832, "mean_step_reward": 0.9773, "min_step_reward": -10.562, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 59.793, "mean_ep_length": 62.2, "mean_collisions": 5.58, "mean_intent_rate": 0.798, "policy_loss": -0.000748, "value_loss": 214.25152, "entropy": 0.372306, "total_loss": 107.121289, "clip_frac": 0.10791, "sps": 391.3, "elapsed_s": 2238.1}
+{"update": 435, "steps_done": 890880, "mean_step_reward": 1.2156, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 75.404, "mean_ep_length": 63.2, "mean_collisions": 4.06, "mean_intent_rate": 0.784, "policy_loss": 0.013439, "value_loss": 159.36045, "entropy": 0.365963, "total_loss": 79.690004, "clip_frac": 0.125732, "sps": 344.8, "elapsed_s": 2244.1}
+{"update": 436, "steps_done": 892928, "mean_step_reward": 0.834, "min_step_reward": -10.582, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 52.301, "mean_ep_length": 60.9, "mean_collisions": 5.97, "mean_intent_rate": 0.795, "policy_loss": 0.000443, "value_loss": 169.105184, "entropy": 0.354645, "total_loss": 84.549488, "clip_frac": 0.124878, "sps": 346.0, "elapsed_s": 2250.0}
+{"update": 437, "steps_done": 894976, "mean_step_reward": 1.2397, "min_step_reward": -10.5882, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 69.07, "mean_ep_length": 55.2, "mean_collisions": 3.38, "mean_intent_rate": 0.796, "policy_loss": 0.018564, "value_loss": 124.605829, "entropy": 0.371728, "total_loss": 62.317761, "clip_frac": 0.15625, "sps": 356.9, "elapsed_s": 2255.7}
+{"update": 438, "steps_done": 897024, "mean_step_reward": 0.7108, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 42.357, "mean_ep_length": 61.1, "mean_collisions": 7.42, "mean_intent_rate": 0.821, "policy_loss": 0.006252, "value_loss": 152.060772, "entropy": 0.357469, "total_loss": 76.033064, "clip_frac": 0.120117, "sps": 345.2, "elapsed_s": 2261.7}
+{"update": 439, "steps_done": 899072, "mean_step_reward": 1.0725, "min_step_reward": -10.5587, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 67.389, "mean_ep_length": 62.1, "mean_collisions": 5.35, "mean_intent_rate": 0.829, "policy_loss": 0.006663, "value_loss": 170.065459, "entropy": 0.35554, "total_loss": 85.035837, "clip_frac": 0.113647, "sps": 357.3, "elapsed_s": 2267.4}
+{"update": 440, "steps_done": 901120, "mean_step_reward": 0.8346, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 52.472, "mean_ep_length": 63.6, "mean_collisions": 7.03, "mean_intent_rate": 0.817, "policy_loss": 0.002576, "value_loss": 199.324647, "entropy": 0.357979, "total_loss": 99.661319, "clip_frac": 0.084717, "sps": 380.3, "elapsed_s": 2272.8}
+{"update": 441, "steps_done": 903168, "mean_step_reward": 1.2654, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 79.138, "mean_ep_length": 62.9, "mean_collisions": 3.81, "mean_intent_rate": 0.805, "policy_loss": 0.013051, "value_loss": 143.037099, "entropy": 0.376722, "total_loss": 71.527832, "clip_frac": 0.135864, "sps": 376.2, "elapsed_s": 2278.2}
+{"update": 442, "steps_done": 905216, "mean_step_reward": 1.1339, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 73.984, "mean_ep_length": 64.1, "mean_collisions": 4.5, "mean_intent_rate": 0.806, "policy_loss": 0.002533, "value_loss": 142.631789, "entropy": 0.370132, "total_loss": 71.314726, "clip_frac": 0.080688, "sps": 358.4, "elapsed_s": 2284.0}
+{"update": 443, "steps_done": 907264, "mean_step_reward": 0.8031, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 49.163, "mean_ep_length": 62.1, "mean_collisions": 6.73, "mean_intent_rate": 0.808, "policy_loss": 0.001007, "value_loss": 207.475374, "entropy": 0.365134, "total_loss": 103.735043, "clip_frac": 0.103638, "sps": 367.6, "elapsed_s": 2289.5}
+{"update": 444, "steps_done": 909312, "mean_step_reward": 0.952, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 62.002, "mean_ep_length": 64.6, "mean_collisions": 5.84, "mean_intent_rate": 0.8, "policy_loss": 0.005167, "value_loss": 174.572405, "entropy": 0.353388, "total_loss": 87.287836, "clip_frac": 0.118896, "sps": 356.6, "elapsed_s": 2295.3}
+{"update": 445, "steps_done": 911360, "mean_step_reward": 0.6866, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 40.01, "mean_ep_length": 58.3, "mean_collisions": 6.46, "mean_intent_rate": 0.776, "policy_loss": 0.029583, "value_loss": 159.028627, "entropy": 0.387629, "total_loss": 79.54002, "clip_frac": 0.274292, "sps": 336.1, "elapsed_s": 2301.4}
+{"update": 446, "steps_done": 913408, "mean_step_reward": 0.8013, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 49.766, "mean_ep_length": 62.8, "mean_collisions": 7.31, "mean_intent_rate": 0.831, "policy_loss": 0.021916, "value_loss": 281.46545, "entropy": 0.348912, "total_loss": 140.751153, "clip_frac": 0.172852, "sps": 384.0, "elapsed_s": 2306.7}
+{"update": 447, "steps_done": 915456, "mean_step_reward": 0.43, "min_step_reward": -10.5873, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 28.357, "mean_ep_length": 62.1, "mean_collisions": 9.41, "mean_intent_rate": 0.84, "policy_loss": -0.004248, "value_loss": 224.605703, "entropy": 0.376394, "total_loss": 112.294839, "clip_frac": 0.125854, "sps": 375.4, "elapsed_s": 2312.2}
+{"update": 448, "steps_done": 917504, "mean_step_reward": 0.8055, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 53.026, "mean_ep_length": 65.0, "mean_collisions": 7.23, "mean_intent_rate": 0.825, "policy_loss": 0.002649, "value_loss": 164.767356, "entropy": 0.376031, "total_loss": 82.382566, "clip_frac": 0.12854, "sps": 366.7, "elapsed_s": 2317.7}
+{"update": 449, "steps_done": 919552, "mean_step_reward": 0.585, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 33.31, "mean_ep_length": 55.8, "mean_collisions": 7.35, "mean_intent_rate": 0.813, "policy_loss": 0.00067, "value_loss": 282.255618, "entropy": 0.386331, "total_loss": 141.124616, "clip_frac": 0.091309, "sps": 354.0, "elapsed_s": 2323.5}
+{"update": 450, "steps_done": 921600, "mean_step_reward": 1.1379, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 76.733, "mean_ep_length": 65.6, "mean_collisions": 4.35, "mean_intent_rate": 0.791, "policy_loss": 0.02078, "value_loss": 140.403504, "entropy": 0.402055, "total_loss": 70.218512, "clip_frac": 0.213745, "sps": 368.6, "elapsed_s": 2329.1}
+{"update": 451, "steps_done": 923648, "mean_step_reward": 0.9023, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 50.535, "mean_ep_length": 59.5, "mean_collisions": 5.71, "mean_intent_rate": 0.782, "policy_loss": 0.00676, "value_loss": 131.633852, "entropy": 0.392853, "total_loss": 65.819758, "clip_frac": 0.093628, "sps": 363.7, "elapsed_s": 2334.7}
+{"update": 452, "steps_done": 925696, "mean_step_reward": 0.8659, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 57.399, "mean_ep_length": 64.8, "mean_collisions": 6.1, "mean_intent_rate": 0.79, "policy_loss": 0.005068, "value_loss": 140.857607, "entropy": 0.377754, "total_loss": 70.430093, "clip_frac": 0.113525, "sps": 383.6, "elapsed_s": 2340.1}
+{"update": 453, "steps_done": 927744, "mean_step_reward": 1.0325, "min_step_reward": -10.5587, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 65.892, "mean_ep_length": 63.8, "mean_collisions": 5.28, "mean_intent_rate": 0.805, "policy_loss": 0.014149, "value_loss": 226.566051, "entropy": 0.319579, "total_loss": 113.29398, "clip_frac": 0.157715, "sps": 371.5, "elapsed_s": 2345.6}
+{"update": 454, "steps_done": 929792, "mean_step_reward": 0.8296, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 50.039, "mean_ep_length": 61.7, "mean_collisions": 6.94, "mean_intent_rate": 0.821, "policy_loss": 0.025072, "value_loss": 182.089697, "entropy": 0.363425, "total_loss": 91.066285, "clip_frac": 0.212036, "sps": 361.2, "elapsed_s": 2351.3}
+{"update": 455, "steps_done": 931840, "mean_step_reward": 1.2588, "min_step_reward": -10.5816, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 73.492, "mean_ep_length": 59.3, "mean_collisions": 3.97, "mean_intent_rate": 0.812, "policy_loss": 0.004294, "value_loss": 149.907405, "entropy": 0.332171, "total_loss": 74.954675, "clip_frac": 0.097168, "sps": 364.8, "elapsed_s": 2356.9}
+{"update": 456, "steps_done": 933888, "mean_step_reward": 1.0165, "min_step_reward": -10.5424, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 61.273, "mean_ep_length": 59.2, "mean_collisions": 5.15, "mean_intent_rate": 0.817, "policy_loss": 0.006756, "value_loss": 213.415787, "entropy": 0.33261, "total_loss": 106.711323, "clip_frac": 0.110596, "sps": 366.2, "elapsed_s": 2362.5}
+{"update": 457, "steps_done": 935936, "mean_step_reward": 0.8806, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 56.425, "mean_ep_length": 63.5, "mean_collisions": 6.36, "mean_intent_rate": 0.814, "policy_loss": 0.006307, "value_loss": 190.048027, "entropy": 0.325486, "total_loss": 95.027066, "clip_frac": 0.102173, "sps": 373.0, "elapsed_s": 2368.0}
+{"update": 458, "steps_done": 937984, "mean_step_reward": 0.9854, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 60.996, "mean_ep_length": 62.1, "mean_collisions": 5.24, "mean_intent_rate": 0.797, "policy_loss": 0.008579, "value_loss": 192.902866, "entropy": 0.3162, "total_loss": 96.456851, "clip_frac": 0.130005, "sps": 382.2, "elapsed_s": 2373.3}
+{"update": 459, "steps_done": 940032, "mean_step_reward": 1.3403, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 87.677, "mean_ep_length": 65.8, "mean_collisions": 3.23, "mean_intent_rate": 0.795, "policy_loss": 0.056542, "value_loss": 133.576208, "entropy": 0.345329, "total_loss": 66.841193, "clip_frac": 0.266846, "sps": 390.8, "elapsed_s": 2378.6}
+{"update": 460, "steps_done": 942080, "mean_step_reward": 0.886, "min_step_reward": -10.5711, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 53.75, "mean_ep_length": 60.0, "mean_collisions": 5.88, "mean_intent_rate": 0.803, "policy_loss": 0.006804, "value_loss": 191.989647, "entropy": 0.365594, "total_loss": 95.997973, "clip_frac": 0.172729, "sps": 382.0, "elapsed_s": 2383.9}
+{"update": 461, "steps_done": 944128, "mean_step_reward": 1.1949, "min_step_reward": -10.5355, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 68.742, "mean_ep_length": 57.7, "mean_collisions": 4.47, "mean_intent_rate": 0.832, "policy_loss": -0.001403, "value_loss": 156.014651, "entropy": 0.315963, "total_loss": 78.002763, "clip_frac": 0.044556, "sps": 375.6, "elapsed_s": 2389.4}
+{"update": 462, "steps_done": 946176, "mean_step_reward": 0.658, "min_step_reward": -10.5888, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 41.176, "mean_ep_length": 62.3, "mean_collisions": 7.88, "mean_intent_rate": 0.821, "policy_loss": 0.008657, "value_loss": 217.928733, "entropy": 0.300475, "total_loss": 108.970019, "clip_frac": 0.094482, "sps": 363.5, "elapsed_s": 2395.0}
+{"update": 463, "steps_done": 948224, "mean_step_reward": 1.2938, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 78.393, "mean_ep_length": 61.2, "mean_collisions": 3.61, "mean_intent_rate": 0.805, "policy_loss": 0.007096, "value_loss": 178.99248, "entropy": 0.334047, "total_loss": 89.499996, "clip_frac": 0.135742, "sps": 373.4, "elapsed_s": 2400.5}
+{"update": 464, "steps_done": 950272, "mean_step_reward": 1.0349, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 61.245, "mean_ep_length": 58.0, "mean_collisions": 4.37, "mean_intent_rate": 0.784, "policy_loss": 0.006087, "value_loss": 139.876029, "entropy": 0.334384, "total_loss": 69.940757, "clip_frac": 0.083374, "sps": 357.6, "elapsed_s": 2406.2}
+{"update": 465, "steps_done": 952320, "mean_step_reward": 1.1578, "min_step_reward": -10.5745, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 67.006, "mean_ep_length": 58.2, "mean_collisions": 3.94, "mean_intent_rate": 0.793, "policy_loss": 0.002453, "value_loss": 130.134748, "entropy": 0.345874, "total_loss": 65.066369, "clip_frac": 0.088623, "sps": 366.3, "elapsed_s": 2411.8}
+{"update": 466, "steps_done": 954368, "mean_step_reward": 1.3172, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 79.893, "mean_ep_length": 62.2, "mean_collisions": 3.75, "mean_intent_rate": 0.812, "policy_loss": 0.013554, "value_loss": 140.793597, "entropy": 0.323414, "total_loss": 70.407117, "clip_frac": 0.124023, "sps": 374.2, "elapsed_s": 2417.3}
+{"update": 467, "steps_done": 956416, "mean_step_reward": 1.2254, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 73.809, "mean_ep_length": 60.0, "mean_collisions": 4.44, "mean_intent_rate": 0.834, "policy_loss": 0.012276, "value_loss": 166.559458, "entropy": 0.281456, "total_loss": 83.289191, "clip_frac": 0.135132, "sps": 378.2, "elapsed_s": 2422.7}
+{"update": 468, "steps_done": 958464, "mean_step_reward": 0.9991, "min_step_reward": -10.5835, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 64.116, "mean_ep_length": 62.1, "mean_collisions": 5.79, "mean_intent_rate": 0.834, "policy_loss": 0.016544, "value_loss": 218.014458, "entropy": 0.280246, "total_loss": 109.02097, "clip_frac": 0.181152, "sps": 384.5, "elapsed_s": 2428.0}
+{"update": 469, "steps_done": 960512, "mean_step_reward": 1.1926, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 77.629, "mean_ep_length": 65.5, "mean_collisions": 3.65, "mean_intent_rate": 0.762, "policy_loss": 0.009601, "value_loss": 168.364526, "entropy": 0.309622, "total_loss": 84.188768, "clip_frac": 0.119263, "sps": 377.5, "elapsed_s": 2433.4}
+{"update": 470, "steps_done": 962560, "mean_step_reward": 1.4554, "min_step_reward": -10.5234, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 90.376, "mean_ep_length": 62.1, "mean_collisions": 1.61, "mean_intent_rate": 0.752, "policy_loss": 0.049096, "value_loss": 144.567775, "entropy": 0.361485, "total_loss": 72.329369, "clip_frac": 0.297852, "sps": 376.5, "elapsed_s": 2438.9}
+{"update": 471, "steps_done": 964608, "mean_step_reward": 1.1187, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 73.701, "mean_ep_length": 65.8, "mean_collisions": 4.45, "mean_intent_rate": 0.784, "policy_loss": 0.031071, "value_loss": 152.926101, "entropy": 0.358721, "total_loss": 76.490535, "clip_frac": 0.244385, "sps": 374.4, "elapsed_s": 2444.4}
+{"update": 472, "steps_done": 966656, "mean_step_reward": 0.8892, "min_step_reward": -10.5676, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 56.563, "mean_ep_length": 62.8, "mean_collisions": 6.76, "mean_intent_rate": 0.839, "policy_loss": 0.017575, "value_loss": 198.992864, "entropy": 0.272676, "total_loss": 99.51128, "clip_frac": 0.141968, "sps": 388.6, "elapsed_s": 2449.6}
+{"update": 473, "steps_done": 968704, "mean_step_reward": 0.7812, "min_step_reward": -10.5739, "max_step_reward": 2.5, "n_episodes_finished": 28, "mean_ep_return": 55.473, "mean_ep_length": 72.1, "mean_collisions": 8.04, "mean_intent_rate": 0.806, "policy_loss": 0.01348, "value_loss": 179.605793, "entropy": 0.324695, "total_loss": 89.81313, "clip_frac": 0.216797, "sps": 379.3, "elapsed_s": 2455.0}
+{"update": 474, "steps_done": 970752, "mean_step_reward": 0.7587, "min_step_reward": -10.5637, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 50.529, "mean_ep_length": 61.2, "mean_collisions": 6.7, "mean_intent_rate": 0.816, "policy_loss": -0.003907, "value_loss": 162.535716, "entropy": 0.289138, "total_loss": 81.26106, "clip_frac": 0.089844, "sps": 365.0, "elapsed_s": 2460.6}
+{"update": 475, "steps_done": 972800, "mean_step_reward": 0.9359, "min_step_reward": -10.5515, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 54.871, "mean_ep_length": 64.3, "mean_collisions": 7.16, "mean_intent_rate": 0.838, "policy_loss": 0.00316, "value_loss": 240.45402, "entropy": 0.264687, "total_loss": 120.227523, "clip_frac": 0.071289, "sps": 367.2, "elapsed_s": 2466.2}
+{"update": 476, "steps_done": 974848, "mean_step_reward": 1.0748, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.459, "mean_ep_length": 62.7, "mean_collisions": 5.27, "mean_intent_rate": 0.829, "policy_loss": 0.012215, "value_loss": 234.866993, "entropy": 0.315306, "total_loss": 117.442558, "clip_frac": 0.126221, "sps": 370.7, "elapsed_s": 2471.7}
+{"update": 477, "steps_done": 976896, "mean_step_reward": 1.2131, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 76.994, "mean_ep_length": 64.1, "mean_collisions": 4.16, "mean_intent_rate": 0.799, "policy_loss": 0.004508, "value_loss": 135.699358, "entropy": 0.317957, "total_loss": 67.851007, "clip_frac": 0.088013, "sps": 370.8, "elapsed_s": 2477.3}
+{"update": 478, "steps_done": 978944, "mean_step_reward": 1.0022, "min_step_reward": -10.5712, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 61.596, "mean_ep_length": 62.2, "mean_collisions": 6.12, "mean_intent_rate": 0.833, "policy_loss": 0.046009, "value_loss": 229.027447, "entropy": 0.247537, "total_loss": 114.557257, "clip_frac": 0.240479, "sps": 358.2, "elapsed_s": 2483.0}
+{"update": 479, "steps_done": 980992, "mean_step_reward": 0.9839, "min_step_reward": -10.5439, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 63.162, "mean_ep_length": 64.1, "mean_collisions": 6.12, "mean_intent_rate": 0.822, "policy_loss": 0.022511, "value_loss": 180.874096, "entropy": 0.31098, "total_loss": 90.45645, "clip_frac": 0.204224, "sps": 352.4, "elapsed_s": 2488.8}
+{"update": 480, "steps_done": 983040, "mean_step_reward": 0.7495, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 44.782, "mean_ep_length": 58.3, "mean_collisions": 7.34, "mean_intent_rate": 0.858, "policy_loss": 0.003903, "value_loss": 206.456971, "entropy": 0.294566, "total_loss": 103.229443, "clip_frac": 0.174438, "sps": 374.3, "elapsed_s": 2494.3}
+{"update": 481, "steps_done": 985088, "mean_step_reward": 0.8599, "min_step_reward": -10.5823, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 53.338, "mean_ep_length": 65.1, "mean_collisions": 7.81, "mean_intent_rate": 0.856, "policy_loss": 0.00188, "value_loss": 184.892874, "entropy": 0.278667, "total_loss": 92.445531, "clip_frac": 0.074951, "sps": 367.3, "elapsed_s": 2499.8}
+{"update": 482, "steps_done": 987136, "mean_step_reward": 0.712, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 48.841, "mean_ep_length": 66.8, "mean_collisions": 8.23, "mean_intent_rate": 0.835, "policy_loss": 0.022326, "value_loss": 217.605212, "entropy": 0.331296, "total_loss": 108.821619, "clip_frac": 0.188354, "sps": 380.7, "elapsed_s": 2505.2}
+{"update": 483, "steps_done": 989184, "mean_step_reward": 1.181, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 70.613, "mean_ep_length": 59.5, "mean_collisions": 4.97, "mean_intent_rate": 0.85, "policy_loss": 0.005771, "value_loss": 147.140994, "entropy": 0.284842, "total_loss": 73.573419, "clip_frac": 0.062378, "sps": 379.8, "elapsed_s": 2510.6}
+{"update": 484, "steps_done": 991232, "mean_step_reward": 0.9604, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 65.956, "mean_ep_length": 64.7, "mean_collisions": 5.81, "mean_intent_rate": 0.818, "policy_loss": 0.005307, "value_loss": 146.530632, "entropy": 0.309517, "total_loss": 73.267528, "clip_frac": 0.111938, "sps": 380.0, "elapsed_s": 2516.0}
+{"update": 485, "steps_done": 993280, "mean_step_reward": 1.0322, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 62.193, "mean_ep_length": 63.3, "mean_collisions": 5.72, "mean_intent_rate": 0.814, "policy_loss": 0.001703, "value_loss": 151.092388, "entropy": 0.320118, "total_loss": 75.544696, "clip_frac": 0.070557, "sps": 368.8, "elapsed_s": 2521.6}
+{"update": 486, "steps_done": 995328, "mean_step_reward": 1.3085, "min_step_reward": -10.5325, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 78.413, "mean_ep_length": 60.2, "mean_collisions": 3.5, "mean_intent_rate": 0.805, "policy_loss": 0.009825, "value_loss": 136.482917, "entropy": 0.306396, "total_loss": 68.248219, "clip_frac": 0.09729, "sps": 379.7, "elapsed_s": 2527.0}
+{"update": 487, "steps_done": 997376, "mean_step_reward": 0.9252, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 59.263, "mean_ep_length": 64.4, "mean_collisions": 6.41, "mean_intent_rate": 0.818, "policy_loss": 0.000264, "value_loss": 171.279902, "entropy": 0.284789, "total_loss": 85.637366, "clip_frac": 0.071167, "sps": 355.0, "elapsed_s": 2532.7}
+{"update": 488, "steps_done": 999424, "mean_step_reward": 1.4333, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 90.838, "mean_ep_length": 63.9, "mean_collisions": 2.59, "mean_intent_rate": 0.785, "policy_loss": 0.026583, "value_loss": 143.230633, "entropy": 0.343845, "total_loss": 71.638462, "clip_frac": 0.181152, "sps": 337.0, "elapsed_s": 2538.8}
+{"update": 489, "steps_done": 1001472, "mean_step_reward": 1.249, "min_step_reward": -10.548, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 83.876, "mean_ep_length": 66.7, "mean_collisions": 3.81, "mean_intent_rate": 0.785, "policy_loss": 0.00484, "value_loss": 170.173449, "entropy": 0.299096, "total_loss": 85.088574, "clip_frac": 0.13147, "sps": 354.8, "elapsed_s": 2544.6}
+{"update": 490, "steps_done": 1003520, "mean_step_reward": 1.3842, "min_step_reward": -10.5293, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 84.717, "mean_ep_length": 61.2, "mean_collisions": 2.94, "mean_intent_rate": 0.804, "policy_loss": 0.003694, "value_loss": 146.299865, "entropy": 0.288269, "total_loss": 73.150744, "clip_frac": 0.053955, "sps": 372.6, "elapsed_s": 2550.1}
+{"update": 491, "steps_done": 1005568, "mean_step_reward": 1.1186, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 70.656, "mean_ep_length": 63.0, "mean_collisions": 5.12, "mean_intent_rate": 0.818, "policy_loss": -0.002072, "value_loss": 184.940741, "entropy": 0.282472, "total_loss": 92.465473, "clip_frac": 0.072754, "sps": 395.0, "elapsed_s": 2555.3}
+{"update": 492, "steps_done": 1007616, "mean_step_reward": 1.1407, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 75.077, "mean_ep_length": 63.5, "mean_collisions": 4.16, "mean_intent_rate": 0.794, "policy_loss": 0.005342, "value_loss": 147.615513, "entropy": 0.298942, "total_loss": 73.81011, "clip_frac": 0.099976, "sps": 406.9, "elapsed_s": 2560.3}
+{"update": 493, "steps_done": 1009664, "mean_step_reward": 1.2849, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 78.02, "mean_ep_length": 62.4, "mean_collisions": 3.15, "mean_intent_rate": 0.772, "policy_loss": 0.007954, "value_loss": 154.807214, "entropy": 0.272582, "total_loss": 77.408835, "clip_frac": 0.092896, "sps": 415.4, "elapsed_s": 2565.2}
+{"update": 494, "steps_done": 1011712, "mean_step_reward": 1.0693, "min_step_reward": -10.5719, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 68.366, "mean_ep_length": 63.0, "mean_collisions": 5.19, "mean_intent_rate": 0.818, "policy_loss": 0.030237, "value_loss": 178.299997, "entropy": 0.28985, "total_loss": 89.177337, "clip_frac": 0.228394, "sps": 416.2, "elapsed_s": 2570.2}
+{"update": 495, "steps_done": 1013760, "mean_step_reward": 0.9603, "min_step_reward": -10.5364, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 58.742, "mean_ep_length": 61.8, "mean_collisions": 6.15, "mean_intent_rate": 0.827, "policy_loss": 0.011466, "value_loss": 246.15732, "entropy": 0.244368, "total_loss": 123.087682, "clip_frac": 0.120117, "sps": 408.0, "elapsed_s": 2575.2}
+{"update": 496, "steps_done": 1015808, "mean_step_reward": 0.8305, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 52.119, "mean_ep_length": 63.6, "mean_collisions": 6.38, "mean_intent_rate": 0.792, "policy_loss": 0.001243, "value_loss": 227.474821, "entropy": 0.271921, "total_loss": 113.735934, "clip_frac": 0.103394, "sps": 402.1, "elapsed_s": 2580.3}
+{"update": 497, "steps_done": 1017856, "mean_step_reward": 1.247, "min_step_reward": -10.5752, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 78.945, "mean_ep_length": 63.4, "mean_collisions": 4.81, "mean_intent_rate": 0.85, "policy_loss": 0.005308, "value_loss": 159.962514, "entropy": 0.269723, "total_loss": 79.983867, "clip_frac": 0.06958, "sps": 411.2, "elapsed_s": 2585.2}
+{"update": 498, "steps_done": 1019904, "mean_step_reward": 1.1215, "min_step_reward": -10.5819, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.724, "mean_ep_length": 60.9, "mean_collisions": 4.58, "mean_intent_rate": 0.809, "policy_loss": -0.001218, "value_loss": 172.143484, "entropy": 0.249817, "total_loss": 86.068025, "clip_frac": 0.079102, "sps": 415.2, "elapsed_s": 2590.2}
+{"update": 499, "steps_done": 1021952, "mean_step_reward": 1.2768, "min_step_reward": -10.5531, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 77.818, "mean_ep_length": 62.0, "mean_collisions": 4.53, "mean_intent_rate": 0.835, "policy_loss": 0.001653, "value_loss": 212.122897, "entropy": 0.250725, "total_loss": 106.060594, "clip_frac": 0.073608, "sps": 416.6, "elapsed_s": 2595.1}
+{"update": 500, "steps_done": 1024000, "mean_step_reward": 1.3551, "min_step_reward": -10.5893, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 85.316, "mean_ep_length": 62.9, "mean_collisions": 3.78, "mean_intent_rate": 0.836, "policy_loss": 0.003846, "value_loss": 175.45621, "entropy": 0.239128, "total_loss": 87.72956, "clip_frac": 0.064941, "sps": 413.2, "elapsed_s": 2600.1}
+{"update": 501, "steps_done": 1026048, "mean_step_reward": 1.1995, "min_step_reward": -10.5836, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 70.281, "mean_ep_length": 57.2, "mean_collisions": 4.56, "mean_intent_rate": 0.857, "policy_loss": 0.003863, "value_loss": 205.610718, "entropy": 0.234224, "total_loss": 102.80688, "clip_frac": 0.065186, "sps": 411.9, "elapsed_s": 2605.0}
+{"update": 502, "steps_done": 1028096, "mean_step_reward": 1.658, "min_step_reward": -10.5597, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 102.13, "mean_ep_length": 62.5, "mean_collisions": 2.33, "mean_intent_rate": 0.846, "policy_loss": 0.010935, "value_loss": 163.413617, "entropy": 0.22637, "total_loss": 81.715479, "clip_frac": 0.098999, "sps": 416.6, "elapsed_s": 2610.0}
+{"update": 503, "steps_done": 1030144, "mean_step_reward": 1.1141, "min_step_reward": -10.5737, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 73.827, "mean_ep_length": 66.5, "mean_collisions": 5.55, "mean_intent_rate": 0.826, "policy_loss": 0.013738, "value_loss": 218.34198, "entropy": 0.239571, "total_loss": 109.182332, "clip_frac": 0.13916, "sps": 412.8, "elapsed_s": 2614.9}
+{"update": 504, "steps_done": 1032192, "mean_step_reward": 1.1518, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 68.991, "mean_ep_length": 59.6, "mean_collisions": 5.06, "mean_intent_rate": 0.849, "policy_loss": 0.001948, "value_loss": 212.408975, "entropy": 0.246379, "total_loss": 106.203972, "clip_frac": 0.158081, "sps": 415.4, "elapsed_s": 2619.8}
+{"update": 505, "steps_done": 1034240, "mean_step_reward": 0.9418, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 65.333, "mean_ep_length": 65.6, "mean_collisions": 5.94, "mean_intent_rate": 0.823, "policy_loss": 0.002291, "value_loss": 195.739448, "entropy": 0.255705, "total_loss": 97.869458, "clip_frac": 0.078735, "sps": 401.6, "elapsed_s": 2624.9}
+{"update": 506, "steps_done": 1036288, "mean_step_reward": 0.5602, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 33.303, "mean_ep_length": 63.6, "mean_collisions": 9.09, "mean_intent_rate": 0.83, "policy_loss": 0.008141, "value_loss": 251.420421, "entropy": 0.278004, "total_loss": 125.715571, "clip_frac": 0.169434, "sps": 417.1, "elapsed_s": 2629.9}
+{"update": 507, "steps_done": 1038336, "mean_step_reward": 1.3722, "min_step_reward": -10.5784, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 88.785, "mean_ep_length": 64.3, "mean_collisions": 3.26, "mean_intent_rate": 0.809, "policy_loss": 0.025227, "value_loss": 141.678245, "entropy": 0.275452, "total_loss": 70.861595, "clip_frac": 0.185425, "sps": 419.4, "elapsed_s": 2634.7}
+{"update": 508, "steps_done": 1040384, "mean_step_reward": 1.1292, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.863, "mean_ep_length": 62.4, "mean_collisions": 5.58, "mean_intent_rate": 0.85, "policy_loss": 0.00051, "value_loss": 176.551616, "entropy": 0.242677, "total_loss": 88.273892, "clip_frac": 0.058105, "sps": 412.6, "elapsed_s": 2639.7}
+{"update": 509, "steps_done": 1042432, "mean_step_reward": 1.0589, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 69.232, "mean_ep_length": 65.7, "mean_collisions": 6.1, "mean_intent_rate": 0.843, "policy_loss": 0.007533, "value_loss": 178.883821, "entropy": 0.259249, "total_loss": 89.446851, "clip_frac": 0.107666, "sps": 408.4, "elapsed_s": 2644.7}
+{"update": 510, "steps_done": 1044480, "mean_step_reward": 0.9519, "min_step_reward": -10.5777, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 57.219, "mean_ep_length": 60.8, "mean_collisions": 6.94, "mean_intent_rate": 0.876, "policy_loss": 0.037041, "value_loss": 248.401832, "entropy": 0.186053, "total_loss": 124.236096, "clip_frac": 0.15564, "sps": 415.8, "elapsed_s": 2649.7}
+{"update": 511, "steps_done": 1046528, "mean_step_reward": 1.0805, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 67.132, "mean_ep_length": 61.1, "mean_collisions": 5.82, "mean_intent_rate": 0.864, "policy_loss": 0.006368, "value_loss": 206.87601, "entropy": 0.219705, "total_loss": 103.442176, "clip_frac": 0.096313, "sps": 413.3, "elapsed_s": 2654.6}
+{"update": 512, "steps_done": 1048576, "mean_step_reward": 0.9841, "min_step_reward": -10.5798, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 58.1, "mean_ep_length": 60.8, "mean_collisions": 6.42, "mean_intent_rate": 0.853, "policy_loss": -0.00151, "value_loss": 189.129741, "entropy": 0.223842, "total_loss": 94.561122, "clip_frac": 0.060913, "sps": 412.9, "elapsed_s": 2659.6}
+{"update": 513, "steps_done": 1050624, "mean_step_reward": 0.8826, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 58.396, "mean_ep_length": 64.7, "mean_collisions": 7.03, "mean_intent_rate": 0.843, "policy_loss": 0.006784, "value_loss": 246.808947, "entropy": 0.216927, "total_loss": 123.409088, "clip_frac": 0.098755, "sps": 377.5, "elapsed_s": 2665.0}
+{"update": 514, "steps_done": 1052672, "mean_step_reward": 1.0297, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 68.009, "mean_ep_length": 64.1, "mean_collisions": 5.41, "mean_intent_rate": 0.811, "policy_loss": 0.033188, "value_loss": 160.969811, "entropy": 0.270277, "total_loss": 80.51539, "clip_frac": 0.203857, "sps": 358.6, "elapsed_s": 2670.7}
+{"update": 515, "steps_done": 1054720, "mean_step_reward": 0.8749, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 57.039, "mean_ep_length": 65.8, "mean_collisions": 7.45, "mean_intent_rate": 0.852, "policy_loss": 0.003572, "value_loss": 200.232101, "entropy": 0.261967, "total_loss": 100.117003, "clip_frac": 0.12793, "sps": 370.9, "elapsed_s": 2676.2}
+{"update": 516, "steps_done": 1056768, "mean_step_reward": 1.2135, "min_step_reward": -10.5427, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 70.128, "mean_ep_length": 57.7, "mean_collisions": 4.69, "mean_intent_rate": 0.854, "policy_loss": 0.001917, "value_loss": 154.252963, "entropy": 0.246693, "total_loss": 77.125932, "clip_frac": 0.044922, "sps": 397.7, "elapsed_s": 2681.4}
+{"update": 517, "steps_done": 1058816, "mean_step_reward": 1.3609, "min_step_reward": -10.5644, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 80.858, "mean_ep_length": 60.5, "mean_collisions": 3.76, "mean_intent_rate": 0.837, "policy_loss": 0.01258, "value_loss": 178.572217, "entropy": 0.251349, "total_loss": 89.296175, "clip_frac": 0.127808, "sps": 413.2, "elapsed_s": 2686.3}
+{"update": 518, "steps_done": 1060864, "mean_step_reward": 1.2689, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 80.044, "mean_ep_length": 64.1, "mean_collisions": 5.16, "mean_intent_rate": 0.867, "policy_loss": 0.002353, "value_loss": 208.749685, "entropy": 0.232079, "total_loss": 104.374875, "clip_frac": 0.056641, "sps": 418.8, "elapsed_s": 2691.2}
+{"update": 519, "steps_done": 1062912, "mean_step_reward": 1.1722, "min_step_reward": -10.5516, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 77.87, "mean_ep_length": 65.7, "mean_collisions": 5.45, "mean_intent_rate": 0.853, "policy_loss": 0.020654, "value_loss": 201.982841, "entropy": 0.203601, "total_loss": 101.010038, "clip_frac": 0.147583, "sps": 409.1, "elapsed_s": 2696.2}
+{"update": 520, "steps_done": 1064960, "mean_step_reward": 1.3583, "min_step_reward": -10.5626, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 89.124, "mean_ep_length": 65.5, "mean_collisions": 3.94, "mean_intent_rate": 0.833, "policy_loss": 0.013165, "value_loss": 179.735876, "entropy": 0.247014, "total_loss": 89.878633, "clip_frac": 0.112061, "sps": 389.6, "elapsed_s": 2701.5}
+{"update": 521, "steps_done": 1067008, "mean_step_reward": 1.5934, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 98.809, "mean_ep_length": 62.0, "mean_collisions": 2.64, "mean_intent_rate": 0.85, "policy_loss": 0.011548, "value_loss": 176.959526, "entropy": 0.206095, "total_loss": 88.489249, "clip_frac": 0.088257, "sps": 394.8, "elapsed_s": 2706.7}
+{"update": 522, "steps_done": 1069056, "mean_step_reward": 1.4949, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 90.99, "mean_ep_length": 61.3, "mean_collisions": 2.76, "mean_intent_rate": 0.817, "policy_loss": 0.013119, "value_loss": 171.006623, "entropy": 0.256247, "total_loss": 85.513867, "clip_frac": 0.123413, "sps": 403.6, "elapsed_s": 2711.8}
+{"update": 523, "steps_done": 1071104, "mean_step_reward": 1.391, "min_step_reward": -10.5727, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 95.722, "mean_ep_length": 65.1, "mean_collisions": 3.65, "mean_intent_rate": 0.851, "policy_loss": 0.014715, "value_loss": 227.209143, "entropy": 0.233781, "total_loss": 113.61695, "clip_frac": 0.105835, "sps": 412.4, "elapsed_s": 2716.7}
+{"update": 524, "steps_done": 1073152, "mean_step_reward": 1.0903, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 66.691, "mean_ep_length": 64.1, "mean_collisions": 6.7, "mean_intent_rate": 0.874, "policy_loss": 0.000747, "value_loss": 255.689336, "entropy": 0.212366, "total_loss": 127.843292, "clip_frac": 0.095947, "sps": 393.9, "elapsed_s": 2721.9}
+{"update": 525, "steps_done": 1075200, "mean_step_reward": 1.2615, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 76.281, "mean_ep_length": 60.4, "mean_collisions": 4.41, "mean_intent_rate": 0.842, "policy_loss": 0.005908, "value_loss": 204.876154, "entropy": 0.217209, "total_loss": 102.441812, "clip_frac": 0.06543, "sps": 352.0, "elapsed_s": 2727.7}
+{"update": 526, "steps_done": 1077248, "mean_step_reward": 0.8587, "min_step_reward": -10.5857, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 51.61, "mean_ep_length": 61.1, "mean_collisions": 7.3, "mean_intent_rate": 0.86, "policy_loss": 0.015064, "value_loss": 278.822838, "entropy": 0.204501, "total_loss": 139.424438, "clip_frac": 0.123169, "sps": 366.2, "elapsed_s": 2733.3}
+{"update": 527, "steps_done": 1079296, "mean_step_reward": 1.4267, "min_step_reward": -10.5672, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 96.005, "mean_ep_length": 66.2, "mean_collisions": 3.0, "mean_intent_rate": 0.815, "policy_loss": 0.011175, "value_loss": 184.304129, "entropy": 0.251664, "total_loss": 92.160724, "clip_frac": 0.104614, "sps": 373.2, "elapsed_s": 2738.8}
+{"update": 528, "steps_done": 1081344, "mean_step_reward": 1.4439, "min_step_reward": -10.5725, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 87.127, "mean_ep_length": 61.6, "mean_collisions": 3.58, "mean_intent_rate": 0.846, "policy_loss": 0.014135, "value_loss": 188.473802, "entropy": 0.249815, "total_loss": 94.248538, "clip_frac": 0.117554, "sps": 405.5, "elapsed_s": 2743.9}
+{"update": 529, "steps_done": 1083392, "mean_step_reward": 1.1803, "min_step_reward": -10.5682, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 77.294, "mean_ep_length": 64.8, "mean_collisions": 5.03, "mean_intent_rate": 0.841, "policy_loss": 0.012715, "value_loss": 156.872294, "entropy": 0.239979, "total_loss": 78.446463, "clip_frac": 0.119751, "sps": 407.8, "elapsed_s": 2748.9}
+{"update": 530, "steps_done": 1085440, "mean_step_reward": 0.9858, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 65.738, "mean_ep_length": 65.1, "mean_collisions": 6.77, "mean_intent_rate": 0.863, "policy_loss": 0.002796, "value_loss": 235.672531, "entropy": 0.247315, "total_loss": 117.836588, "clip_frac": 0.112305, "sps": 401.0, "elapsed_s": 2754.0}
+{"update": 531, "steps_done": 1087488, "mean_step_reward": 1.4387, "min_step_reward": -10.5143, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 90.862, "mean_ep_length": 65.0, "mean_collisions": 4.06, "mean_intent_rate": 0.852, "policy_loss": 0.008127, "value_loss": 174.400514, "entropy": 0.23789, "total_loss": 87.206005, "clip_frac": 0.112061, "sps": 409.0, "elapsed_s": 2759.0}
+{"update": 532, "steps_done": 1089536, "mean_step_reward": 1.3904, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 92.952, "mean_ep_length": 67.4, "mean_collisions": 4.9, "mean_intent_rate": 0.88, "policy_loss": 0.002759, "value_loss": 196.969277, "entropy": 0.210344, "total_loss": 98.485293, "clip_frac": 0.05249, "sps": 409.2, "elapsed_s": 2764.0}
+{"update": 533, "steps_done": 1091584, "mean_step_reward": 1.3084, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 82.926, "mean_ep_length": 60.8, "mean_collisions": 3.53, "mean_intent_rate": 0.825, "policy_loss": 0.004332, "value_loss": 188.76381, "entropy": 0.246963, "total_loss": 94.383768, "clip_frac": 0.078125, "sps": 414.7, "elapsed_s": 2769.0}
+{"update": 534, "steps_done": 1093632, "mean_step_reward": 1.1166, "min_step_reward": -10.5877, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 71.762, "mean_ep_length": 66.8, "mean_collisions": 5.9, "mean_intent_rate": 0.834, "policy_loss": 0.006829, "value_loss": 218.661995, "entropy": 0.245038, "total_loss": 109.335376, "clip_frac": 0.084595, "sps": 414.4, "elapsed_s": 2773.9}
+{"update": 535, "steps_done": 1095680, "mean_step_reward": 1.5256, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 96.986, "mean_ep_length": 63.8, "mean_collisions": 2.7, "mean_intent_rate": 0.829, "policy_loss": 0.042876, "value_loss": 176.479992, "entropy": 0.277862, "total_loss": 88.280093, "clip_frac": 0.241211, "sps": 412.3, "elapsed_s": 2778.9}
+{"update": 536, "steps_done": 1097728, "mean_step_reward": 0.6671, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 44.183, "mean_ep_length": 63.7, "mean_collisions": 8.94, "mean_intent_rate": 0.881, "policy_loss": 0.006296, "value_loss": 280.129166, "entropy": 0.250213, "total_loss": 140.068377, "clip_frac": 0.131714, "sps": 415.1, "elapsed_s": 2783.8}
+{"update": 537, "steps_done": 1099776, "mean_step_reward": 1.4573, "min_step_reward": -10.5437, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 87.673, "mean_ep_length": 62.2, "mean_collisions": 4.12, "mean_intent_rate": 0.873, "policy_loss": 0.002143, "value_loss": 179.527333, "entropy": 0.206658, "total_loss": 89.763743, "clip_frac": 0.060303, "sps": 410.9, "elapsed_s": 2788.8}
+{"update": 538, "steps_done": 1101824, "mean_step_reward": 1.2588, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 74.604, "mean_ep_length": 59.8, "mean_collisions": 4.71, "mean_intent_rate": 0.86, "policy_loss": 0.007572, "value_loss": 179.68528, "entropy": 0.231217, "total_loss": 89.8479, "clip_frac": 0.110962, "sps": 409.1, "elapsed_s": 2793.8}
+{"update": 539, "steps_done": 1103872, "mean_step_reward": 1.092, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 64.492, "mean_ep_length": 58.9, "mean_collisions": 5.86, "mean_intent_rate": 0.879, "policy_loss": 0.006732, "value_loss": 246.388742, "entropy": 0.214225, "total_loss": 123.19896, "clip_frac": 0.129517, "sps": 405.7, "elapsed_s": 2798.8}
+{"update": 540, "steps_done": 1105920, "mean_step_reward": 1.4017, "min_step_reward": -10.5775, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 99.331, "mean_ep_length": 70.8, "mean_collisions": 4.83, "mean_intent_rate": 0.873, "policy_loss": 0.011725, "value_loss": 190.986279, "entropy": 0.178579, "total_loss": 95.503079, "clip_frac": 0.081421, "sps": 410.1, "elapsed_s": 2803.8}
+{"update": 541, "steps_done": 1107968, "mean_step_reward": 1.3583, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 84.908, "mean_ep_length": 62.0, "mean_collisions": 4.22, "mean_intent_rate": 0.861, "policy_loss": 0.017565, "value_loss": 210.071842, "entropy": 0.217242, "total_loss": 105.051313, "clip_frac": 0.140259, "sps": 410.7, "elapsed_s": 2808.8}
+{"update": 542, "steps_done": 1110016, "mean_step_reward": 1.2981, "min_step_reward": -10.5802, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 83.207, "mean_ep_length": 64.3, "mean_collisions": 4.79, "mean_intent_rate": 0.857, "policy_loss": 0.00268, "value_loss": 203.006719, "entropy": 0.200732, "total_loss": 101.504032, "clip_frac": 0.059937, "sps": 415.5, "elapsed_s": 2813.8}
+{"update": 543, "steps_done": 1112064, "mean_step_reward": 1.1215, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.595, "mean_ep_length": 62.1, "mean_collisions": 6.09, "mean_intent_rate": 0.879, "policy_loss": 0.000237, "value_loss": 242.110398, "entropy": 0.20264, "total_loss": 121.053409, "clip_frac": 0.106934, "sps": 412.0, "elapsed_s": 2818.7}
+{"update": 544, "steps_done": 1114112, "mean_step_reward": 1.3281, "min_step_reward": -10.5775, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 81.683, "mean_ep_length": 61.7, "mean_collisions": 5.15, "mean_intent_rate": 0.903, "policy_loss": 0.007546, "value_loss": 257.123363, "entropy": 0.178216, "total_loss": 128.567446, "clip_frac": 0.068848, "sps": 408.8, "elapsed_s": 2823.7}
+{"update": 545, "steps_done": 1116160, "mean_step_reward": 1.4606, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 88.479, "mean_ep_length": 60.5, "mean_collisions": 3.44, "mean_intent_rate": 0.855, "policy_loss": 0.011146, "value_loss": 205.081639, "entropy": 0.227602, "total_loss": 102.549689, "clip_frac": 0.119263, "sps": 405.8, "elapsed_s": 2828.8}
+{"update": 546, "steps_done": 1118208, "mean_step_reward": 1.3246, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 77.794, "mean_ep_length": 58.6, "mean_collisions": 3.83, "mean_intent_rate": 0.843, "policy_loss": 0.00676, "value_loss": 152.253656, "entropy": 0.226623, "total_loss": 76.131322, "clip_frac": 0.07605, "sps": 411.7, "elapsed_s": 2833.8}
+{"update": 547, "steps_done": 1120256, "mean_step_reward": 1.3613, "min_step_reward": -10.5883, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 81.708, "mean_ep_length": 60.9, "mean_collisions": 4.0, "mean_intent_rate": 0.848, "policy_loss": 0.004525, "value_loss": 205.295804, "entropy": 0.234296, "total_loss": 102.650084, "clip_frac": 0.053345, "sps": 404.1, "elapsed_s": 2838.8}
+{"update": 548, "steps_done": 1122304, "mean_step_reward": 1.3471, "min_step_reward": -10.5379, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 89.387, "mean_ep_length": 65.3, "mean_collisions": 3.53, "mean_intent_rate": 0.82, "policy_loss": 0.008871, "value_loss": 167.397139, "entropy": 0.265009, "total_loss": 83.704791, "clip_frac": 0.14209, "sps": 413.9, "elapsed_s": 2843.8}
+{"update": 549, "steps_done": 1124352, "mean_step_reward": 1.4226, "min_step_reward": -10.5263, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 85.348, "mean_ep_length": 60.2, "mean_collisions": 3.03, "mean_intent_rate": 0.818, "policy_loss": 0.003659, "value_loss": 156.441742, "entropy": 0.224051, "total_loss": 78.222289, "clip_frac": 0.054688, "sps": 410.3, "elapsed_s": 2848.8}
+{"update": 550, "steps_done": 1126400, "mean_step_reward": 1.3201, "min_step_reward": -10.537, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 87.975, "mean_ep_length": 66.1, "mean_collisions": 4.0, "mean_intent_rate": 0.829, "policy_loss": 0.011793, "value_loss": 184.054133, "entropy": 0.237953, "total_loss": 92.03648, "clip_frac": 0.118896, "sps": 415.6, "elapsed_s": 2853.7}
+{"update": 551, "steps_done": 1128448, "mean_step_reward": 1.5646, "min_step_reward": -10.5824, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 96.179, "mean_ep_length": 61.9, "mean_collisions": 3.38, "mean_intent_rate": 0.877, "policy_loss": 0.005074, "value_loss": 196.868948, "entropy": 0.191919, "total_loss": 98.43763, "clip_frac": 0.059937, "sps": 413.5, "elapsed_s": 2858.7}
+{"update": 552, "steps_done": 1130496, "mean_step_reward": 1.6061, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 101.775, "mean_ep_length": 63.6, "mean_collisions": 2.72, "mean_intent_rate": 0.856, "policy_loss": 0.001531, "value_loss": 189.480265, "entropy": 0.205442, "total_loss": 94.739609, "clip_frac": 0.052368, "sps": 414.1, "elapsed_s": 2863.6}
+{"update": 553, "steps_done": 1132544, "mean_step_reward": 1.1135, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.863, "mean_ep_length": 62.4, "mean_collisions": 5.73, "mean_intent_rate": 0.862, "policy_loss": 0.008944, "value_loss": 254.002322, "entropy": 0.213108, "total_loss": 127.007974, "clip_frac": 0.098633, "sps": 411.8, "elapsed_s": 2868.6}
+{"update": 554, "steps_done": 1134592, "mean_step_reward": 1.3765, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 86.552, "mean_ep_length": 63.3, "mean_collisions": 4.28, "mean_intent_rate": 0.859, "policy_loss": 0.044868, "value_loss": 199.469119, "entropy": 0.218276, "total_loss": 99.777244, "clip_frac": 0.109619, "sps": 410.1, "elapsed_s": 2873.6}
+{"update": 555, "steps_done": 1136640, "mean_step_reward": 1.5333, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 93.884, "mean_ep_length": 60.9, "mean_collisions": 3.47, "mean_intent_rate": 0.885, "policy_loss": 0.000965, "value_loss": 183.327538, "entropy": 0.193623, "total_loss": 91.662798, "clip_frac": 0.082397, "sps": 414.3, "elapsed_s": 2878.5}
+{"update": 556, "steps_done": 1138688, "mean_step_reward": 1.3296, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 81.226, "mean_ep_length": 61.5, "mean_collisions": 4.61, "mean_intent_rate": 0.871, "policy_loss": 0.004248, "value_loss": 188.433936, "entropy": 0.195744, "total_loss": 94.219259, "clip_frac": 0.080322, "sps": 406.7, "elapsed_s": 2883.6}
+{"update": 557, "steps_done": 1140736, "mean_step_reward": 1.2952, "min_step_reward": -10.531, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 75.745, "mean_ep_length": 57.4, "mean_collisions": 4.69, "mean_intent_rate": 0.886, "policy_loss": 0.004575, "value_loss": 235.170098, "entropy": 0.175507, "total_loss": 117.587869, "clip_frac": 0.051636, "sps": 413.4, "elapsed_s": 2888.5}
+{"update": 558, "steps_done": 1142784, "mean_step_reward": 1.0567, "min_step_reward": -10.5811, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 64.285, "mean_ep_length": 61.5, "mean_collisions": 6.64, "mean_intent_rate": 0.884, "policy_loss": 0.003008, "value_loss": 221.953667, "entropy": 0.163916, "total_loss": 110.978203, "clip_frac": 0.044678, "sps": 414.2, "elapsed_s": 2893.5}
+{"update": 559, "steps_done": 1144832, "mean_step_reward": 1.1424, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 76.952, "mean_ep_length": 67.9, "mean_collisions": 7.0, "mean_intent_rate": 0.903, "policy_loss": 0.003346, "value_loss": 305.784552, "entropy": 0.174146, "total_loss": 152.89388, "clip_frac": 0.09375, "sps": 416.8, "elapsed_s": 2898.4}
+{"update": 560, "steps_done": 1146880, "mean_step_reward": 0.9478, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 64.028, "mean_ep_length": 66.7, "mean_collisions": 8.13, "mean_intent_rate": 0.904, "policy_loss": 0.002241, "value_loss": 309.431633, "entropy": 0.168278, "total_loss": 154.716375, "clip_frac": 0.075439, "sps": 414.4, "elapsed_s": 2903.3}
+{"update": 561, "steps_done": 1148928, "mean_step_reward": 1.2186, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 75.985, "mean_ep_length": 62.2, "mean_collisions": 5.73, "mean_intent_rate": 0.898, "policy_loss": -0.001408, "value_loss": 229.219491, "entropy": 0.186303, "total_loss": 114.606474, "clip_frac": 0.078247, "sps": 410.6, "elapsed_s": 2908.3}
+{"update": 562, "steps_done": 1150976, "mean_step_reward": 0.5911, "min_step_reward": -10.5793, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 36.912, "mean_ep_length": 63.6, "mean_collisions": 9.62, "mean_intent_rate": 0.874, "policy_loss": 0.007174, "value_loss": 298.150511, "entropy": 0.226336, "total_loss": 149.080165, "clip_frac": 0.111694, "sps": 417.1, "elapsed_s": 2913.2}
+{"update": 563, "steps_done": 1153024, "mean_step_reward": 1.1619, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 70.63, "mean_ep_length": 60.5, "mean_collisions": 6.0, "mean_intent_rate": 0.898, "policy_loss": 0.003352, "value_loss": 229.743334, "entropy": 0.175234, "total_loss": 114.873266, "clip_frac": 0.058105, "sps": 413.0, "elapsed_s": 2918.2}
+{"update": 564, "steps_done": 1155072, "mean_step_reward": 1.6147, "min_step_reward": -10.5876, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 99.903, "mean_ep_length": 61.9, "mean_collisions": 3.0, "mean_intent_rate": 0.877, "policy_loss": 0.013127, "value_loss": 178.411346, "entropy": 0.199321, "total_loss": 89.216807, "clip_frac": 0.086792, "sps": 404.2, "elapsed_s": 2923.2}
+{"update": 565, "steps_done": 1157120, "mean_step_reward": 1.5412, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 98.249, "mean_ep_length": 63.8, "mean_collisions": 4.09, "mean_intent_rate": 0.91, "policy_loss": -0.000178, "value_loss": 215.952008, "entropy": 0.170559, "total_loss": 107.974121, "clip_frac": 0.065186, "sps": 412.4, "elapsed_s": 2928.2}
+{"update": 566, "steps_done": 1159168, "mean_step_reward": 1.5887, "min_step_reward": -10.5898, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 95.411, "mean_ep_length": 60.1, "mean_collisions": 3.18, "mean_intent_rate": 0.888, "policy_loss": 0.010434, "value_loss": 185.458992, "entropy": 0.171196, "total_loss": 92.738219, "clip_frac": 0.070312, "sps": 411.9, "elapsed_s": 2933.2}
+{"update": 567, "steps_done": 1161216, "mean_step_reward": 1.2999, "min_step_reward": -10.5818, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 79.007, "mean_ep_length": 61.4, "mean_collisions": 5.42, "mean_intent_rate": 0.905, "policy_loss": 0.007483, "value_loss": 240.878141, "entropy": 0.164294, "total_loss": 120.444911, "clip_frac": 0.060913, "sps": 415.4, "elapsed_s": 2938.1}
+{"update": 568, "steps_done": 1163264, "mean_step_reward": 1.0885, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 69.103, "mean_ep_length": 62.3, "mean_collisions": 6.52, "mean_intent_rate": 0.898, "policy_loss": 0.003702, "value_loss": 245.958568, "entropy": 0.187422, "total_loss": 122.981112, "clip_frac": 0.114014, "sps": 417.7, "elapsed_s": 2943.0}
+{"update": 569, "steps_done": 1165312, "mean_step_reward": 1.3806, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 85.802, "mean_ep_length": 62.5, "mean_collisions": 5.0, "mean_intent_rate": 0.904, "policy_loss": 0.00183, "value_loss": 258.335226, "entropy": 0.162465, "total_loss": 129.167819, "clip_frac": 0.055298, "sps": 415.9, "elapsed_s": 2947.9}
+{"update": 570, "steps_done": 1167360, "mean_step_reward": 1.556, "min_step_reward": -10.5813, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 100.138, "mean_ep_length": 64.2, "mean_collisions": 4.28, "mean_intent_rate": 0.921, "policy_loss": 0.005415, "value_loss": 219.150633, "entropy": 0.152408, "total_loss": 109.579208, "clip_frac": 0.057617, "sps": 412.3, "elapsed_s": 2952.9}
+{"update": 571, "steps_done": 1169408, "mean_step_reward": 1.7251, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 105.156, "mean_ep_length": 59.8, "mean_collisions": 2.45, "mean_intent_rate": 0.896, "policy_loss": 0.007421, "value_loss": 199.587508, "entropy": 0.145084, "total_loss": 99.799724, "clip_frac": 0.048828, "sps": 407.5, "elapsed_s": 2957.9}
+{"update": 572, "steps_done": 1171456, "mean_step_reward": 1.3121, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 86.46, "mean_ep_length": 66.5, "mean_collisions": 5.25, "mean_intent_rate": 0.879, "policy_loss": 0.003495, "value_loss": 246.061274, "entropy": 0.162923, "total_loss": 123.032503, "clip_frac": 0.062012, "sps": 411.0, "elapsed_s": 2962.9}
+{"update": 573, "steps_done": 1173504, "mean_step_reward": 1.3432, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 82.817, "mean_ep_length": 61.3, "mean_collisions": 3.67, "mean_intent_rate": 0.837, "policy_loss": 0.016685, "value_loss": 193.46604, "entropy": 0.197423, "total_loss": 96.747732, "clip_frac": 0.110596, "sps": 407.2, "elapsed_s": 2968.0}
+{"update": 574, "steps_done": 1175552, "mean_step_reward": 1.4951, "min_step_reward": -10.5267, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 86.071, "mean_ep_length": 58.5, "mean_collisions": 3.97, "mean_intent_rate": 0.897, "policy_loss": 0.009057, "value_loss": 175.988497, "entropy": 0.192927, "total_loss": 88.001375, "clip_frac": 0.093262, "sps": 411.6, "elapsed_s": 2972.9}
+{"update": 575, "steps_done": 1177600, "mean_step_reward": 1.5067, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 96.338, "mean_ep_length": 63.9, "mean_collisions": 4.03, "mean_intent_rate": 0.896, "policy_loss": 0.019639, "value_loss": 200.882798, "entropy": 0.153662, "total_loss": 100.459502, "clip_frac": 0.125, "sps": 415.3, "elapsed_s": 2977.9}
+{"update": 576, "steps_done": 1179648, "mean_step_reward": 1.3687, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 88.346, "mean_ep_length": 64.3, "mean_collisions": 5.27, "mean_intent_rate": 0.909, "policy_loss": 0.000677, "value_loss": 230.35052, "entropy": 0.158058, "total_loss": 115.174356, "clip_frac": 0.021851, "sps": 416.5, "elapsed_s": 2982.8}
+{"update": 577, "steps_done": 1181696, "mean_step_reward": 1.3899, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 98.204, "mean_ep_length": 68.1, "mean_collisions": 4.7, "mean_intent_rate": 0.889, "policy_loss": -0.000236, "value_loss": 278.899817, "entropy": 0.171359, "total_loss": 139.447959, "clip_frac": 0.050049, "sps": 408.3, "elapsed_s": 2987.8}
+{"update": 578, "steps_done": 1183744, "mean_step_reward": 1.182, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 71.755, "mean_ep_length": 62.8, "mean_collisions": 5.67, "mean_intent_rate": 0.86, "policy_loss": 0.001613, "value_loss": 258.046668, "entropy": 0.180376, "total_loss": 129.023144, "clip_frac": 0.045898, "sps": 408.9, "elapsed_s": 2992.8}
+{"update": 579, "steps_done": 1185792, "mean_step_reward": 1.4395, "min_step_reward": -10.545, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 88.743, "mean_ep_length": 60.8, "mean_collisions": 4.42, "mean_intent_rate": 0.909, "policy_loss": 0.001284, "value_loss": 232.987222, "entropy": 0.156869, "total_loss": 116.493327, "clip_frac": 0.045044, "sps": 414.2, "elapsed_s": 2997.7}
+{"update": 580, "steps_done": 1187840, "mean_step_reward": 1.2467, "min_step_reward": -10.5371, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 72.118, "mean_ep_length": 59.8, "mean_collisions": 5.24, "mean_intent_rate": 0.877, "policy_loss": 0.006903, "value_loss": 258.12847, "entropy": 0.189315, "total_loss": 129.069245, "clip_frac": 0.07251, "sps": 415.9, "elapsed_s": 3002.7}
+{"update": 581, "steps_done": 1189888, "mean_step_reward": 1.4234, "min_step_reward": -10.537, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 87.423, "mean_ep_length": 61.0, "mean_collisions": 4.06, "mean_intent_rate": 0.873, "policy_loss": 0.007572, "value_loss": 256.914037, "entropy": 0.184307, "total_loss": 128.462748, "clip_frac": 0.077393, "sps": 396.8, "elapsed_s": 3007.8}
+{"update": 582, "steps_done": 1191936, "mean_step_reward": 1.3762, "min_step_reward": -10.576, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 87.145, "mean_ep_length": 63.7, "mean_collisions": 5.16, "mean_intent_rate": 0.905, "policy_loss": 0.002266, "value_loss": 242.960439, "entropy": 0.151532, "total_loss": 121.48097, "clip_frac": 0.039673, "sps": 410.8, "elapsed_s": 3012.8}
+{"update": 583, "steps_done": 1193984, "mean_step_reward": 1.6786, "min_step_reward": -10.5645, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 98.08, "mean_ep_length": 58.4, "mean_collisions": 2.6, "mean_intent_rate": 0.885, "policy_loss": 0.000634, "value_loss": 187.791154, "entropy": 0.1582, "total_loss": 93.894629, "clip_frac": 0.044189, "sps": 412.0, "elapsed_s": 3017.8}
+{"update": 584, "steps_done": 1196032, "mean_step_reward": 1.7319, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 104.576, "mean_ep_length": 60.4, "mean_collisions": 2.74, "mean_intent_rate": 0.904, "policy_loss": 0.014457, "value_loss": 214.140358, "entropy": 0.174724, "total_loss": 107.08289, "clip_frac": 0.10791, "sps": 409.2, "elapsed_s": 3022.8}
+{"update": 585, "steps_done": 1198080, "mean_step_reward": 1.7512, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 107.308, "mean_ep_length": 61.0, "mean_collisions": 2.56, "mean_intent_rate": 0.905, "policy_loss": 0.003887, "value_loss": 219.700931, "entropy": 0.140398, "total_loss": 109.852948, "clip_frac": 0.056519, "sps": 409.1, "elapsed_s": 3027.8}
+{"update": 586, "steps_done": 1200128, "mean_step_reward": 1.5522, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 102.87, "mean_ep_length": 65.5, "mean_collisions": 3.19, "mean_intent_rate": 0.873, "policy_loss": 0.003946, "value_loss": 184.826285, "entropy": 0.169112, "total_loss": 92.415398, "clip_frac": 0.046509, "sps": 408.6, "elapsed_s": 3032.8}
+{"update": 587, "steps_done": 1202176, "mean_step_reward": 1.6495, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 106.345, "mean_ep_length": 64.9, "mean_collisions": 2.75, "mean_intent_rate": 0.866, "policy_loss": 0.003723, "value_loss": 183.87418, "entropy": 0.172635, "total_loss": 91.939087, "clip_frac": 0.053711, "sps": 421.5, "elapsed_s": 3037.7}
+{"update": 588, "steps_done": 1204224, "mean_step_reward": 1.4606, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 97.429, "mean_ep_length": 67.3, "mean_collisions": 4.27, "mean_intent_rate": 0.876, "policy_loss": 0.005451, "value_loss": 245.441637, "entropy": 0.15522, "total_loss": 122.724716, "clip_frac": 0.087646, "sps": 415.1, "elapsed_s": 3042.6}
+{"update": 589, "steps_done": 1206272, "mean_step_reward": 1.6407, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 100.715, "mean_ep_length": 61.6, "mean_collisions": 2.73, "mean_intent_rate": 0.866, "policy_loss": 0.002663, "value_loss": 213.668371, "entropy": 0.154197, "total_loss": 106.835306, "clip_frac": 0.045044, "sps": 414.4, "elapsed_s": 3047.6}
+{"update": 590, "steps_done": 1208320, "mean_step_reward": 1.4785, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 88.073, "mean_ep_length": 59.0, "mean_collisions": 3.29, "mean_intent_rate": 0.86, "policy_loss": 0.006033, "value_loss": 268.232701, "entropy": 0.149995, "total_loss": 134.120884, "clip_frac": 0.069336, "sps": 406.2, "elapsed_s": 3052.6}
+{"update": 591, "steps_done": 1210368, "mean_step_reward": 1.6152, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 92.833, "mean_ep_length": 57.8, "mean_collisions": 3.09, "mean_intent_rate": 0.893, "policy_loss": 0.004525, "value_loss": 241.598125, "entropy": 0.168089, "total_loss": 120.801907, "clip_frac": 0.097656, "sps": 409.4, "elapsed_s": 3057.6}
+{"update": 592, "steps_done": 1212416, "mean_step_reward": 1.7432, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 101.068, "mean_ep_length": 58.4, "mean_collisions": 2.26, "mean_intent_rate": 0.886, "policy_loss": 0.000725, "value_loss": 219.768995, "entropy": 0.151893, "total_loss": 109.883704, "clip_frac": 0.034912, "sps": 407.2, "elapsed_s": 3062.6}
+{"update": 593, "steps_done": 1214464, "mean_step_reward": 1.5209, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 98.215, "mean_ep_length": 64.2, "mean_collisions": 4.12, "mean_intent_rate": 0.905, "policy_loss": 0.009058, "value_loss": 281.526244, "entropy": 0.132934, "total_loss": 140.77085, "clip_frac": 0.078857, "sps": 409.0, "elapsed_s": 3067.6}
+{"update": 594, "steps_done": 1216512, "mean_step_reward": 1.3707, "min_step_reward": -10.5426, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 93.334, "mean_ep_length": 66.4, "mean_collisions": 5.03, "mean_intent_rate": 0.902, "policy_loss": 0.002156, "value_loss": 253.556663, "entropy": 0.15782, "total_loss": 126.778911, "clip_frac": 0.064819, "sps": 408.0, "elapsed_s": 3072.7}
+{"update": 595, "steps_done": 1218560, "mean_step_reward": 1.0481, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 63.811, "mean_ep_length": 58.5, "mean_collisions": 6.34, "mean_intent_rate": 0.906, "policy_loss": 0.004288, "value_loss": 308.476622, "entropy": 0.154187, "total_loss": 154.241059, "clip_frac": 0.072144, "sps": 416.9, "elapsed_s": 3077.6}
+{"update": 596, "steps_done": 1220608, "mean_step_reward": 1.6434, "min_step_reward": -10.5508, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 102.86, "mean_ep_length": 65.2, "mean_collisions": 3.66, "mean_intent_rate": 0.891, "policy_loss": 0.005268, "value_loss": 224.707635, "entropy": 0.147097, "total_loss": 112.357615, "clip_frac": 0.055908, "sps": 416.4, "elapsed_s": 3082.5}
+{"update": 597, "steps_done": 1222656, "mean_step_reward": 1.2824, "min_step_reward": -10.5513, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 77.256, "mean_ep_length": 59.0, "mean_collisions": 4.59, "mean_intent_rate": 0.882, "policy_loss": 0.000793, "value_loss": 235.881383, "entropy": 0.153368, "total_loss": 117.93995, "clip_frac": 0.057495, "sps": 413.8, "elapsed_s": 3087.4}
+{"update": 598, "steps_done": 1224704, "mean_step_reward": 1.4844, "min_step_reward": -10.5648, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 92.636, "mean_ep_length": 63.5, "mean_collisions": 4.67, "mean_intent_rate": 0.908, "policy_loss": 0.003282, "value_loss": 265.36694, "entropy": 0.145365, "total_loss": 132.685298, "clip_frac": 0.036255, "sps": 414.3, "elapsed_s": 3092.4}
+{"update": 599, "steps_done": 1226752, "mean_step_reward": 1.233, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 78.151, "mean_ep_length": 63.7, "mean_collisions": 5.5, "mean_intent_rate": 0.878, "policy_loss": 0.025148, "value_loss": 254.192119, "entropy": 0.187917, "total_loss": 127.119327, "clip_frac": 0.129761, "sps": 408.4, "elapsed_s": 3097.4}
+{"update": 600, "steps_done": 1228800, "mean_step_reward": 1.6003, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 92.202, "mean_ep_length": 58.0, "mean_collisions": 3.77, "mean_intent_rate": 0.922, "policy_loss": 0.014544, "value_loss": 256.021303, "entropy": 0.132304, "total_loss": 128.023873, "clip_frac": 0.07666, "sps": 412.3, "elapsed_s": 3102.4}
+{"update": 601, "steps_done": 1230848, "mean_step_reward": 1.2678, "min_step_reward": -10.5839, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 80.607, "mean_ep_length": 63.8, "mean_collisions": 5.69, "mean_intent_rate": 0.902, "policy_loss": 0.003672, "value_loss": 240.962089, "entropy": 0.145462, "total_loss": 120.483262, "clip_frac": 0.09314, "sps": 408.4, "elapsed_s": 3107.4}
+{"update": 602, "steps_done": 1232896, "mean_step_reward": 0.9657, "min_step_reward": -10.5556, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 64.245, "mean_ep_length": 62.2, "mean_collisions": 7.03, "mean_intent_rate": 0.901, "policy_loss": 0.006741, "value_loss": 338.961469, "entropy": 0.132801, "total_loss": 169.486146, "clip_frac": 0.059814, "sps": 413.5, "elapsed_s": 3112.4}
+{"update": 603, "steps_done": 1234944, "mean_step_reward": 1.0314, "min_step_reward": -10.575, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 57.645, "mean_ep_length": 59.9, "mean_collisions": 6.82, "mean_intent_rate": 0.877, "policy_loss": 0.00347, "value_loss": 231.72301, "entropy": 0.16006, "total_loss": 115.863374, "clip_frac": 0.058472, "sps": 416.7, "elapsed_s": 3117.3}
+{"update": 604, "steps_done": 1236992, "mean_step_reward": 1.3413, "min_step_reward": -10.5297, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 91.206, "mean_ep_length": 63.8, "mean_collisions": 5.22, "mean_intent_rate": 0.923, "policy_loss": 0.005511, "value_loss": 271.769117, "entropy": 0.121684, "total_loss": 135.888851, "clip_frac": 0.052734, "sps": 413.1, "elapsed_s": 3122.2}
+{"update": 605, "steps_done": 1239040, "mean_step_reward": 1.0202, "min_step_reward": -10.5788, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 60.46, "mean_ep_length": 64.4, "mean_collisions": 7.72, "mean_intent_rate": 0.888, "policy_loss": 0.005078, "value_loss": 219.169849, "entropy": 0.177023, "total_loss": 109.588233, "clip_frac": 0.11792, "sps": 411.0, "elapsed_s": 3127.2}
+{"update": 606, "steps_done": 1241088, "mean_step_reward": 1.3553, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 82.675, "mean_ep_length": 60.6, "mean_collisions": 4.76, "mean_intent_rate": 0.884, "policy_loss": 0.004115, "value_loss": 255.025531, "entropy": 0.164797, "total_loss": 127.51523, "clip_frac": 0.120972, "sps": 407.5, "elapsed_s": 3132.2}
+{"update": 607, "steps_done": 1243136, "mean_step_reward": 1.5326, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 92.987, "mean_ep_length": 60.8, "mean_collisions": 3.55, "mean_intent_rate": 0.88, "policy_loss": 0.004692, "value_loss": 226.274796, "entropy": 0.153845, "total_loss": 113.140551, "clip_frac": 0.061523, "sps": 405.0, "elapsed_s": 3137.3}
+{"update": 608, "steps_done": 1245184, "mean_step_reward": 1.5375, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 97.602, "mean_ep_length": 63.2, "mean_collisions": 3.88, "mean_intent_rate": 0.895, "policy_loss": 0.00128, "value_loss": 209.567038, "entropy": 0.146491, "total_loss": 104.783334, "clip_frac": 0.026367, "sps": 409.4, "elapsed_s": 3142.3}
+{"update": 609, "steps_done": 1247232, "mean_step_reward": 1.6606, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 103.598, "mean_ep_length": 62.4, "mean_collisions": 3.09, "mean_intent_rate": 0.893, "policy_loss": 0.015417, "value_loss": 218.670878, "entropy": 0.15499, "total_loss": 109.349306, "clip_frac": 0.089111, "sps": 414.0, "elapsed_s": 3147.2}
+{"update": 610, "steps_done": 1249280, "mean_step_reward": 1.4586, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 92.689, "mean_ep_length": 63.7, "mean_collisions": 4.39, "mean_intent_rate": 0.894, "policy_loss": 0.005082, "value_loss": 211.016819, "entropy": 0.173448, "total_loss": 105.511757, "clip_frac": 0.069458, "sps": 413.1, "elapsed_s": 3152.2}
+{"update": 611, "steps_done": 1251328, "mean_step_reward": 1.4545, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 87.309, "mean_ep_length": 60.2, "mean_collisions": 4.66, "mean_intent_rate": 0.923, "policy_loss": 0.001525, "value_loss": 285.477426, "entropy": 0.133646, "total_loss": 142.738901, "clip_frac": 0.054199, "sps": 408.0, "elapsed_s": 3157.2}
+{"update": 612, "steps_done": 1253376, "mean_step_reward": 1.4406, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 88.811, "mean_ep_length": 61.1, "mean_collisions": 4.38, "mean_intent_rate": 0.907, "policy_loss": 0.002157, "value_loss": 227.135156, "entropy": 0.154539, "total_loss": 113.568188, "clip_frac": 0.047607, "sps": 414.1, "elapsed_s": 3162.2}
+{"update": 613, "steps_done": 1255424, "mean_step_reward": 1.3653, "min_step_reward": -10.5275, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 81.948, "mean_ep_length": 60.1, "mean_collisions": 5.47, "mean_intent_rate": 0.932, "policy_loss": 0.009766, "value_loss": 333.322122, "entropy": 0.108344, "total_loss": 166.669744, "clip_frac": 0.042114, "sps": 410.3, "elapsed_s": 3167.2}
+{"update": 614, "steps_done": 1257472, "mean_step_reward": 1.6209, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 97.562, "mean_ep_length": 60.2, "mean_collisions": 3.41, "mean_intent_rate": 0.907, "policy_loss": 0.005047, "value_loss": 231.422822, "entropy": 0.140986, "total_loss": 115.715049, "clip_frac": 0.065918, "sps": 406.8, "elapsed_s": 3172.2}
+{"update": 615, "steps_done": 1259520, "mean_step_reward": 1.6197, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 98.78, "mean_ep_length": 61.2, "mean_collisions": 3.94, "mean_intent_rate": 0.928, "policy_loss": 0.002157, "value_loss": 273.938381, "entropy": 0.112615, "total_loss": 136.970221, "clip_frac": 0.028198, "sps": 412.8, "elapsed_s": 3177.2}
+{"update": 616, "steps_done": 1261568, "mean_step_reward": 1.5439, "min_step_reward": -10.5513, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 91.482, "mean_ep_length": 59.1, "mean_collisions": 3.86, "mean_intent_rate": 0.911, "policy_loss": 0.011518, "value_loss": 257.630345, "entropy": 0.132646, "total_loss": 128.825364, "clip_frac": 0.081543, "sps": 404.1, "elapsed_s": 3182.2}
+{"update": 617, "steps_done": 1263616, "mean_step_reward": 1.6018, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 101.122, "mean_ep_length": 63.4, "mean_collisions": 4.38, "mean_intent_rate": 0.939, "policy_loss": 0.001061, "value_loss": 313.162014, "entropy": 0.101713, "total_loss": 156.581051, "clip_frac": 0.020996, "sps": 408.9, "elapsed_s": 3187.2}
+{"update": 618, "steps_done": 1265664, "mean_step_reward": 1.1638, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 71.268, "mean_ep_length": 60.7, "mean_collisions": 6.21, "mean_intent_rate": 0.907, "policy_loss": 0.004832, "value_loss": 332.592522, "entropy": 0.121564, "total_loss": 166.299877, "clip_frac": 0.060913, "sps": 416.1, "elapsed_s": 3192.2}
+{"update": 619, "steps_done": 1267712, "mean_step_reward": 1.7956, "min_step_reward": -10.5622, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 110.361, "mean_ep_length": 61.6, "mean_collisions": 2.88, "mean_intent_rate": 0.932, "policy_loss": -0.001159, "value_loss": 236.333768, "entropy": 0.112607, "total_loss": 118.164599, "clip_frac": 0.039429, "sps": 411.3, "elapsed_s": 3197.1}
+{"update": 620, "steps_done": 1269760, "mean_step_reward": 1.5644, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 93.137, "mean_ep_length": 59.2, "mean_collisions": 4.11, "mean_intent_rate": 0.931, "policy_loss": 0.001603, "value_loss": 301.705361, "entropy": 0.106199, "total_loss": 150.853222, "clip_frac": 0.031128, "sps": 412.4, "elapsed_s": 3202.1}
+{"update": 621, "steps_done": 1271808, "mean_step_reward": 1.6793, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 106.4, "mean_ep_length": 64.2, "mean_collisions": 3.84, "mean_intent_rate": 0.931, "policy_loss": 0.005499, "value_loss": 257.900378, "entropy": 0.107054, "total_loss": 128.954618, "clip_frac": 0.056396, "sps": 404.6, "elapsed_s": 3207.2}
+{"update": 622, "steps_done": 1273856, "mean_step_reward": 1.5898, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 99.949, "mean_ep_length": 61.2, "mean_collisions": 3.53, "mean_intent_rate": 0.92, "policy_loss": 0.004394, "value_loss": 243.240108, "entropy": 0.129389, "total_loss": 121.623155, "clip_frac": 0.075195, "sps": 413.5, "elapsed_s": 3212.1}
+{"update": 623, "steps_done": 1275904, "mean_step_reward": 1.9673, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 124.606, "mean_ep_length": 64.2, "mean_collisions": 2.28, "mean_intent_rate": 0.938, "policy_loss": 0.001289, "value_loss": 263.645586, "entropy": 0.090065, "total_loss": 131.823181, "clip_frac": 0.019043, "sps": 413.3, "elapsed_s": 3217.1}
+{"update": 624, "steps_done": 1277952, "mean_step_reward": 1.4431, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 89.827, "mean_ep_length": 62.0, "mean_collisions": 4.94, "mean_intent_rate": 0.929, "policy_loss": -0.000561, "value_loss": 300.120444, "entropy": 0.120582, "total_loss": 150.058455, "clip_frac": 0.089722, "sps": 406.3, "elapsed_s": 3222.1}
+{"update": 625, "steps_done": 1280000, "mean_step_reward": 1.5842, "min_step_reward": -10.5689, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 93.775, "mean_ep_length": 59.8, "mean_collisions": 3.94, "mean_intent_rate": 0.92, "policy_loss": 0.011517, "value_loss": 277.115107, "entropy": 0.113752, "total_loss": 138.567934, "clip_frac": 0.057495, "sps": 412.5, "elapsed_s": 3227.1}
+{"update": 626, "steps_done": 1282048, "mean_step_reward": 1.7714, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 111.428, "mean_ep_length": 62.6, "mean_collisions": 2.85, "mean_intent_rate": 0.919, "policy_loss": 0.003177, "value_loss": 267.218837, "entropy": 0.109116, "total_loss": 133.611504, "clip_frac": 0.052002, "sps": 413.5, "elapsed_s": 3232.0}
+{"update": 627, "steps_done": 1284096, "mean_step_reward": 1.7412, "min_step_reward": -10.5857, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 107.244, "mean_ep_length": 61.5, "mean_collisions": 3.39, "mean_intent_rate": 0.94, "policy_loss": 0.003621, "value_loss": 279.628013, "entropy": 0.093411, "total_loss": 139.816693, "clip_frac": 0.019409, "sps": 413.8, "elapsed_s": 3237.0}
+{"update": 628, "steps_done": 1286144, "mean_step_reward": 1.9111, "min_step_reward": -10.577, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 110.027, "mean_ep_length": 57.7, "mean_collisions": 1.94, "mean_intent_rate": 0.929, "policy_loss": 0.004452, "value_loss": 255.539721, "entropy": 0.100553, "total_loss": 127.773306, "clip_frac": 0.043945, "sps": 410.5, "elapsed_s": 3242.0}
+{"update": 629, "steps_done": 1288192, "mean_step_reward": 1.8395, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 111.029, "mean_ep_length": 60.3, "mean_collisions": 2.5, "mean_intent_rate": 0.924, "policy_loss": -0.000288, "value_loss": 259.03009, "entropy": 0.099256, "total_loss": 129.513764, "clip_frac": 0.025635, "sps": 408.8, "elapsed_s": 3247.0}
+{"update": 630, "steps_done": 1290240, "mean_step_reward": 1.8956, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 132.092, "mean_ep_length": 69.1, "mean_collisions": 2.93, "mean_intent_rate": 0.951, "policy_loss": 0.000506, "value_loss": 270.474233, "entropy": 0.083688, "total_loss": 135.236785, "clip_frac": 0.03479, "sps": 417.6, "elapsed_s": 3251.9}
+{"update": 631, "steps_done": 1292288, "mean_step_reward": 1.7383, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 105.719, "mean_ep_length": 61.3, "mean_collisions": 3.71, "mean_intent_rate": 0.957, "policy_loss": 0.001234, "value_loss": 284.219958, "entropy": 0.083271, "total_loss": 142.11038, "clip_frac": 0.034302, "sps": 410.4, "elapsed_s": 3256.9}
+{"update": 632, "steps_done": 1294336, "mean_step_reward": 1.8247, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 113.189, "mean_ep_length": 61.3, "mean_collisions": 2.42, "mean_intent_rate": 0.929, "policy_loss": -0.000339, "value_loss": 254.861852, "entropy": 0.097713, "total_loss": 127.429611, "clip_frac": 0.064575, "sps": 418.9, "elapsed_s": 3261.8}
+{"update": 633, "steps_done": 1296384, "mean_step_reward": 1.683, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 102.228, "mean_ep_length": 61.3, "mean_collisions": 3.62, "mean_intent_rate": 0.929, "policy_loss": -0.001479, "value_loss": 309.329924, "entropy": 0.103949, "total_loss": 154.662444, "clip_frac": 0.044678, "sps": 402.3, "elapsed_s": 3266.9}
+{"update": 634, "steps_done": 1298432, "mean_step_reward": 2.1427, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 140.116, "mean_ep_length": 65.5, "mean_collisions": 1.26, "mean_intent_rate": 0.949, "policy_loss": 0.003231, "value_loss": 294.804818, "entropy": 0.095168, "total_loss": 147.404688, "clip_frac": 0.036743, "sps": 411.0, "elapsed_s": 3271.9}
+{"update": 635, "steps_done": 1300480, "mean_step_reward": 1.7745, "min_step_reward": -10.5513, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 98.784, "mean_ep_length": 55.8, "mean_collisions": 2.42, "mean_intent_rate": 0.917, "policy_loss": 0.00248, "value_loss": 204.655479, "entropy": 0.106295, "total_loss": 102.329156, "clip_frac": 0.025757, "sps": 414.0, "elapsed_s": 3276.8}
+{"update": 636, "steps_done": 1302528, "mean_step_reward": 1.9784, "min_step_reward": -10.5772, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 125.06, "mean_ep_length": 63.3, "mean_collisions": 2.12, "mean_intent_rate": 0.946, "policy_loss": 0.015767, "value_loss": 237.946753, "entropy": 0.083694, "total_loss": 118.988305, "clip_frac": 0.054443, "sps": 413.5, "elapsed_s": 3281.8}
+{"update": 637, "steps_done": 1304576, "mean_step_reward": 1.4591, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 91.838, "mean_ep_length": 62.6, "mean_collisions": 4.64, "mean_intent_rate": 0.92, "policy_loss": 0.022734, "value_loss": 290.644869, "entropy": 0.136598, "total_loss": 145.343802, "clip_frac": 0.112427, "sps": 415.4, "elapsed_s": 3286.7}
+{"update": 638, "steps_done": 1306624, "mean_step_reward": 1.6863, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 103.882, "mean_ep_length": 61.5, "mean_collisions": 2.45, "mean_intent_rate": 0.873, "policy_loss": 0.017416, "value_loss": 250.870413, "entropy": 0.157684, "total_loss": 125.451044, "clip_frac": 0.099365, "sps": 410.4, "elapsed_s": 3291.7}
+{"update": 639, "steps_done": 1308672, "mean_step_reward": 1.8306, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 124.234, "mean_ep_length": 66.5, "mean_collisions": 2.93, "mean_intent_rate": 0.949, "policy_loss": 0.001239, "value_loss": 304.049182, "entropy": 0.088225, "total_loss": 152.024948, "clip_frac": 0.022827, "sps": 410.4, "elapsed_s": 3296.7}
+{"update": 640, "steps_done": 1310720, "mean_step_reward": 1.7066, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 106.373, "mean_ep_length": 60.5, "mean_collisions": 2.85, "mean_intent_rate": 0.924, "policy_loss": 0.001369, "value_loss": 334.588417, "entropy": 0.110317, "total_loss": 167.294473, "clip_frac": 0.038086, "sps": 408.3, "elapsed_s": 3301.7}
+{"update": 641, "steps_done": 1312768, "mean_step_reward": 1.7191, "min_step_reward": -10.5785, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 96.086, "mean_ep_length": 58.8, "mean_collisions": 3.67, "mean_intent_rate": 0.932, "policy_loss": 0.005969, "value_loss": 232.49801, "entropy": 0.100032, "total_loss": 116.253973, "clip_frac": 0.029419, "sps": 409.8, "elapsed_s": 3306.7}
+{"update": 642, "steps_done": 1314816, "mean_step_reward": 1.8059, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 111.122, "mean_ep_length": 60.2, "mean_collisions": 2.33, "mean_intent_rate": 0.922, "policy_loss": 0.000512, "value_loss": 256.005888, "entropy": 0.111424, "total_loss": 128.002341, "clip_frac": 0.020874, "sps": 410.3, "elapsed_s": 3311.7}
+{"update": 643, "steps_done": 1316864, "mean_step_reward": 1.5461, "min_step_reward": -10.5672, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 103.845, "mean_ep_length": 69.0, "mean_collisions": 4.83, "mean_intent_rate": 0.914, "policy_loss": 0.001528, "value_loss": 254.379107, "entropy": 0.12512, "total_loss": 127.18983, "clip_frac": 0.049072, "sps": 425.8, "elapsed_s": 3316.5}
+{"update": 644, "steps_done": 1318912, "mean_step_reward": 1.5397, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 87.853, "mean_ep_length": 56.7, "mean_collisions": 3.5, "mean_intent_rate": 0.905, "policy_loss": 0.007469, "value_loss": 241.786437, "entropy": 0.130173, "total_loss": 120.899386, "clip_frac": 0.061523, "sps": 417.2, "elapsed_s": 3321.4}
+{"update": 645, "steps_done": 1320960, "mean_step_reward": 1.8066, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 117.751, "mean_ep_length": 65.4, "mean_collisions": 2.47, "mean_intent_rate": 0.912, "policy_loss": 0.006182, "value_loss": 197.198828, "entropy": 0.139538, "total_loss": 98.604201, "clip_frac": 0.056519, "sps": 410.4, "elapsed_s": 3326.4}
+{"update": 646, "steps_done": 1323008, "mean_step_reward": 1.6914, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 113.07, "mean_ep_length": 67.3, "mean_collisions": 2.83, "mean_intent_rate": 0.883, "policy_loss": 0.000805, "value_loss": 175.375554, "entropy": 0.147173, "total_loss": 87.68711, "clip_frac": 0.035278, "sps": 415.2, "elapsed_s": 3331.3}
+{"update": 647, "steps_done": 1325056, "mean_step_reward": 1.8221, "min_step_reward": -10.5442, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 114.132, "mean_ep_length": 62.5, "mean_collisions": 2.21, "mean_intent_rate": 0.905, "policy_loss": 0.002496, "value_loss": 224.048639, "entropy": 0.137601, "total_loss": 112.02544, "clip_frac": 0.043457, "sps": 412.3, "elapsed_s": 3336.3}
+{"update": 648, "steps_done": 1327104, "mean_step_reward": 1.529, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 101.341, "mean_ep_length": 66.2, "mean_collisions": 3.74, "mean_intent_rate": 0.88, "policy_loss": 0.002322, "value_loss": 286.371919, "entropy": 0.122446, "total_loss": 143.187058, "clip_frac": 0.055298, "sps": 415.8, "elapsed_s": 3341.2}
+{"update": 649, "steps_done": 1329152, "mean_step_reward": 1.7241, "min_step_reward": -10.547, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 119.021, "mean_ep_length": 68.8, "mean_collisions": 2.63, "mean_intent_rate": 0.887, "policy_loss": 0.012114, "value_loss": 174.569024, "entropy": 0.149446, "total_loss": 87.295131, "clip_frac": 0.077515, "sps": 413.2, "elapsed_s": 3346.2}
+{"update": 650, "steps_done": 1331200, "mean_step_reward": 1.7416, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 107.163, "mean_ep_length": 60.1, "mean_collisions": 2.48, "mean_intent_rate": 0.912, "policy_loss": -0.001236, "value_loss": 206.16779, "entropy": 0.123946, "total_loss": 103.081419, "clip_frac": 0.051636, "sps": 404.9, "elapsed_s": 3351.2}
+{"update": 651, "steps_done": 1333248, "mean_step_reward": 1.9059, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 119.389, "mean_ep_length": 64.4, "mean_collisions": 2.75, "mean_intent_rate": 0.936, "policy_loss": 0.00191, "value_loss": 282.480207, "entropy": 0.096193, "total_loss": 141.241052, "clip_frac": 0.020508, "sps": 412.3, "elapsed_s": 3356.2}
+{"update": 652, "steps_done": 1335296, "mean_step_reward": 2.1464, "min_step_reward": -10.5612, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 120.527, "mean_ep_length": 56.2, "mean_collisions": 1.08, "mean_intent_rate": 0.95, "policy_loss": 0.000624, "value_loss": 262.659098, "entropy": 0.07229, "total_loss": 131.32945, "clip_frac": 0.018799, "sps": 408.0, "elapsed_s": 3361.2}
+{"update": 653, "steps_done": 1337344, "mean_step_reward": 1.8217, "min_step_reward": -10.5586, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 118.414, "mean_ep_length": 64.1, "mean_collisions": 2.53, "mean_intent_rate": 0.927, "policy_loss": 0.000324, "value_loss": 290.580081, "entropy": 0.09836, "total_loss": 145.289383, "clip_frac": 0.022461, "sps": 412.0, "elapsed_s": 3366.2}
+{"update": 654, "steps_done": 1339392, "mean_step_reward": 1.9778, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 120.96, "mean_ep_length": 62.0, "mean_collisions": 2.3, "mean_intent_rate": 0.952, "policy_loss": 0.000281, "value_loss": 292.291618, "entropy": 0.079664, "total_loss": 146.145293, "clip_frac": 0.021851, "sps": 408.0, "elapsed_s": 3371.2}
+{"update": 655, "steps_done": 1341440, "mean_step_reward": 1.9066, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 124.291, "mean_ep_length": 65.4, "mean_collisions": 2.58, "mean_intent_rate": 0.944, "policy_loss": 0.003719, "value_loss": 272.837724, "entropy": 0.089036, "total_loss": 136.421691, "clip_frac": 0.026978, "sps": 412.7, "elapsed_s": 3376.2}
+{"update": 656, "steps_done": 1343488, "mean_step_reward": 1.9499, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 125.218, "mean_ep_length": 61.2, "mean_collisions": 1.73, "mean_intent_rate": 0.954, "policy_loss": 0.000427, "value_loss": 271.597173, "entropy": 0.082154, "total_loss": 135.798192, "clip_frac": 0.012573, "sps": 412.9, "elapsed_s": 3381.1}
+{"update": 657, "steps_done": 1345536, "mean_step_reward": 1.7449, "min_step_reward": -10.5645, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 105.883, "mean_ep_length": 63.6, "mean_collisions": 3.73, "mean_intent_rate": 0.93, "policy_loss": 0.003235, "value_loss": 256.720861, "entropy": 0.089601, "total_loss": 128.362768, "clip_frac": 0.042725, "sps": 412.7, "elapsed_s": 3386.1}
+{"update": 658, "steps_done": 1347584, "mean_step_reward": 1.905, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 121.074, "mean_ep_length": 63.7, "mean_collisions": 2.62, "mean_intent_rate": 0.949, "policy_loss": 0.004615, "value_loss": 283.590188, "entropy": 0.085117, "total_loss": 141.79886, "clip_frac": 0.030151, "sps": 407.7, "elapsed_s": 3391.1}
+{"update": 659, "steps_done": 1349632, "mean_step_reward": 1.8882, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 117.925, "mean_ep_length": 62.4, "mean_collisions": 2.55, "mean_intent_rate": 0.941, "policy_loss": -0.000784, "value_loss": 280.403491, "entropy": 0.084716, "total_loss": 140.200114, "clip_frac": 0.01416, "sps": 415.3, "elapsed_s": 3396.1}
+{"update": 660, "steps_done": 1351680, "mean_step_reward": 2.1157, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 127.609, "mean_ep_length": 60.3, "mean_collisions": 1.56, "mean_intent_rate": 0.965, "policy_loss": 0.000551, "value_loss": 283.056077, "entropy": 0.064377, "total_loss": 141.527945, "clip_frac": 0.015381, "sps": 416.8, "elapsed_s": 3401.0}
+{"update": 661, "steps_done": 1353728, "mean_step_reward": 2.0727, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 130.461, "mean_ep_length": 63.2, "mean_collisions": 1.75, "mean_intent_rate": 0.959, "policy_loss": -6e-06, "value_loss": 288.365812, "entropy": 0.075407, "total_loss": 144.182147, "clip_frac": 0.016968, "sps": 409.8, "elapsed_s": 3406.0}
+{"update": 662, "steps_done": 1355776, "mean_step_reward": 2.0315, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 127.957, "mean_ep_length": 62.8, "mean_collisions": 1.7, "mean_intent_rate": 0.938, "policy_loss": 0.000883, "value_loss": 267.043846, "entropy": 0.081176, "total_loss": 133.521993, "clip_frac": 0.014648, "sps": 409.8, "elapsed_s": 3411.0}
+{"update": 663, "steps_done": 1357824, "mean_step_reward": 1.9988, "min_step_reward": -10.5534, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 132.08, "mean_ep_length": 66.1, "mean_collisions": 1.94, "mean_intent_rate": 0.942, "policy_loss": 0.000533, "value_loss": 274.76464, "entropy": 0.082163, "total_loss": 137.382031, "clip_frac": 0.018433, "sps": 411.7, "elapsed_s": 3415.9}
+{"update": 664, "steps_done": 1359872, "mean_step_reward": 1.8285, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 117.964, "mean_ep_length": 64.1, "mean_collisions": 2.68, "mean_intent_rate": 0.934, "policy_loss": 0.004236, "value_loss": 231.09114, "entropy": 0.101657, "total_loss": 115.548789, "clip_frac": 0.035278, "sps": 413.3, "elapsed_s": 3420.9}
+{"update": 665, "steps_done": 1361920, "mean_step_reward": 1.9464, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 113.28, "mean_ep_length": 58.6, "mean_collisions": 2.11, "mean_intent_rate": 0.946, "policy_loss": 0.003016, "value_loss": 199.359871, "entropy": 0.084533, "total_loss": 99.682107, "clip_frac": 0.017944, "sps": 414.2, "elapsed_s": 3425.9}
+{"update": 666, "steps_done": 1363968, "mean_step_reward": 1.8486, "min_step_reward": -10.5593, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 108.734, "mean_ep_length": 58.7, "mean_collisions": 2.31, "mean_intent_rate": 0.923, "policy_loss": 0.010427, "value_loss": 243.709703, "entropy": 0.110778, "total_loss": 121.864169, "clip_frac": 0.050293, "sps": 411.7, "elapsed_s": 3430.8}
+{"update": 667, "steps_done": 1366016, "mean_step_reward": 1.7843, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 107.561, "mean_ep_length": 60.3, "mean_collisions": 3.24, "mean_intent_rate": 0.949, "policy_loss": -0.001212, "value_loss": 323.471645, "entropy": 0.078581, "total_loss": 161.733823, "clip_frac": 0.019287, "sps": 400.7, "elapsed_s": 3435.9}
+{"update": 668, "steps_done": 1368064, "mean_step_reward": 1.8216, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 120.895, "mean_ep_length": 66.3, "mean_collisions": 3.06, "mean_intent_rate": 0.937, "policy_loss": 0.001907, "value_loss": 261.358311, "entropy": 0.093622, "total_loss": 130.680125, "clip_frac": 0.035034, "sps": 414.6, "elapsed_s": 3440.9}
+{"update": 669, "steps_done": 1370112, "mean_step_reward": 2.097, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 140.749, "mean_ep_length": 67.3, "mean_collisions": 1.37, "mean_intent_rate": 0.942, "policy_loss": 0.00512, "value_loss": 270.852081, "entropy": 0.073159, "total_loss": 135.430429, "clip_frac": 0.036865, "sps": 415.1, "elapsed_s": 3445.8}
+{"update": 670, "steps_done": 1372160, "mean_step_reward": 2.0544, "min_step_reward": -10.5647, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 121.213, "mean_ep_length": 58.9, "mean_collisions": 1.51, "mean_intent_rate": 0.948, "policy_loss": 0.008274, "value_loss": 229.143181, "entropy": 0.080563, "total_loss": 114.579059, "clip_frac": 0.029297, "sps": 420.8, "elapsed_s": 3450.7}
+{"update": 671, "steps_done": 1374208, "mean_step_reward": 1.9004, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 108.785, "mean_ep_length": 57.5, "mean_collisions": 2.6, "mean_intent_rate": 0.959, "policy_loss": 0.000252, "value_loss": 327.828885, "entropy": 0.074387, "total_loss": 163.913951, "clip_frac": 0.017578, "sps": 414.6, "elapsed_s": 3455.6}
+{"update": 672, "steps_done": 1376256, "mean_step_reward": 1.8229, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 115.657, "mean_ep_length": 61.2, "mean_collisions": 2.67, "mean_intent_rate": 0.951, "policy_loss": 0.00252, "value_loss": 330.515078, "entropy": 0.070558, "total_loss": 165.259356, "clip_frac": 0.029785, "sps": 412.5, "elapsed_s": 3460.6}
+{"update": 673, "steps_done": 1378304, "mean_step_reward": 1.8402, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 116.705, "mean_ep_length": 65.5, "mean_collisions": 3.5, "mean_intent_rate": 0.951, "policy_loss": 0.010153, "value_loss": 319.041776, "entropy": 0.106909, "total_loss": 159.529972, "clip_frac": 0.083984, "sps": 413.4, "elapsed_s": 3465.5}
+{"update": 674, "steps_done": 1380352, "mean_step_reward": 2.1936, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 133.882, "mean_ep_length": 61.2, "mean_collisions": 1.09, "mean_intent_rate": 0.963, "policy_loss": 0.000598, "value_loss": 231.03063, "entropy": 0.05483, "total_loss": 115.515365, "clip_frac": 0.018188, "sps": 400.9, "elapsed_s": 3470.6}
+{"update": 675, "steps_done": 1382400, "mean_step_reward": 1.94, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 124.697, "mean_ep_length": 63.8, "mean_collisions": 2.15, "mean_intent_rate": 0.941, "policy_loss": 0.005737, "value_loss": 282.374918, "entropy": 0.092741, "total_loss": 141.192267, "clip_frac": 0.028809, "sps": 403.5, "elapsed_s": 3475.7}
+{"update": 676, "steps_done": 1384448, "mean_step_reward": 1.9573, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 129.224, "mean_ep_length": 64.3, "mean_collisions": 1.9, "mean_intent_rate": 0.943, "policy_loss": 0.008335, "value_loss": 296.248088, "entropy": 0.08464, "total_loss": 148.131535, "clip_frac": 0.032959, "sps": 408.7, "elapsed_s": 3480.7}
+{"update": 677, "steps_done": 1386496, "mean_step_reward": 1.9896, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 117.33, "mean_ep_length": 60.8, "mean_collisions": 2.53, "mean_intent_rate": 0.959, "policy_loss": 0.001925, "value_loss": 270.58547, "entropy": 0.061966, "total_loss": 135.294039, "clip_frac": 0.013428, "sps": 410.7, "elapsed_s": 3485.7}
+{"update": 678, "steps_done": 1388544, "mean_step_reward": 1.8896, "min_step_reward": -10.5362, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 116.036, "mean_ep_length": 61.1, "mean_collisions": 2.71, "mean_intent_rate": 0.955, "policy_loss": 0.004994, "value_loss": 296.171957, "entropy": 0.075073, "total_loss": 148.090221, "clip_frac": 0.029175, "sps": 407.7, "elapsed_s": 3490.7}
+{"update": 679, "steps_done": 1390592, "mean_step_reward": 2.059, "min_step_reward": -10.5889, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 119.553, "mean_ep_length": 57.7, "mean_collisions": 1.4, "mean_intent_rate": 0.946, "policy_loss": 0.003291, "value_loss": 237.231817, "entropy": 0.077079, "total_loss": 118.618428, "clip_frac": 0.026733, "sps": 408.1, "elapsed_s": 3495.8}
+{"update": 680, "steps_done": 1392640, "mean_step_reward": 1.4922, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 96.672, "mean_ep_length": 65.0, "mean_collisions": 4.78, "mean_intent_rate": 0.918, "policy_loss": -0.001682, "value_loss": 309.935868, "entropy": 0.100723, "total_loss": 154.965245, "clip_frac": 0.050781, "sps": 407.9, "elapsed_s": 3500.8}
+{"update": 681, "steps_done": 1394688, "mean_step_reward": 2.0353, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 125.562, "mean_ep_length": 61.8, "mean_collisions": 1.97, "mean_intent_rate": 0.959, "policy_loss": 0.000288, "value_loss": 228.05567, "entropy": 0.061172, "total_loss": 114.027511, "clip_frac": 0.012207, "sps": 413.7, "elapsed_s": 3505.7}
+{"update": 682, "steps_done": 1396736, "mean_step_reward": 1.8901, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 118.476, "mean_ep_length": 63.0, "mean_collisions": 2.81, "mean_intent_rate": 0.951, "policy_loss": 0.002274, "value_loss": 251.559235, "entropy": 0.073652, "total_loss": 125.781155, "clip_frac": 0.020752, "sps": 414.3, "elapsed_s": 3510.7}
+{"update": 683, "steps_done": 1398784, "mean_step_reward": 1.8234, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 114.17, "mean_ep_length": 62.4, "mean_collisions": 3.24, "mean_intent_rate": 0.96, "policy_loss": -0.000597, "value_loss": 275.256248, "entropy": 0.066236, "total_loss": 137.626864, "clip_frac": 0.024658, "sps": 409.6, "elapsed_s": 3515.7}
+{"update": 684, "steps_done": 1400832, "mean_step_reward": 1.8893, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 119.863, "mean_ep_length": 63.2, "mean_collisions": 2.88, "mean_intent_rate": 0.96, "policy_loss": 0.001116, "value_loss": 282.718253, "entropy": 0.076122, "total_loss": 141.359479, "clip_frac": 0.03186, "sps": 405.1, "elapsed_s": 3520.7}
+{"update": 685, "steps_done": 1402880, "mean_step_reward": 1.9537, "min_step_reward": -10.5249, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 123.616, "mean_ep_length": 63.9, "mean_collisions": 2.42, "mean_intent_rate": 0.947, "policy_loss": 0.005098, "value_loss": 258.835883, "entropy": 0.072236, "total_loss": 129.422316, "clip_frac": 0.019775, "sps": 415.2, "elapsed_s": 3525.7}
+{"update": 686, "steps_done": 1404928, "mean_step_reward": 2.3389, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 147.377, "mean_ep_length": 62.9, "mean_collisions": 0.33, "mean_intent_rate": 0.969, "policy_loss": 0.000773, "value_loss": 233.035004, "entropy": 0.049896, "total_loss": 116.517775, "clip_frac": 0.008667, "sps": 410.5, "elapsed_s": 3530.7}
+{"update": 687, "steps_done": 1406976, "mean_step_reward": 1.9343, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 128.512, "mean_ep_length": 66.9, "mean_collisions": 2.63, "mean_intent_rate": 0.949, "policy_loss": -0.000603, "value_loss": 320.313963, "entropy": 0.065811, "total_loss": 160.155721, "clip_frac": 0.026001, "sps": 408.1, "elapsed_s": 3535.7}
+{"update": 688, "steps_done": 1409024, "mean_step_reward": 2.0589, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 141.647, "mean_ep_length": 68.3, "mean_collisions": 1.84, "mean_intent_rate": 0.958, "policy_loss": 8.4e-05, "value_loss": 259.474291, "entropy": 0.063792, "total_loss": 129.736592, "clip_frac": 0.021484, "sps": 409.4, "elapsed_s": 3540.7}
+{"update": 689, "steps_done": 1411072, "mean_step_reward": 1.8598, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 116.163, "mean_ep_length": 62.8, "mean_collisions": 3.03, "mean_intent_rate": 0.952, "policy_loss": -0.000537, "value_loss": 321.859682, "entropy": 0.069953, "total_loss": 160.928606, "clip_frac": 0.02478, "sps": 411.0, "elapsed_s": 3545.7}
+{"update": 690, "steps_done": 1413120, "mean_step_reward": 1.9783, "min_step_reward": -10.5133, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 118.804, "mean_ep_length": 59.7, "mean_collisions": 1.77, "mean_intent_rate": 0.94, "policy_loss": -2.6e-05, "value_loss": 242.602973, "entropy": 0.069087, "total_loss": 121.30077, "clip_frac": 0.013672, "sps": 409.3, "elapsed_s": 3550.7}
+{"update": 691, "steps_done": 1415168, "mean_step_reward": 1.9812, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 131.287, "mean_ep_length": 66.2, "mean_collisions": 2.13, "mean_intent_rate": 0.942, "policy_loss": 0.005232, "value_loss": 256.670469, "entropy": 0.08637, "total_loss": 128.339603, "clip_frac": 0.023926, "sps": 408.6, "elapsed_s": 3555.7}
+{"update": 692, "steps_done": 1417216, "mean_step_reward": 2.0472, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 122.801, "mean_ep_length": 60.0, "mean_collisions": 1.88, "mean_intent_rate": 0.962, "policy_loss": 0.001872, "value_loss": 258.384396, "entropy": 0.057037, "total_loss": 129.193501, "clip_frac": 0.018799, "sps": 404.8, "elapsed_s": 3560.8}
+{"update": 693, "steps_done": 1419264, "mean_step_reward": 2.1011, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 130.097, "mean_ep_length": 61.9, "mean_collisions": 1.7, "mean_intent_rate": 0.967, "policy_loss": 0.000539, "value_loss": 285.776101, "entropy": 0.059965, "total_loss": 142.88799, "clip_frac": 0.013306, "sps": 407.2, "elapsed_s": 3565.8}
+{"update": 694, "steps_done": 1421312, "mean_step_reward": 1.9439, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 125.004, "mean_ep_length": 64.7, "mean_collisions": 2.55, "mean_intent_rate": 0.951, "policy_loss": 0.002123, "value_loss": 298.134103, "entropy": 0.068096, "total_loss": 149.068493, "clip_frac": 0.024536, "sps": 410.4, "elapsed_s": 3570.8}
+{"update": 695, "steps_done": 1423360, "mean_step_reward": 2.0998, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 126.883, "mean_ep_length": 60.4, "mean_collisions": 1.44, "mean_intent_rate": 0.958, "policy_loss": 0.001018, "value_loss": 235.432048, "entropy": 0.067366, "total_loss": 117.716368, "clip_frac": 0.012817, "sps": 413.4, "elapsed_s": 3575.7}
+{"update": 696, "steps_done": 1425408, "mean_step_reward": 2.1421, "min_step_reward": -7.5878, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 124.293, "mean_ep_length": 57.9, "mean_collisions": 1.39, "mean_intent_rate": 0.97, "policy_loss": 0.001534, "value_loss": 262.697478, "entropy": 0.04891, "total_loss": 131.349784, "clip_frac": 0.007324, "sps": 411.5, "elapsed_s": 3580.7}
+{"update": 697, "steps_done": 1427456, "mean_step_reward": 1.909, "min_step_reward": -10.5863, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 123.111, "mean_ep_length": 64.4, "mean_collisions": 2.88, "mean_intent_rate": 0.964, "policy_loss": -0.000469, "value_loss": 341.690223, "entropy": 0.056101, "total_loss": 170.844082, "clip_frac": 0.02124, "sps": 412.4, "elapsed_s": 3585.7}
+{"update": 698, "steps_done": 1429504, "mean_step_reward": 1.5508, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 95.194, "mean_ep_length": 61.6, "mean_collisions": 4.85, "mean_intent_rate": 0.957, "policy_loss": -0.001755, "value_loss": 362.824189, "entropy": 0.063223, "total_loss": 181.409708, "clip_frac": 0.017334, "sps": 415.2, "elapsed_s": 3590.6}
+{"update": 699, "steps_done": 1431552, "mean_step_reward": 1.7479, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 97.915, "mean_ep_length": 56.3, "mean_collisions": 3.56, "mean_intent_rate": 0.966, "policy_loss": -0.000335, "value_loss": 400.006926, "entropy": 0.057385, "total_loss": 200.002553, "clip_frac": 0.016113, "sps": 409.7, "elapsed_s": 3595.6}
+{"update": 700, "steps_done": 1433600, "mean_step_reward": 1.9172, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 125.246, "mean_ep_length": 63.6, "mean_collisions": 2.59, "mean_intent_rate": 0.964, "policy_loss": 0.004173, "value_loss": 301.046595, "entropy": 0.054491, "total_loss": 150.526926, "clip_frac": 0.014648, "sps": 414.4, "elapsed_s": 3600.5}
+{"update": 701, "steps_done": 1435648, "mean_step_reward": 1.9469, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 119.094, "mean_ep_length": 62.8, "mean_collisions": 2.61, "mean_intent_rate": 0.942, "policy_loss": 0.003595, "value_loss": 332.700442, "entropy": 0.070421, "total_loss": 166.353113, "clip_frac": 0.025635, "sps": 401.6, "elapsed_s": 3605.7}
+{"update": 702, "steps_done": 1437696, "mean_step_reward": 2.052, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 125.1, "mean_ep_length": 61.2, "mean_collisions": 1.76, "mean_intent_rate": 0.957, "policy_loss": 0.000939, "value_loss": 258.443203, "entropy": 0.058002, "total_loss": 129.22196, "clip_frac": 0.013184, "sps": 410.7, "elapsed_s": 3610.6}
+{"update": 703, "steps_done": 1439744, "mean_step_reward": 1.964, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 119.789, "mean_ep_length": 60.8, "mean_collisions": 2.35, "mean_intent_rate": 0.96, "policy_loss": -0.000825, "value_loss": 259.183391, "entropy": 0.05755, "total_loss": 129.590293, "clip_frac": 0.020508, "sps": 405.5, "elapsed_s": 3615.7}
+{"update": 704, "steps_done": 1441792, "mean_step_reward": 2.0897, "min_step_reward": -10.5848, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 132.339, "mean_ep_length": 63.4, "mean_collisions": 1.78, "mean_intent_rate": 0.964, "policy_loss": 0.002714, "value_loss": 241.519853, "entropy": 0.053382, "total_loss": 120.762108, "clip_frac": 0.009033, "sps": 409.9, "elapsed_s": 3620.7}
+{"update": 705, "steps_done": 1443840, "mean_step_reward": 2.1609, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 140.576, "mean_ep_length": 64.7, "mean_collisions": 1.12, "mean_intent_rate": 0.953, "policy_loss": 0.00167, "value_loss": 254.441758, "entropy": 0.063748, "total_loss": 127.221912, "clip_frac": 0.012451, "sps": 411.1, "elapsed_s": 3625.7}
+{"update": 706, "steps_done": 1445888, "mean_step_reward": 1.9032, "min_step_reward": -10.572, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 114.266, "mean_ep_length": 60.3, "mean_collisions": 2.47, "mean_intent_rate": 0.946, "policy_loss": 7.5e-05, "value_loss": 265.452081, "entropy": 0.072773, "total_loss": 132.725388, "clip_frac": 0.011108, "sps": 413.3, "elapsed_s": 3630.6}
+{"update": 707, "steps_done": 1447936, "mean_step_reward": 2.1273, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 129.043, "mean_ep_length": 59.3, "mean_collisions": 1.15, "mean_intent_rate": 0.962, "policy_loss": -0.000505, "value_loss": 259.139332, "entropy": 0.054686, "total_loss": 129.568613, "clip_frac": 0.015625, "sps": 405.7, "elapsed_s": 3635.7}
+{"update": 708, "steps_done": 1449984, "mean_step_reward": 2.2285, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 134.163, "mean_ep_length": 61.5, "mean_collisions": 1.26, "mean_intent_rate": 0.973, "policy_loss": 0.001309, "value_loss": 238.29605, "entropy": 0.043145, "total_loss": 119.148901, "clip_frac": 0.008789, "sps": 413.6, "elapsed_s": 3640.6}
+{"update": 709, "steps_done": 1452032, "mean_step_reward": 2.1485, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 126.15, "mean_ep_length": 58.7, "mean_collisions": 1.17, "mean_intent_rate": 0.958, "policy_loss": 0.000245, "value_loss": 206.567913, "entropy": 0.055452, "total_loss": 103.283648, "clip_frac": 0.009888, "sps": 406.3, "elapsed_s": 3645.7}
+{"update": 710, "steps_done": 1454080, "mean_step_reward": 2.256, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 141.707, "mean_ep_length": 62.8, "mean_collisions": 0.91, "mean_intent_rate": 0.972, "policy_loss": 0.005274, "value_loss": 283.881221, "entropy": 0.0399, "total_loss": 141.945487, "clip_frac": 0.008911, "sps": 410.6, "elapsed_s": 3650.7}
+{"update": 711, "steps_done": 1456128, "mean_step_reward": 1.9713, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 118.868, "mean_ep_length": 60.4, "mean_collisions": 2.44, "mean_intent_rate": 0.967, "policy_loss": -0.001016, "value_loss": 334.263029, "entropy": 0.052242, "total_loss": 167.129976, "clip_frac": 0.017334, "sps": 412.5, "elapsed_s": 3655.6}
+{"update": 712, "steps_done": 1458176, "mean_step_reward": 2.2945, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 145.763, "mean_ep_length": 63.4, "mean_collisions": 0.58, "mean_intent_rate": 0.97, "policy_loss": -0.000269, "value_loss": 195.104136, "entropy": 0.045491, "total_loss": 97.551345, "clip_frac": 0.008057, "sps": 411.3, "elapsed_s": 3660.6}
+{"update": 713, "steps_done": 1460224, "mean_step_reward": 2.058, "min_step_reward": -10.582, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 119.663, "mean_ep_length": 58.5, "mean_collisions": 1.74, "mean_intent_rate": 0.96, "policy_loss": 0.002312, "value_loss": 269.991008, "entropy": 0.053854, "total_loss": 134.997278, "clip_frac": 0.017822, "sps": 411.4, "elapsed_s": 3665.6}
+{"update": 714, "steps_done": 1462272, "mean_step_reward": 2.1787, "min_step_reward": -10.5854, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 143.623, "mean_ep_length": 65.7, "mean_collisions": 1.28, "mean_intent_rate": 0.969, "policy_loss": -0.000955, "value_loss": 256.555096, "entropy": 0.052816, "total_loss": 128.276065, "clip_frac": 0.009277, "sps": 405.9, "elapsed_s": 3670.6}
+{"update": 715, "steps_done": 1464320, "mean_step_reward": 2.0101, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 127.486, "mean_ep_length": 63.5, "mean_collisions": 2.03, "mean_intent_rate": 0.952, "policy_loss": -0.000763, "value_loss": 280.581802, "entropy": 0.059961, "total_loss": 140.289539, "clip_frac": 0.015137, "sps": 414.4, "elapsed_s": 3675.6}
+{"update": 716, "steps_done": 1466368, "mean_step_reward": 2.1748, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 136.971, "mean_ep_length": 63.1, "mean_collisions": 1.28, "mean_intent_rate": 0.965, "policy_loss": 0.001413, "value_loss": 232.557301, "entropy": 0.05502, "total_loss": 116.279513, "clip_frac": 0.015259, "sps": 408.9, "elapsed_s": 3680.6}
+{"update": 717, "steps_done": 1468416, "mean_step_reward": 1.8894, "min_step_reward": -10.5514, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 120.054, "mean_ep_length": 63.0, "mean_collisions": 2.82, "mean_intent_rate": 0.957, "policy_loss": 0.003793, "value_loss": 286.154593, "entropy": 0.064628, "total_loss": 143.080443, "clip_frac": 0.035767, "sps": 413.0, "elapsed_s": 3685.5}
+{"update": 718, "steps_done": 1470464, "mean_step_reward": 1.9275, "min_step_reward": -10.5768, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 131.195, "mean_ep_length": 68.4, "mean_collisions": 2.67, "mean_intent_rate": 0.947, "policy_loss": 0.003556, "value_loss": 276.154562, "entropy": 0.062978, "total_loss": 138.080208, "clip_frac": 0.023438, "sps": 409.2, "elapsed_s": 3690.6}
+{"update": 719, "steps_done": 1472512, "mean_step_reward": 2.1148, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 140.209, "mean_ep_length": 66.6, "mean_collisions": 1.7, "mean_intent_rate": 0.962, "policy_loss": 0.001272, "value_loss": 265.93836, "entropy": 0.054499, "total_loss": 132.969907, "clip_frac": 0.010986, "sps": 408.5, "elapsed_s": 3695.6}
+{"update": 720, "steps_done": 1474560, "mean_step_reward": 2.0534, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 123.145, "mean_ep_length": 59.7, "mean_collisions": 1.8, "mean_intent_rate": 0.961, "policy_loss": 0.001231, "value_loss": 266.241789, "entropy": 0.048674, "total_loss": 133.121639, "clip_frac": 0.015381, "sps": 416.0, "elapsed_s": 3700.5}
+{"update": 721, "steps_done": 1476608, "mean_step_reward": 1.9162, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 124.706, "mean_ep_length": 65.3, "mean_collisions": 2.87, "mean_intent_rate": 0.963, "policy_loss": 0.003677, "value_loss": 271.908037, "entropy": 0.054834, "total_loss": 135.957147, "clip_frac": 0.015747, "sps": 406.0, "elapsed_s": 3705.5}
+{"update": 722, "steps_done": 1478656, "mean_step_reward": 2.224, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 133.334, "mean_ep_length": 60.0, "mean_collisions": 1.0, "mean_intent_rate": 0.972, "policy_loss": 0.001437, "value_loss": 232.895971, "entropy": 0.040349, "total_loss": 116.449019, "clip_frac": 0.006104, "sps": 408.1, "elapsed_s": 3710.6}
+{"update": 723, "steps_done": 1480704, "mean_step_reward": 2.2013, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 145.769, "mean_ep_length": 66.2, "mean_collisions": 1.16, "mean_intent_rate": 0.964, "policy_loss": -0.000209, "value_loss": 309.228284, "entropy": 0.045424, "total_loss": 154.613481, "clip_frac": 0.010132, "sps": 413.8, "elapsed_s": 3715.5}
+{"update": 724, "steps_done": 1482752, "mean_step_reward": 1.8996, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 114.141, "mean_ep_length": 59.7, "mean_collisions": 2.77, "mean_intent_rate": 0.964, "policy_loss": 0.001098, "value_loss": 359.761359, "entropy": 0.057223, "total_loss": 179.881207, "clip_frac": 0.024292, "sps": 405.6, "elapsed_s": 3720.6}
+{"update": 725, "steps_done": 1484800, "mean_step_reward": 2.084, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 130.965, "mean_ep_length": 63.0, "mean_collisions": 1.88, "mean_intent_rate": 0.969, "policy_loss": 0.001399, "value_loss": 301.804659, "entropy": 0.048347, "total_loss": 150.903243, "clip_frac": 0.009644, "sps": 410.6, "elapsed_s": 3725.5}
+{"update": 726, "steps_done": 1486848, "mean_step_reward": 2.0633, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 130.079, "mean_ep_length": 62.9, "mean_collisions": 1.76, "mean_intent_rate": 0.957, "policy_loss": 0.001358, "value_loss": 252.362077, "entropy": 0.057423, "total_loss": 126.181823, "clip_frac": 0.013184, "sps": 405.5, "elapsed_s": 3730.6}
+{"update": 727, "steps_done": 1488896, "mean_step_reward": 2.0458, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 128.671, "mean_ep_length": 63.1, "mean_collisions": 2.03, "mean_intent_rate": 0.961, "policy_loss": 0.000295, "value_loss": 320.70655, "entropy": 0.050755, "total_loss": 160.353063, "clip_frac": 0.014893, "sps": 413.1, "elapsed_s": 3735.6}
+{"update": 728, "steps_done": 1490944, "mean_step_reward": 2.2477, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 131.904, "mean_ep_length": 58.6, "mean_collisions": 0.71, "mean_intent_rate": 0.97, "policy_loss": -0.000691, "value_loss": 214.481829, "entropy": 0.049478, "total_loss": 107.239729, "clip_frac": 0.019653, "sps": 412.6, "elapsed_s": 3740.5}
+{"update": 729, "steps_done": 1492992, "mean_step_reward": 2.3129, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.072, "mean_ep_length": 64.1, "mean_collisions": 0.66, "mean_intent_rate": 0.978, "policy_loss": 0.001307, "value_loss": 252.645025, "entropy": 0.033361, "total_loss": 126.323486, "clip_frac": 0.006592, "sps": 409.3, "elapsed_s": 3745.5}
+{"update": 730, "steps_done": 1495040, "mean_step_reward": 2.2111, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 139.719, "mean_ep_length": 63.1, "mean_collisions": 1.03, "mean_intent_rate": 0.969, "policy_loss": 0.001576, "value_loss": 217.982292, "entropy": 0.046804, "total_loss": 108.992254, "clip_frac": 0.013794, "sps": 415.2, "elapsed_s": 3750.5}
+{"update": 731, "steps_done": 1497088, "mean_step_reward": 2.0217, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 123.895, "mean_ep_length": 61.4, "mean_collisions": 2.09, "mean_intent_rate": 0.959, "policy_loss": 0.00062, "value_loss": 246.170938, "entropy": 0.047974, "total_loss": 123.08561, "clip_frac": 0.012329, "sps": 411.9, "elapsed_s": 3755.4}
+{"update": 732, "steps_done": 1499136, "mean_step_reward": 2.1632, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 129.646, "mean_ep_length": 60.0, "mean_collisions": 1.26, "mean_intent_rate": 0.966, "policy_loss": -0.000171, "value_loss": 212.789104, "entropy": 0.045626, "total_loss": 106.393925, "clip_frac": 0.014282, "sps": 416.2, "elapsed_s": 3760.4}
+{"update": 733, "steps_done": 1501184, "mean_step_reward": 2.2974, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 135.023, "mean_ep_length": 58.9, "mean_collisions": 0.65, "mean_intent_rate": 0.975, "policy_loss": 9e-06, "value_loss": 262.872878, "entropy": 0.036611, "total_loss": 131.43608, "clip_frac": 0.007935, "sps": 413.0, "elapsed_s": 3765.3}
+{"update": 734, "steps_done": 1503232, "mean_step_reward": 2.1793, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 135.417, "mean_ep_length": 61.9, "mean_collisions": 1.09, "mean_intent_rate": 0.96, "policy_loss": 0.001681, "value_loss": 219.114075, "entropy": 0.051838, "total_loss": 109.558199, "clip_frac": 0.014038, "sps": 414.4, "elapsed_s": 3770.3}
+{"update": 735, "steps_done": 1505280, "mean_step_reward": 1.8623, "min_step_reward": -10.5608, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 114.148, "mean_ep_length": 61.4, "mean_collisions": 3.03, "mean_intent_rate": 0.964, "policy_loss": 0.00346, "value_loss": 289.723843, "entropy": 0.0508, "total_loss": 144.864874, "clip_frac": 0.019165, "sps": 405.1, "elapsed_s": 3775.3}
+{"update": 736, "steps_done": 1507328, "mean_step_reward": 1.9718, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 131.098, "mean_ep_length": 63.5, "mean_collisions": 1.97, "mean_intent_rate": 0.966, "policy_loss": 2e-05, "value_loss": 245.93219, "entropy": 0.054107, "total_loss": 122.965574, "clip_frac": 0.011353, "sps": 410.1, "elapsed_s": 3780.3}
+{"update": 737, "steps_done": 1509376, "mean_step_reward": 2.1316, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 130.735, "mean_ep_length": 64.3, "mean_collisions": 2.06, "mean_intent_rate": 0.959, "policy_loss": 0.000109, "value_loss": 226.962672, "entropy": 0.043918, "total_loss": 113.481005, "clip_frac": 0.013794, "sps": 406.9, "elapsed_s": 3785.3}
+{"update": 738, "steps_done": 1511424, "mean_step_reward": 2.1345, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 137.082, "mean_ep_length": 64.2, "mean_collisions": 1.47, "mean_intent_rate": 0.966, "policy_loss": 0.000848, "value_loss": 233.028806, "entropy": 0.047926, "total_loss": 116.514772, "clip_frac": 0.013062, "sps": 413.6, "elapsed_s": 3790.3}
+{"update": 739, "steps_done": 1513472, "mean_step_reward": 2.1686, "min_step_reward": -10.563, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 137.388, "mean_ep_length": 63.4, "mean_collisions": 1.38, "mean_intent_rate": 0.968, "policy_loss": 0.000921, "value_loss": 220.824576, "entropy": 0.045919, "total_loss": 110.412751, "clip_frac": 0.011597, "sps": 408.9, "elapsed_s": 3795.3}
+{"update": 740, "steps_done": 1515520, "mean_step_reward": 2.2083, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 144.61, "mean_ep_length": 65.3, "mean_collisions": 1.0, "mean_intent_rate": 0.966, "policy_loss": 0.001228, "value_loss": 221.449482, "entropy": 0.050754, "total_loss": 110.725463, "clip_frac": 0.010132, "sps": 408.7, "elapsed_s": 3800.3}
+{"update": 741, "steps_done": 1517568, "mean_step_reward": 2.2628, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 137.341, "mean_ep_length": 60.8, "mean_collisions": 0.64, "mean_intent_rate": 0.963, "policy_loss": 0.00082, "value_loss": 203.340797, "entropy": 0.043112, "total_loss": 101.670788, "clip_frac": 0.010864, "sps": 411.5, "elapsed_s": 3805.3}
+{"update": 742, "steps_done": 1519616, "mean_step_reward": 2.2554, "min_step_reward": -10.5848, "max_step_reward": 2.5, "n_episodes_finished": 37, "mean_ep_return": 128.741, "mean_ep_length": 56.9, "mean_collisions": 0.68, "mean_intent_rate": 0.968, "policy_loss": 0.001197, "value_loss": 215.118337, "entropy": 0.04384, "total_loss": 107.559927, "clip_frac": 0.013672, "sps": 410.2, "elapsed_s": 3810.3}
+{"update": 743, "steps_done": 1521664, "mean_step_reward": 1.895, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 118.274, "mean_ep_length": 62.8, "mean_collisions": 2.88, "mean_intent_rate": 0.958, "policy_loss": 0.001282, "value_loss": 252.917989, "entropy": 0.050928, "total_loss": 126.459765, "clip_frac": 0.010254, "sps": 405.9, "elapsed_s": 3815.3}
+{"update": 744, "steps_done": 1523712, "mean_step_reward": 2.2253, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 135.486, "mean_ep_length": 60.8, "mean_collisions": 1.0, "mean_intent_rate": 0.971, "policy_loss": -0.000291, "value_loss": 221.523157, "entropy": 0.046461, "total_loss": 110.760823, "clip_frac": 0.007935, "sps": 411.0, "elapsed_s": 3820.3}
+{"update": 745, "steps_done": 1525760, "mean_step_reward": 2.1832, "min_step_reward": -10.5433, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 133.03, "mean_ep_length": 61.1, "mean_collisions": 1.18, "mean_intent_rate": 0.965, "policy_loss": -0.000139, "value_loss": 244.563519, "entropy": 0.049525, "total_loss": 122.281125, "clip_frac": 0.011841, "sps": 408.2, "elapsed_s": 3825.3}
+{"update": 746, "steps_done": 1527808, "mean_step_reward": 2.2479, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 127.98, "mean_ep_length": 56.9, "mean_collisions": 0.58, "mean_intent_rate": 0.958, "policy_loss": 0.000176, "value_loss": 236.173355, "entropy": 0.047824, "total_loss": 118.086375, "clip_frac": 0.013916, "sps": 415.3, "elapsed_s": 3830.3}
+{"update": 747, "steps_done": 1529856, "mean_step_reward": 2.2239, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 136.962, "mean_ep_length": 60.6, "mean_collisions": 0.71, "mean_intent_rate": 0.967, "policy_loss": 0.000687, "value_loss": 225.301947, "entropy": 0.052406, "total_loss": 112.651138, "clip_frac": 0.009888, "sps": 411.9, "elapsed_s": 3835.2}
+{"update": 748, "steps_done": 1531904, "mean_step_reward": 2.2801, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 141.363, "mean_ep_length": 62.9, "mean_collisions": 0.79, "mean_intent_rate": 0.966, "policy_loss": -0.000122, "value_loss": 194.10582, "entropy": 0.044609, "total_loss": 97.052342, "clip_frac": 0.013794, "sps": 416.3, "elapsed_s": 3840.2}
+{"update": 749, "steps_done": 1533952, "mean_step_reward": 2.0332, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 127.537, "mean_ep_length": 63.0, "mean_collisions": 2.16, "mean_intent_rate": 0.963, "policy_loss": 0.000133, "value_loss": 272.482906, "entropy": 0.054165, "total_loss": 136.241042, "clip_frac": 0.015503, "sps": 410.6, "elapsed_s": 3845.1}
+{"update": 750, "steps_done": 1536000, "mean_step_reward": 2.2092, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 128.285, "mean_ep_length": 58.1, "mean_collisions": 1.09, "mean_intent_rate": 0.971, "policy_loss": 0.000616, "value_loss": 256.678998, "entropy": 0.04205, "total_loss": 128.339696, "clip_frac": 0.006226, "sps": 411.9, "elapsed_s": 3850.1}
+{"update": 751, "steps_done": 1538048, "mean_step_reward": 2.3843, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 138.759, "mean_ep_length": 58.1, "mean_collisions": 0.11, "mean_intent_rate": 0.973, "policy_loss": 0.000612, "value_loss": 211.645803, "entropy": 0.034659, "total_loss": 105.823168, "clip_frac": 0.005371, "sps": 416.3, "elapsed_s": 3855.0}
+{"update": 752, "steps_done": 1540096, "mean_step_reward": 2.2317, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 126.328, "mean_ep_length": 56.6, "mean_collisions": 0.86, "mean_intent_rate": 0.964, "policy_loss": 5.1e-05, "value_loss": 248.390842, "entropy": 0.036498, "total_loss": 124.195107, "clip_frac": 0.007324, "sps": 401.4, "elapsed_s": 3860.2}
+{"update": 753, "steps_done": 1542144, "mean_step_reward": 2.1078, "min_step_reward": -10.581, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 121.711, "mean_ep_length": 57.6, "mean_collisions": 1.44, "mean_intent_rate": 0.961, "policy_loss": -0.000389, "value_loss": 270.18937, "entropy": 0.044866, "total_loss": 135.093848, "clip_frac": 0.015747, "sps": 411.7, "elapsed_s": 3865.1}
+{"update": 754, "steps_done": 1544192, "mean_step_reward": 1.8863, "min_step_reward": -10.5849, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 114.098, "mean_ep_length": 60.8, "mean_collisions": 3.0, "mean_intent_rate": 0.964, "policy_loss": -0.001509, "value_loss": 338.931832, "entropy": 0.06072, "total_loss": 169.4638, "clip_frac": 0.013794, "sps": 414.2, "elapsed_s": 3870.1}
+{"update": 755, "steps_done": 1546240, "mean_step_reward": 2.2207, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 148.669, "mean_ep_length": 66.9, "mean_collisions": 1.1, "mean_intent_rate": 0.967, "policy_loss": 0.000368, "value_loss": 309.419113, "entropy": 0.040733, "total_loss": 154.709518, "clip_frac": 0.005127, "sps": 409.2, "elapsed_s": 3875.1}
+{"update": 756, "steps_done": 1548288, "mean_step_reward": 2.1514, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 138.537, "mean_ep_length": 64.3, "mean_collisions": 1.41, "mean_intent_rate": 0.969, "policy_loss": 0.002384, "value_loss": 264.042058, "entropy": 0.042343, "total_loss": 132.022988, "clip_frac": 0.010254, "sps": 418.2, "elapsed_s": 3880.0}
+{"update": 757, "steps_done": 1550336, "mean_step_reward": 2.1508, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 130.098, "mean_ep_length": 60.7, "mean_collisions": 1.36, "mean_intent_rate": 0.967, "policy_loss": 4e-05, "value_loss": 242.302335, "entropy": 0.047329, "total_loss": 121.150735, "clip_frac": 0.011719, "sps": 409.6, "elapsed_s": 3885.0}
+{"update": 758, "steps_done": 1552384, "mean_step_reward": 2.1231, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 122.573, "mean_ep_length": 57.3, "mean_collisions": 1.11, "mean_intent_rate": 0.954, "policy_loss": 0.000708, "value_loss": 229.282941, "entropy": 0.056975, "total_loss": 114.641608, "clip_frac": 0.006104, "sps": 414.0, "elapsed_s": 3889.9}
+{"update": 759, "steps_done": 1554432, "mean_step_reward": 2.1562, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 134.722, "mean_ep_length": 62.8, "mean_collisions": 1.38, "mean_intent_rate": 0.964, "policy_loss": 0.002839, "value_loss": 237.512819, "entropy": 0.053919, "total_loss": 118.75871, "clip_frac": 0.019287, "sps": 409.4, "elapsed_s": 3894.9}
+{"update": 760, "steps_done": 1556480, "mean_step_reward": 2.2335, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 143.412, "mean_ep_length": 64.2, "mean_collisions": 0.91, "mean_intent_rate": 0.963, "policy_loss": 0.000496, "value_loss": 241.840031, "entropy": 0.039598, "total_loss": 120.920115, "clip_frac": 0.018799, "sps": 400.8, "elapsed_s": 3900.0}
+{"update": 761, "steps_done": 1558528, "mean_step_reward": 2.1858, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 134.368, "mean_ep_length": 61.8, "mean_collisions": 1.25, "mean_intent_rate": 0.966, "policy_loss": -1.1e-05, "value_loss": 220.410028, "entropy": 0.051074, "total_loss": 110.204491, "clip_frac": 0.013916, "sps": 417.9, "elapsed_s": 3904.9}
+{"update": 762, "steps_done": 1560576, "mean_step_reward": 2.1181, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 133.266, "mean_ep_length": 62.8, "mean_collisions": 1.33, "mean_intent_rate": 0.955, "policy_loss": 0.001826, "value_loss": 203.808702, "entropy": 0.050851, "total_loss": 101.905668, "clip_frac": 0.013916, "sps": 414.5, "elapsed_s": 3909.9}
+{"update": 763, "steps_done": 1562624, "mean_step_reward": 2.1585, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 141.949, "mean_ep_length": 65.6, "mean_collisions": 1.25, "mean_intent_rate": 0.96, "policy_loss": 0.000816, "value_loss": 242.056902, "entropy": 0.048019, "total_loss": 121.028788, "clip_frac": 0.007324, "sps": 411.2, "elapsed_s": 3914.9}
+{"update": 764, "steps_done": 1564672, "mean_step_reward": 2.2401, "min_step_reward": -7.5847, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 144.872, "mean_ep_length": 64.8, "mean_collisions": 0.94, "mean_intent_rate": 0.967, "policy_loss": 0.001361, "value_loss": 224.044322, "entropy": 0.043766, "total_loss": 112.023084, "clip_frac": 0.012207, "sps": 408.8, "elapsed_s": 3919.9}
+{"update": 765, "steps_done": 1566720, "mean_step_reward": 2.0614, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 136.571, "mean_ep_length": 66.2, "mean_collisions": 2.03, "mean_intent_rate": 0.963, "policy_loss": 0.002191, "value_loss": 294.820036, "entropy": 0.046829, "total_loss": 147.41174, "clip_frac": 0.015991, "sps": 411.6, "elapsed_s": 3924.8}
+{"update": 766, "steps_done": 1568768, "mean_step_reward": 2.249, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 144.32, "mean_ep_length": 64.2, "mean_collisions": 0.91, "mean_intent_rate": 0.973, "policy_loss": -0.000327, "value_loss": 221.404764, "entropy": 0.035755, "total_loss": 110.701698, "clip_frac": 0.011353, "sps": 414.8, "elapsed_s": 3929.8}
+{"update": 767, "steps_done": 1570816, "mean_step_reward": 2.154, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 133.082, "mean_ep_length": 61.8, "mean_collisions": 1.48, "mean_intent_rate": 0.973, "policy_loss": 0.000557, "value_loss": 251.869582, "entropy": 0.036679, "total_loss": 125.934982, "clip_frac": 0.006226, "sps": 418.3, "elapsed_s": 3934.7}
+{"update": 768, "steps_done": 1572864, "mean_step_reward": 2.3555, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 143.379, "mean_ep_length": 60.9, "mean_collisions": 0.36, "mean_intent_rate": 0.978, "policy_loss": 0.00079, "value_loss": 216.422828, "entropy": 0.034062, "total_loss": 108.211865, "clip_frac": 0.005127, "sps": 409.2, "elapsed_s": 3939.7}
+{"update": 769, "steps_done": 1574912, "mean_step_reward": 2.2312, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 142.309, "mean_ep_length": 63.6, "mean_collisions": 0.88, "mean_intent_rate": 0.967, "policy_loss": 0.001265, "value_loss": 260.898745, "entropy": 0.043484, "total_loss": 130.450202, "clip_frac": 0.010864, "sps": 406.2, "elapsed_s": 3944.7}
+{"update": 770, "steps_done": 1576960, "mean_step_reward": 2.0668, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 126.442, "mean_ep_length": 61.3, "mean_collisions": 1.94, "mean_intent_rate": 0.969, "policy_loss": 0.006056, "value_loss": 271.348691, "entropy": 0.045656, "total_loss": 135.679944, "clip_frac": 0.01416, "sps": 412.0, "elapsed_s": 3949.7}
+{"update": 771, "steps_done": 1579008, "mean_step_reward": 2.1473, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 126.27, "mean_ep_length": 58.7, "mean_collisions": 1.46, "mean_intent_rate": 0.971, "policy_loss": 0.00049, "value_loss": 272.226114, "entropy": 0.036999, "total_loss": 136.113177, "clip_frac": 0.005493, "sps": 413.1, "elapsed_s": 3954.7}
+{"update": 772, "steps_done": 1581056, "mean_step_reward": 2.1726, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 137.146, "mean_ep_length": 63.0, "mean_collisions": 1.18, "mean_intent_rate": 0.965, "policy_loss": -0.000407, "value_loss": 239.042689, "entropy": 0.045431, "total_loss": 119.520484, "clip_frac": 0.007812, "sps": 412.8, "elapsed_s": 3959.6}
+{"update": 773, "steps_done": 1583104, "mean_step_reward": 2.3874, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 152.03, "mean_ep_length": 63.7, "mean_collisions": 0.19, "mean_intent_rate": 0.98, "policy_loss": 0.000316, "value_loss": 157.899971, "entropy": 0.027179, "total_loss": 78.95003, "clip_frac": 0.004028, "sps": 416.5, "elapsed_s": 3964.5}
+{"update": 774, "steps_done": 1585152, "mean_step_reward": 2.2011, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 126.455, "mean_ep_length": 57.4, "mean_collisions": 1.06, "mean_intent_rate": 0.971, "policy_loss": 0.000592, "value_loss": 224.868801, "entropy": 0.046609, "total_loss": 112.434525, "clip_frac": 0.016113, "sps": 418.3, "elapsed_s": 3969.4}
+{"update": 775, "steps_done": 1587200, "mean_step_reward": 2.231, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 137.716, "mean_ep_length": 61.7, "mean_collisions": 1.09, "mean_intent_rate": 0.974, "policy_loss": -0.000224, "value_loss": 206.028863, "entropy": 0.036135, "total_loss": 103.013846, "clip_frac": 0.011475, "sps": 409.6, "elapsed_s": 3974.4}
+{"update": 776, "steps_done": 1589248, "mean_step_reward": 2.2627, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 136.509, "mean_ep_length": 60.2, "mean_collisions": 0.68, "mean_intent_rate": 0.968, "policy_loss": 0.001209, "value_loss": 186.347756, "entropy": 0.040212, "total_loss": 93.174687, "clip_frac": 0.006836, "sps": 410.4, "elapsed_s": 3979.4}
+{"update": 777, "steps_done": 1591296, "mean_step_reward": 2.2738, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 136.145, "mean_ep_length": 60.0, "mean_collisions": 0.76, "mean_intent_rate": 0.973, "policy_loss": 0.001865, "value_loss": 199.895546, "entropy": 0.04027, "total_loss": 99.949236, "clip_frac": 0.008545, "sps": 415.7, "elapsed_s": 3984.4}
+{"update": 778, "steps_done": 1593344, "mean_step_reward": 2.2658, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 142.925, "mean_ep_length": 63.2, "mean_collisions": 0.84, "mean_intent_rate": 0.971, "policy_loss": -0.000192, "value_loss": 196.714828, "entropy": 0.039926, "total_loss": 98.356823, "clip_frac": 0.004639, "sps": 405.7, "elapsed_s": 3989.4}
+{"update": 779, "steps_done": 1595392, "mean_step_reward": 2.3386, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 142.095, "mean_ep_length": 60.7, "mean_collisions": 0.44, "mean_intent_rate": 0.975, "policy_loss": -1.4e-05, "value_loss": 228.61623, "entropy": 0.036782, "total_loss": 114.307733, "clip_frac": 0.003906, "sps": 411.3, "elapsed_s": 3994.4}
+{"update": 780, "steps_done": 1597440, "mean_step_reward": 2.3733, "min_step_reward": -7.5867, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 151.6, "mean_ep_length": 63.9, "mean_collisions": 0.22, "mean_intent_rate": 0.977, "policy_loss": 4.4e-05, "value_loss": 261.257518, "entropy": 0.033883, "total_loss": 130.628466, "clip_frac": 0.003174, "sps": 410.1, "elapsed_s": 3999.4}
+{"update": 781, "steps_done": 1599488, "mean_step_reward": 2.3499, "min_step_reward": -10.5776, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 146.797, "mean_ep_length": 62.5, "mean_collisions": 0.39, "mean_intent_rate": 0.979, "policy_loss": 0.000193, "value_loss": 255.591683, "entropy": 0.027619, "total_loss": 127.795758, "clip_frac": 0.001953, "sps": 412.2, "elapsed_s": 4004.3}
+{"update": 782, "steps_done": 1601536, "mean_step_reward": 2.1936, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 138.334, "mean_ep_length": 63.1, "mean_collisions": 1.31, "mean_intent_rate": 0.972, "policy_loss": 0.000914, "value_loss": 303.237875, "entropy": 0.037687, "total_loss": 151.619475, "clip_frac": 0.007812, "sps": 409.5, "elapsed_s": 4009.4}
+{"update": 783, "steps_done": 1603584, "mean_step_reward": 2.1415, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 124.05, "mean_ep_length": 58.1, "mean_collisions": 1.34, "mean_intent_rate": 0.967, "policy_loss": -0.000171, "value_loss": 248.685287, "entropy": 0.039771, "total_loss": 124.342074, "clip_frac": 0.006592, "sps": 418.9, "elapsed_s": 4014.2}
+{"update": 784, "steps_done": 1605632, "mean_step_reward": 2.0294, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 132.472, "mean_ep_length": 65.0, "mean_collisions": 2.34, "mean_intent_rate": 0.973, "policy_loss": 0.000583, "value_loss": 301.977625, "entropy": 0.040746, "total_loss": 150.988989, "clip_frac": 0.010132, "sps": 412.9, "elapsed_s": 4019.2}
+{"update": 785, "steps_done": 1607680, "mean_step_reward": 2.1012, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 135.638, "mean_ep_length": 64.5, "mean_collisions": 1.88, "mean_intent_rate": 0.975, "policy_loss": -0.000694, "value_loss": 259.687195, "entropy": 0.035875, "total_loss": 129.842544, "clip_frac": 0.003174, "sps": 406.5, "elapsed_s": 4024.2}
+{"update": 786, "steps_done": 1609728, "mean_step_reward": 2.1884, "min_step_reward": -10.5612, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 130.78, "mean_ep_length": 59.8, "mean_collisions": 1.21, "mean_intent_rate": 0.971, "policy_loss": -0.000652, "value_loss": 214.197532, "entropy": 0.032546, "total_loss": 107.097789, "clip_frac": 0.005859, "sps": 413.1, "elapsed_s": 4029.2}
+{"update": 787, "steps_done": 1611776, "mean_step_reward": 2.3096, "min_step_reward": -7.588, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 149.796, "mean_ep_length": 64.7, "mean_collisions": 0.48, "mean_intent_rate": 0.972, "policy_loss": 0.000706, "value_loss": 205.807374, "entropy": 0.036553, "total_loss": 102.904027, "clip_frac": 0.012939, "sps": 411.3, "elapsed_s": 4034.2}
+{"update": 788, "steps_done": 1613824, "mean_step_reward": 2.3696, "min_step_reward": -7.5868, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 137.522, "mean_ep_length": 58.3, "mean_collisions": 0.2, "mean_intent_rate": 0.973, "policy_loss": -0.000419, "value_loss": 185.673227, "entropy": 0.028467, "total_loss": 92.83591, "clip_frac": 0.008545, "sps": 413.6, "elapsed_s": 4039.1}
+{"update": 789, "steps_done": 1615872, "mean_step_reward": 2.2523, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.017, "mean_ep_length": 65.6, "mean_collisions": 0.84, "mean_intent_rate": 0.969, "policy_loss": 0.000257, "value_loss": 239.426139, "entropy": 0.033868, "total_loss": 119.712988, "clip_frac": 0.007812, "sps": 410.8, "elapsed_s": 4044.1}
+{"update": 790, "steps_done": 1617920, "mean_step_reward": 2.1261, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 131.424, "mean_ep_length": 61.8, "mean_collisions": 1.7, "mean_intent_rate": 0.974, "policy_loss": -0.00155, "value_loss": 304.639571, "entropy": 0.03459, "total_loss": 152.317891, "clip_frac": 0.008423, "sps": 411.9, "elapsed_s": 4049.1}
+{"update": 791, "steps_done": 1619968, "mean_step_reward": 2.191, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 133.957, "mean_ep_length": 61.0, "mean_collisions": 1.32, "mean_intent_rate": 0.977, "policy_loss": -0.000717, "value_loss": 240.082762, "entropy": 0.0323, "total_loss": 120.040342, "clip_frac": 0.008301, "sps": 408.3, "elapsed_s": 4054.1}
+{"update": 792, "steps_done": 1622016, "mean_step_reward": 2.4144, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.212, "mean_ep_length": 62.2, "mean_collisions": 0.0, "mean_intent_rate": 0.979, "policy_loss": 0.000391, "value_loss": 181.333243, "entropy": 0.025369, "total_loss": 90.666759, "clip_frac": 0.004395, "sps": 413.4, "elapsed_s": 4059.1}
+{"update": 793, "steps_done": 1624064, "mean_step_reward": 2.1481, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 129.244, "mean_ep_length": 60.2, "mean_collisions": 1.41, "mean_intent_rate": 0.968, "policy_loss": -2.5e-05, "value_loss": 258.239523, "entropy": 0.048541, "total_loss": 129.119253, "clip_frac": 0.019897, "sps": 412.1, "elapsed_s": 4064.0}
+{"update": 794, "steps_done": 1626112, "mean_step_reward": 2.1681, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 136.651, "mean_ep_length": 63.2, "mean_collisions": 1.31, "mean_intent_rate": 0.967, "policy_loss": -0.000243, "value_loss": 220.058413, "entropy": 0.044216, "total_loss": 110.028521, "clip_frac": 0.00647, "sps": 402.5, "elapsed_s": 4069.1}
+{"update": 795, "steps_done": 1628160, "mean_step_reward": 2.166, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 135.947, "mean_ep_length": 62.7, "mean_collisions": 1.42, "mean_intent_rate": 0.973, "policy_loss": 0.001372, "value_loss": 203.320058, "entropy": 0.033255, "total_loss": 101.661068, "clip_frac": 0.009521, "sps": 411.4, "elapsed_s": 4074.1}
+{"update": 796, "steps_done": 1630208, "mean_step_reward": 2.1908, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 124.691, "mean_ep_length": 56.9, "mean_collisions": 1.0, "mean_intent_rate": 0.963, "policy_loss": 0.000567, "value_loss": 202.697217, "entropy": 0.042815, "total_loss": 101.348747, "clip_frac": 0.008911, "sps": 421.3, "elapsed_s": 4079.0}
+{"update": 797, "steps_done": 1632256, "mean_step_reward": 2.2034, "min_step_reward": -10.5245, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 141.019, "mean_ep_length": 64.0, "mean_collisions": 1.25, "mean_intent_rate": 0.974, "policy_loss": -0.000797, "value_loss": 230.404573, "entropy": 0.037003, "total_loss": 115.20112, "clip_frac": 0.004028, "sps": 409.1, "elapsed_s": 4084.0}
+{"update": 798, "steps_done": 1634304, "mean_step_reward": 2.2211, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 142.932, "mean_ep_length": 64.3, "mean_collisions": 1.0, "mean_intent_rate": 0.968, "policy_loss": 0.001203, "value_loss": 186.89062, "entropy": 0.037383, "total_loss": 93.44614, "clip_frac": 0.015015, "sps": 411.2, "elapsed_s": 4089.0}
+{"update": 799, "steps_done": 1636352, "mean_step_reward": 2.1391, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 136.836, "mean_ep_length": 64.3, "mean_collisions": 1.58, "mean_intent_rate": 0.966, "policy_loss": -2e-05, "value_loss": 213.555559, "entropy": 0.044548, "total_loss": 106.777313, "clip_frac": 0.005859, "sps": 408.5, "elapsed_s": 4094.0}
+{"update": 800, "steps_done": 1638400, "mean_step_reward": 2.3607, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.866, "mean_ep_length": 63.8, "mean_collisions": 0.3, "mean_intent_rate": 0.979, "policy_loss": -0.000523, "value_loss": 192.169149, "entropy": 0.028508, "total_loss": 96.083766, "clip_frac": 0.005005, "sps": 406.8, "elapsed_s": 4099.0}
+{"update": 801, "steps_done": 1640448, "mean_step_reward": 2.1711, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 138.644, "mean_ep_length": 63.9, "mean_collisions": 1.41, "mean_intent_rate": 0.971, "policy_loss": 0.00141, "value_loss": 230.826074, "entropy": 0.034259, "total_loss": 115.414105, "clip_frac": 0.006592, "sps": 412.0, "elapsed_s": 4104.0}
+{"update": 802, "steps_done": 1642496, "mean_step_reward": 2.1052, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 133.249, "mean_ep_length": 63.4, "mean_collisions": 1.69, "mean_intent_rate": 0.964, "policy_loss": -0.000509, "value_loss": 244.385384, "entropy": 0.043404, "total_loss": 122.191748, "clip_frac": 0.016235, "sps": 407.1, "elapsed_s": 4109.0}
+{"update": 803, "steps_done": 1644544, "mean_step_reward": 2.2971, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 146.142, "mean_ep_length": 63.5, "mean_collisions": 0.62, "mean_intent_rate": 0.973, "policy_loss": 0.000438, "value_loss": 183.656304, "entropy": 0.033679, "total_loss": 91.828254, "clip_frac": 0.005737, "sps": 413.8, "elapsed_s": 4114.0}
+{"update": 804, "steps_done": 1646592, "mean_step_reward": 2.1473, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 136.59, "mean_ep_length": 63.8, "mean_collisions": 1.47, "mean_intent_rate": 0.966, "policy_loss": 0.00024, "value_loss": 224.822626, "entropy": 0.046155, "total_loss": 112.411092, "clip_frac": 0.008179, "sps": 418.1, "elapsed_s": 4118.9}
+{"update": 805, "steps_done": 1648640, "mean_step_reward": 2.3831, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 138.069, "mean_ep_length": 57.9, "mean_collisions": 0.14, "mean_intent_rate": 0.975, "policy_loss": -0.000593, "value_loss": 203.832202, "entropy": 0.029179, "total_loss": 101.915217, "clip_frac": 0.007446, "sps": 409.7, "elapsed_s": 4123.9}
+{"update": 806, "steps_done": 1650688, "mean_step_reward": 2.3281, "min_step_reward": -10.5565, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 138.628, "mean_ep_length": 59.6, "mean_collisions": 0.35, "mean_intent_rate": 0.966, "policy_loss": 0.00059, "value_loss": 215.219008, "entropy": 0.032017, "total_loss": 107.609774, "clip_frac": 0.004395, "sps": 411.0, "elapsed_s": 4128.8}
+{"update": 807, "steps_done": 1652736, "mean_step_reward": 2.1875, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 132.787, "mean_ep_length": 60.9, "mean_collisions": 1.3, "mean_intent_rate": 0.973, "policy_loss": 0.001135, "value_loss": 222.794919, "entropy": 0.033341, "total_loss": 111.39826, "clip_frac": 0.004028, "sps": 409.2, "elapsed_s": 4133.9}
+{"update": 808, "steps_done": 1654784, "mean_step_reward": 2.1226, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 132.63, "mean_ep_length": 62.1, "mean_collisions": 1.56, "mean_intent_rate": 0.972, "policy_loss": -0.000449, "value_loss": 226.056945, "entropy": 0.039176, "total_loss": 113.027632, "clip_frac": 0.007446, "sps": 413.5, "elapsed_s": 4138.8}
+{"update": 809, "steps_done": 1656832, "mean_step_reward": 2.273, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 138.04, "mean_ep_length": 60.8, "mean_collisions": 0.76, "mean_intent_rate": 0.972, "policy_loss": 0.000424, "value_loss": 222.488059, "entropy": 0.03093, "total_loss": 111.244144, "clip_frac": 0.005859, "sps": 404.5, "elapsed_s": 4143.9}
+{"update": 810, "steps_done": 1658880, "mean_step_reward": 2.3851, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 148.034, "mean_ep_length": 61.9, "mean_collisions": 0.18, "mean_intent_rate": 0.98, "policy_loss": 2.7e-05, "value_loss": 170.343006, "entropy": 0.025347, "total_loss": 85.171276, "clip_frac": 0.004028, "sps": 419.0, "elapsed_s": 4148.8}
+{"update": 811, "steps_done": 1660928, "mean_step_reward": 2.1257, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 129.206, "mean_ep_length": 60.9, "mean_collisions": 1.59, "mean_intent_rate": 0.969, "policy_loss": -0.000638, "value_loss": 214.961135, "entropy": 0.03695, "total_loss": 107.47956, "clip_frac": 0.011719, "sps": 403.9, "elapsed_s": 4153.8}
+{"update": 812, "steps_done": 1662976, "mean_step_reward": 2.2776, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 139.014, "mean_ep_length": 61.0, "mean_collisions": 0.82, "mean_intent_rate": 0.978, "policy_loss": -0.000408, "value_loss": 220.256965, "entropy": 0.029489, "total_loss": 110.12778, "clip_frac": 0.00293, "sps": 415.1, "elapsed_s": 4158.8}
+{"update": 813, "steps_done": 1665024, "mean_step_reward": 2.259, "min_step_reward": -10.5645, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 132.189, "mean_ep_length": 58.5, "mean_collisions": 0.86, "mean_intent_rate": 0.978, "policy_loss": -0.000554, "value_loss": 210.825935, "entropy": 0.031416, "total_loss": 105.4121, "clip_frac": 0.001831, "sps": 411.6, "elapsed_s": 4163.7}
+{"update": 814, "steps_done": 1667072, "mean_step_reward": 2.2769, "min_step_reward": -10.5515, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 137.75, "mean_ep_length": 60.6, "mean_collisions": 0.79, "mean_intent_rate": 0.973, "policy_loss": -0.001003, "value_loss": 211.26576, "entropy": 0.03401, "total_loss": 105.631537, "clip_frac": 0.012939, "sps": 413.3, "elapsed_s": 4168.7}
+{"update": 815, "steps_done": 1669120, "mean_step_reward": 2.2726, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 138.306, "mean_ep_length": 60.8, "mean_collisions": 0.74, "mean_intent_rate": 0.975, "policy_loss": 0.000227, "value_loss": 225.690242, "entropy": 0.032716, "total_loss": 112.84502, "clip_frac": 0.001953, "sps": 410.4, "elapsed_s": 4173.7}
+{"update": 816, "steps_done": 1671168, "mean_step_reward": 2.3956, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.771, "mean_ep_length": 62.9, "mean_collisions": 0.15, "mean_intent_rate": 0.98, "policy_loss": 0.000164, "value_loss": 174.918596, "entropy": 0.02658, "total_loss": 87.459196, "clip_frac": 0.008301, "sps": 413.2, "elapsed_s": 4178.6}
+{"update": 817, "steps_done": 1673216, "mean_step_reward": 2.3233, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 156.265, "mean_ep_length": 67.3, "mean_collisions": 0.63, "mean_intent_rate": 0.981, "policy_loss": 0.000588, "value_loss": 214.815048, "entropy": 0.027527, "total_loss": 107.407836, "clip_frac": 0.01001, "sps": 407.7, "elapsed_s": 4183.7}
+{"update": 818, "steps_done": 1675264, "mean_step_reward": 2.1312, "min_step_reward": -10.583, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 130.246, "mean_ep_length": 61.3, "mean_collisions": 1.48, "mean_intent_rate": 0.967, "policy_loss": -9e-05, "value_loss": 219.72051, "entropy": 0.038365, "total_loss": 109.859782, "clip_frac": 0.012451, "sps": 413.9, "elapsed_s": 4188.6}
+{"update": 819, "steps_done": 1677312, "mean_step_reward": 2.1901, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 131.712, "mean_ep_length": 59.9, "mean_collisions": 1.14, "mean_intent_rate": 0.97, "policy_loss": 3.6e-05, "value_loss": 212.732713, "entropy": 0.033, "total_loss": 106.366063, "clip_frac": 0.007446, "sps": 406.1, "elapsed_s": 4193.7}
+{"update": 820, "steps_done": 1679360, "mean_step_reward": 2.4029, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 142.763, "mean_ep_length": 59.4, "mean_collisions": 0.09, "mean_intent_rate": 0.98, "policy_loss": 0.000231, "value_loss": 190.26143, "entropy": 0.025588, "total_loss": 95.130691, "clip_frac": 0.003418, "sps": 409.2, "elapsed_s": 4198.7}
+{"update": 821, "steps_done": 1681408, "mean_step_reward": 2.2736, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 135.023, "mean_ep_length": 59.3, "mean_collisions": 0.69, "mean_intent_rate": 0.973, "policy_loss": 0.000427, "value_loss": 197.023916, "entropy": 0.034108, "total_loss": 98.512044, "clip_frac": 0.008057, "sps": 414.7, "elapsed_s": 4203.6}
+{"update": 822, "steps_done": 1683456, "mean_step_reward": 2.321, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 136.235, "mean_ep_length": 58.7, "mean_collisions": 0.46, "mean_intent_rate": 0.973, "policy_loss": 0.001137, "value_loss": 174.940022, "entropy": 0.032527, "total_loss": 87.470822, "clip_frac": 0.006104, "sps": 410.7, "elapsed_s": 4208.6}
+{"update": 823, "steps_done": 1685504, "mean_step_reward": 2.3327, "min_step_reward": -10.5612, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 137.035, "mean_ep_length": 58.8, "mean_collisions": 0.41, "mean_intent_rate": 0.975, "policy_loss": -0.000215, "value_loss": 138.125682, "entropy": 0.031173, "total_loss": 69.062314, "clip_frac": 0.005249, "sps": 416.7, "elapsed_s": 4213.5}
+{"update": 824, "steps_done": 1687552, "mean_step_reward": 2.4282, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 159.747, "mean_ep_length": 65.8, "mean_collisions": 0.03, "mean_intent_rate": 0.983, "policy_loss": 1e-06, "value_loss": 157.165502, "entropy": 0.021926, "total_loss": 78.582533, "clip_frac": 0.004028, "sps": 408.1, "elapsed_s": 4218.5}
+{"update": 825, "steps_done": 1689600, "mean_step_reward": 2.3486, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.19, "mean_ep_length": 63.8, "mean_collisions": 0.42, "mean_intent_rate": 0.981, "policy_loss": -0.001159, "value_loss": 137.768979, "entropy": 0.024246, "total_loss": 68.883088, "clip_frac": 0.006592, "sps": 412.5, "elapsed_s": 4223.5}
+{"update": 826, "steps_done": 1691648, "mean_step_reward": 2.3366, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 143.281, "mean_ep_length": 61.4, "mean_collisions": 0.48, "mean_intent_rate": 0.98, "policy_loss": -0.000249, "value_loss": 150.808282, "entropy": 0.02234, "total_loss": 75.403669, "clip_frac": 0.001831, "sps": 405.4, "elapsed_s": 4228.5}
+{"update": 827, "steps_done": 1693696, "mean_step_reward": 2.1197, "min_step_reward": -10.5484, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 132.905, "mean_ep_length": 62.6, "mean_collisions": 1.73, "mean_intent_rate": 0.973, "policy_loss": -0.001134, "value_loss": 271.17001, "entropy": 0.02994, "total_loss": 135.58357, "clip_frac": 0.004272, "sps": 407.2, "elapsed_s": 4233.6}
+{"update": 828, "steps_done": 1695744, "mean_step_reward": 2.3609, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 142.575, "mean_ep_length": 60.4, "mean_collisions": 0.26, "mean_intent_rate": 0.976, "policy_loss": -0.000879, "value_loss": 122.439357, "entropy": 0.031139, "total_loss": 61.218488, "clip_frac": 0.005859, "sps": 405.7, "elapsed_s": 4238.6}
+{"update": 829, "steps_done": 1697792, "mean_step_reward": 2.3148, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 135.784, "mean_ep_length": 58.8, "mean_collisions": 0.5, "mean_intent_rate": 0.972, "policy_loss": 0.000772, "value_loss": 150.577712, "entropy": 0.031657, "total_loss": 75.289311, "clip_frac": 0.004883, "sps": 410.9, "elapsed_s": 4243.6}
+{"update": 830, "steps_done": 1699840, "mean_step_reward": 2.2675, "min_step_reward": -7.5862, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 140.654, "mean_ep_length": 61.4, "mean_collisions": 0.67, "mean_intent_rate": 0.971, "policy_loss": 0.001769, "value_loss": 195.962725, "entropy": 0.031676, "total_loss": 97.982814, "clip_frac": 0.01062, "sps": 414.9, "elapsed_s": 4248.5}
+{"update": 831, "steps_done": 1701888, "mean_step_reward": 2.2561, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 134.204, "mean_ep_length": 60.0, "mean_collisions": 1.0, "mean_intent_rate": 0.977, "policy_loss": -0.000864, "value_loss": 166.930285, "entropy": 0.034303, "total_loss": 83.463935, "clip_frac": 0.009521, "sps": 410.3, "elapsed_s": 4253.5}
+{"update": 832, "steps_done": 1703936, "mean_step_reward": 2.211, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 140.736, "mean_ep_length": 63.7, "mean_collisions": 1.22, "mean_intent_rate": 0.973, "policy_loss": -0.000148, "value_loss": 199.937239, "entropy": 0.026552, "total_loss": 99.968207, "clip_frac": 0.005981, "sps": 411.7, "elapsed_s": 4258.5}
+{"update": 833, "steps_done": 1705984, "mean_step_reward": 2.2673, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 157.256, "mean_ep_length": 69.3, "mean_collisions": 1.0, "mean_intent_rate": 0.98, "policy_loss": 0.000486, "value_loss": 223.439647, "entropy": 0.023008, "total_loss": 111.720079, "clip_frac": 0.002563, "sps": 412.9, "elapsed_s": 4263.5}
+{"update": 834, "steps_done": 1708032, "mean_step_reward": 2.4171, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 152.595, "mean_ep_length": 63.2, "mean_collisions": 0.03, "mean_intent_rate": 0.98, "policy_loss": -0.000128, "value_loss": 183.453351, "entropy": 0.023948, "total_loss": 91.726308, "clip_frac": 0.000854, "sps": 409.4, "elapsed_s": 4268.5}
+{"update": 835, "steps_done": 1710080, "mean_step_reward": 2.2461, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 134.646, "mean_ep_length": 60.0, "mean_collisions": 0.88, "mean_intent_rate": 0.97, "policy_loss": 0.001878, "value_loss": 201.499189, "entropy": 0.02943, "total_loss": 100.751179, "clip_frac": 0.009644, "sps": 412.1, "elapsed_s": 4273.5}
+{"update": 836, "steps_done": 1712128, "mean_step_reward": 2.3389, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 142.851, "mean_ep_length": 61.0, "mean_collisions": 0.35, "mean_intent_rate": 0.976, "policy_loss": 0.000933, "value_loss": 140.991032, "entropy": 0.025518, "total_loss": 70.496194, "clip_frac": 0.009277, "sps": 416.7, "elapsed_s": 4278.4}
+{"update": 837, "steps_done": 1714176, "mean_step_reward": 2.1702, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 141.641, "mean_ep_length": 65.4, "mean_collisions": 1.55, "mean_intent_rate": 0.977, "policy_loss": 0.000118, "value_loss": 251.986641, "entropy": 0.028727, "total_loss": 125.993152, "clip_frac": 0.009277, "sps": 405.4, "elapsed_s": 4283.4}
+{"update": 838, "steps_done": 1716224, "mean_step_reward": 2.3539, "min_step_reward": -7.5867, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 145.452, "mean_ep_length": 61.8, "mean_collisions": 0.39, "mean_intent_rate": 0.978, "policy_loss": 0.000117, "value_loss": 152.567541, "entropy": 0.021962, "total_loss": 76.283669, "clip_frac": 0.003784, "sps": 411.4, "elapsed_s": 4288.4}
+{"update": 839, "steps_done": 1718272, "mean_step_reward": 2.2838, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 143.39, "mean_ep_length": 62.8, "mean_collisions": 0.82, "mean_intent_rate": 0.979, "policy_loss": 0.001016, "value_loss": 180.564183, "entropy": 0.027331, "total_loss": 90.282834, "clip_frac": 0.005615, "sps": 409.5, "elapsed_s": 4293.4}
+{"update": 840, "steps_done": 1720320, "mean_step_reward": 2.4137, "min_step_reward": -7.5822, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 153.592, "mean_ep_length": 63.6, "mean_collisions": 0.03, "mean_intent_rate": 0.979, "policy_loss": -5e-05, "value_loss": 206.68455, "entropy": 0.022101, "total_loss": 103.342004, "clip_frac": 0.007568, "sps": 413.8, "elapsed_s": 4298.4}
+{"update": 841, "steps_done": 1722368, "mean_step_reward": 2.3764, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 144.14, "mean_ep_length": 60.6, "mean_collisions": 0.26, "mean_intent_rate": 0.98, "policy_loss": 1.8e-05, "value_loss": 184.259274, "entropy": 0.022122, "total_loss": 92.129433, "clip_frac": 0.001953, "sps": 413.9, "elapsed_s": 4303.3}
+{"update": 842, "steps_done": 1724416, "mean_step_reward": 2.4025, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 148.48, "mean_ep_length": 61.8, "mean_collisions": 0.09, "mean_intent_rate": 0.979, "policy_loss": -0.000132, "value_loss": 213.82472, "entropy": 0.023169, "total_loss": 106.911997, "clip_frac": 0.001465, "sps": 414.1, "elapsed_s": 4308.2}
+{"update": 843, "steps_done": 1726464, "mean_step_reward": 2.2525, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 149.295, "mean_ep_length": 66.3, "mean_collisions": 1.03, "mean_intent_rate": 0.979, "policy_loss": 8.2e-05, "value_loss": 242.739838, "entropy": 0.026623, "total_loss": 121.369735, "clip_frac": 0.010132, "sps": 406.0, "elapsed_s": 4313.3}
+{"update": 844, "steps_done": 1728512, "mean_step_reward": 2.2122, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 133.829, "mean_ep_length": 60.5, "mean_collisions": 1.15, "mean_intent_rate": 0.974, "policy_loss": 0.000927, "value_loss": 208.542865, "entropy": 0.029682, "total_loss": 104.272063, "clip_frac": 0.006348, "sps": 413.0, "elapsed_s": 4318.2}
+{"update": 845, "steps_done": 1730560, "mean_step_reward": 2.3164, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 147.094, "mean_ep_length": 63.5, "mean_collisions": 0.59, "mean_intent_rate": 0.977, "policy_loss": 4e-05, "value_loss": 167.304887, "entropy": 0.026023, "total_loss": 83.652223, "clip_frac": 0.002441, "sps": 405.5, "elapsed_s": 4323.3}
+{"update": 846, "steps_done": 1732608, "mean_step_reward": 2.3897, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 143.717, "mean_ep_length": 60.1, "mean_collisions": 0.18, "mean_intent_rate": 0.979, "policy_loss": 0.000639, "value_loss": 159.003433, "entropy": 0.022882, "total_loss": 79.502127, "clip_frac": 0.007324, "sps": 416.3, "elapsed_s": 4328.2}
+{"update": 847, "steps_done": 1734656, "mean_step_reward": 2.352, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 144.02, "mean_ep_length": 61.3, "mean_collisions": 0.39, "mean_intent_rate": 0.98, "policy_loss": 0.000468, "value_loss": 201.768904, "entropy": 0.022806, "total_loss": 100.884692, "clip_frac": 0.004761, "sps": 411.5, "elapsed_s": 4333.2}
+{"update": 848, "steps_done": 1736704, "mean_step_reward": 2.3187, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 151.696, "mean_ep_length": 65.0, "mean_collisions": 0.59, "mean_intent_rate": 0.981, "policy_loss": 0.001024, "value_loss": 227.790622, "entropy": 0.026179, "total_loss": 113.896073, "clip_frac": 0.004272, "sps": 411.4, "elapsed_s": 4338.2}
+{"update": 849, "steps_done": 1738752, "mean_step_reward": 1.9647, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 125.774, "mean_ep_length": 64.3, "mean_collisions": 2.5, "mean_intent_rate": 0.961, "policy_loss": -0.000257, "value_loss": 262.355948, "entropy": 0.047044, "total_loss": 131.177247, "clip_frac": 0.017822, "sps": 415.7, "elapsed_s": 4343.1}
+{"update": 850, "steps_done": 1740800, "mean_step_reward": 2.2173, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 132.455, "mean_ep_length": 59.8, "mean_collisions": 0.97, "mean_intent_rate": 0.969, "policy_loss": 0.000188, "value_loss": 183.567904, "entropy": 0.030878, "total_loss": 91.783832, "clip_frac": 0.012695, "sps": 412.3, "elapsed_s": 4348.1}
+{"update": 851, "steps_done": 1742848, "mean_step_reward": 2.3408, "min_step_reward": -7.5849, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 132.814, "mean_ep_length": 56.8, "mean_collisions": 0.39, "mean_intent_rate": 0.976, "policy_loss": -0.000269, "value_loss": 186.578077, "entropy": 0.027005, "total_loss": 93.288499, "clip_frac": 0.002319, "sps": 410.5, "elapsed_s": 4353.1}
+{"update": 852, "steps_done": 1744896, "mean_step_reward": 2.3413, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 137.004, "mean_ep_length": 58.5, "mean_collisions": 0.4, "mean_intent_rate": 0.975, "policy_loss": 0.0003, "value_loss": 155.350005, "entropy": 0.024652, "total_loss": 77.675056, "clip_frac": 0.007812, "sps": 408.3, "elapsed_s": 4358.1}
+{"update": 853, "steps_done": 1746944, "mean_step_reward": 2.3065, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 135.126, "mean_ep_length": 58.6, "mean_collisions": 0.57, "mean_intent_rate": 0.976, "policy_loss": 0.000294, "value_loss": 169.067552, "entropy": 0.026368, "total_loss": 84.533807, "clip_frac": 0.004883, "sps": 411.5, "elapsed_s": 4363.1}
+{"update": 854, "steps_done": 1748992, "mean_step_reward": 2.2496, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 144.822, "mean_ep_length": 64.3, "mean_collisions": 1.03, "mean_intent_rate": 0.977, "policy_loss": 9.8e-05, "value_loss": 192.854236, "entropy": 0.027586, "total_loss": 96.426941, "clip_frac": 0.004639, "sps": 405.9, "elapsed_s": 4368.1}
+{"update": 855, "steps_done": 1751040, "mean_step_reward": 2.2198, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 134.27, "mean_ep_length": 60.7, "mean_collisions": 1.21, "mean_intent_rate": 0.979, "policy_loss": 0.001394, "value_loss": 200.943887, "entropy": 0.028643, "total_loss": 100.473051, "clip_frac": 0.003906, "sps": 412.3, "elapsed_s": 4373.1}
+{"update": 856, "steps_done": 1753088, "mean_step_reward": 2.2571, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 142.564, "mean_ep_length": 63.0, "mean_collisions": 0.91, "mean_intent_rate": 0.976, "policy_loss": -9.8e-05, "value_loss": 211.769612, "entropy": 0.025484, "total_loss": 105.884453, "clip_frac": 0.005737, "sps": 408.0, "elapsed_s": 4378.1}
+{"update": 857, "steps_done": 1755136, "mean_step_reward": 2.1774, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 143.958, "mean_ep_length": 66.1, "mean_collisions": 1.45, "mean_intent_rate": 0.976, "policy_loss": 0.000435, "value_loss": 207.300433, "entropy": 0.026954, "total_loss": 103.650382, "clip_frac": 0.004517, "sps": 409.0, "elapsed_s": 4383.1}
+{"update": 858, "steps_done": 1757184, "mean_step_reward": 2.1845, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 125.78, "mean_ep_length": 57.7, "mean_collisions": 1.14, "mean_intent_rate": 0.966, "policy_loss": 0.000492, "value_loss": 214.808005, "entropy": 0.036545, "total_loss": 107.404129, "clip_frac": 0.006226, "sps": 414.8, "elapsed_s": 4388.0}
+{"update": 859, "steps_done": 1759232, "mean_step_reward": 2.2238, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 135.76, "mean_ep_length": 61.0, "mean_collisions": 0.88, "mean_intent_rate": 0.965, "policy_loss": -0.000655, "value_loss": 189.446548, "entropy": 0.037551, "total_loss": 94.722245, "clip_frac": 0.006836, "sps": 414.7, "elapsed_s": 4393.0}
+{"update": 860, "steps_done": 1761280, "mean_step_reward": 2.2465, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 139.27, "mean_ep_length": 62.0, "mean_collisions": 0.94, "mean_intent_rate": 0.973, "policy_loss": 8.8e-05, "value_loss": 234.738213, "entropy": 0.03665, "total_loss": 117.368829, "clip_frac": 0.010742, "sps": 409.5, "elapsed_s": 4398.0}
+{"update": 861, "steps_done": 1763328, "mean_step_reward": 2.2213, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 133.497, "mean_ep_length": 60.0, "mean_collisions": 0.85, "mean_intent_rate": 0.964, "policy_loss": -5.1e-05, "value_loss": 210.431166, "entropy": 0.037821, "total_loss": 105.215154, "clip_frac": 0.006836, "sps": 411.3, "elapsed_s": 4403.0}
+{"update": 862, "steps_done": 1765376, "mean_step_reward": 2.2565, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 148.987, "mean_ep_length": 66.0, "mean_collisions": 0.97, "mean_intent_rate": 0.974, "policy_loss": -0.000196, "value_loss": 203.273133, "entropy": 0.032262, "total_loss": 101.636048, "clip_frac": 0.00415, "sps": 399.5, "elapsed_s": 4408.1}
+{"update": 863, "steps_done": 1767424, "mean_step_reward": 2.3948, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 152.222, "mean_ep_length": 63.7, "mean_collisions": 0.19, "mean_intent_rate": 0.98, "policy_loss": 0.000111, "value_loss": 157.720394, "entropy": 0.024197, "total_loss": 78.860066, "clip_frac": 0.002563, "sps": 409.1, "elapsed_s": 4413.1}
+{"update": 864, "steps_done": 1769472, "mean_step_reward": 2.2275, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 136.486, "mean_ep_length": 61.2, "mean_collisions": 1.03, "mean_intent_rate": 0.974, "policy_loss": 0.000869, "value_loss": 210.626961, "entropy": 0.029409, "total_loss": 105.314055, "clip_frac": 0.007935, "sps": 411.7, "elapsed_s": 4418.1}
+{"update": 865, "steps_done": 1771520, "mean_step_reward": 2.2617, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 144.85, "mean_ep_length": 64.0, "mean_collisions": 0.91, "mean_intent_rate": 0.975, "policy_loss": -0.000169, "value_loss": 212.607812, "entropy": 0.02896, "total_loss": 106.303447, "clip_frac": 0.010742, "sps": 408.3, "elapsed_s": 4423.1}
+{"update": 866, "steps_done": 1773568, "mean_step_reward": 2.3118, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 144.28, "mean_ep_length": 62.4, "mean_collisions": 0.55, "mean_intent_rate": 0.972, "policy_loss": 0.000168, "value_loss": 179.455365, "entropy": 0.02723, "total_loss": 89.727577, "clip_frac": 0.010742, "sps": 409.8, "elapsed_s": 4428.1}
+{"update": 867, "steps_done": 1775616, "mean_step_reward": 2.3142, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 156.568, "mean_ep_length": 67.7, "mean_collisions": 0.63, "mean_intent_rate": 0.974, "policy_loss": -0.000734, "value_loss": 180.741945, "entropy": 0.026959, "total_loss": 90.369968, "clip_frac": 0.004761, "sps": 413.6, "elapsed_s": 4433.0}
+{"update": 868, "steps_done": 1777664, "mean_step_reward": 2.4283, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 147.522, "mean_ep_length": 60.7, "mean_collisions": 0.0, "mean_intent_rate": 0.982, "policy_loss": 1.8e-05, "value_loss": 173.205619, "entropy": 0.022404, "total_loss": 86.602604, "clip_frac": 0.001343, "sps": 419.8, "elapsed_s": 4437.9}
+{"update": 869, "steps_done": 1779712, "mean_step_reward": 2.4113, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 147.835, "mean_ep_length": 61.3, "mean_collisions": 0.06, "mean_intent_rate": 0.981, "policy_loss": -0.000263, "value_loss": 154.685001, "entropy": 0.023021, "total_loss": 77.342008, "clip_frac": 0.004272, "sps": 413.5, "elapsed_s": 4442.9}
+{"update": 870, "steps_done": 1781760, "mean_step_reward": 2.3872, "min_step_reward": -7.581, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 149.139, "mean_ep_length": 62.5, "mean_collisions": 0.18, "mean_intent_rate": 0.98, "policy_loss": 6.4e-05, "value_loss": 153.358526, "entropy": 0.023055, "total_loss": 76.679097, "clip_frac": 0.001099, "sps": 411.8, "elapsed_s": 4447.9}
+{"update": 871, "steps_done": 1783808, "mean_step_reward": 2.063, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 130.311, "mean_ep_length": 63.3, "mean_collisions": 2.19, "mean_intent_rate": 0.977, "policy_loss": 0.00091, "value_loss": 224.642846, "entropy": 0.033361, "total_loss": 112.321999, "clip_frac": 0.008301, "sps": 405.0, "elapsed_s": 4452.9}
+{"update": 872, "steps_done": 1785856, "mean_step_reward": 2.3945, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 151.394, "mean_ep_length": 63.2, "mean_collisions": 0.15, "mean_intent_rate": 0.981, "policy_loss": 0.000773, "value_loss": 118.729477, "entropy": 0.022588, "total_loss": 59.365285, "clip_frac": 0.005737, "sps": 407.8, "elapsed_s": 4457.9}
+{"update": 873, "steps_done": 1787904, "mean_step_reward": 2.2391, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 139.446, "mean_ep_length": 62.4, "mean_collisions": 1.0, "mean_intent_rate": 0.969, "policy_loss": -0.001189, "value_loss": 192.017311, "entropy": 0.032238, "total_loss": 96.007144, "clip_frac": 0.012573, "sps": 412.3, "elapsed_s": 4462.9}
+{"update": 874, "steps_done": 1789952, "mean_step_reward": 2.2384, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.103, "mean_ep_length": 64.3, "mean_collisions": 0.75, "mean_intent_rate": 0.983, "policy_loss": -0.000148, "value_loss": 192.695406, "entropy": 0.02227, "total_loss": 96.347332, "clip_frac": 0.000732, "sps": 412.1, "elapsed_s": 4467.9}
+{"update": 875, "steps_done": 1792000, "mean_step_reward": 2.1761, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 130.287, "mean_ep_length": 61.5, "mean_collisions": 1.71, "mean_intent_rate": 0.974, "policy_loss": -0.000213, "value_loss": 211.030549, "entropy": 0.031574, "total_loss": 105.514745, "clip_frac": 0.005371, "sps": 413.1, "elapsed_s": 4472.8}
+{"update": 876, "steps_done": 1794048, "mean_step_reward": 2.2767, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 139.641, "mean_ep_length": 61.4, "mean_collisions": 0.82, "mean_intent_rate": 0.974, "policy_loss": -0.00024, "value_loss": 183.254326, "entropy": 0.027791, "total_loss": 91.626645, "clip_frac": 0.004395, "sps": 411.8, "elapsed_s": 4477.8}
+{"update": 877, "steps_done": 1796096, "mean_step_reward": 2.2275, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 144.25, "mean_ep_length": 64.9, "mean_collisions": 1.13, "mean_intent_rate": 0.975, "policy_loss": 0.000478, "value_loss": 198.876825, "entropy": 0.033052, "total_loss": 99.43856, "clip_frac": 0.00354, "sps": 407.8, "elapsed_s": 4482.8}
+{"update": 878, "steps_done": 1798144, "mean_step_reward": 2.2047, "min_step_reward": -10.5238, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 144.314, "mean_ep_length": 65.5, "mean_collisions": 1.26, "mean_intent_rate": 0.969, "policy_loss": 0.001336, "value_loss": 221.864702, "entropy": 0.03458, "total_loss": 110.933341, "clip_frac": 0.009033, "sps": 407.8, "elapsed_s": 4487.9}
+{"update": 879, "steps_done": 1800192, "mean_step_reward": 2.243, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 134.22, "mean_ep_length": 59.7, "mean_collisions": 0.83, "mean_intent_rate": 0.971, "policy_loss": -0.000593, "value_loss": 203.513547, "entropy": 0.035581, "total_loss": 101.755825, "clip_frac": 0.012329, "sps": 404.0, "elapsed_s": 4492.9}
+{"update": 880, "steps_done": 1802240, "mean_step_reward": 2.278, "min_step_reward": -7.5892, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 139.852, "mean_ep_length": 61.3, "mean_collisions": 0.74, "mean_intent_rate": 0.972, "policy_loss": 0.001328, "value_loss": 188.900658, "entropy": 0.029789, "total_loss": 94.451359, "clip_frac": 0.005249, "sps": 414.1, "elapsed_s": 4497.9}
+{"update": 881, "steps_done": 1804288, "mean_step_reward": 2.3727, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 144.382, "mean_ep_length": 60.9, "mean_collisions": 0.3, "mean_intent_rate": 0.981, "policy_loss": -0.000504, "value_loss": 132.488222, "entropy": 0.023248, "total_loss": 66.243375, "clip_frac": 0.007812, "sps": 405.7, "elapsed_s": 4502.9}
+{"update": 882, "steps_done": 1806336, "mean_step_reward": 2.3354, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 150.004, "mean_ep_length": 64.2, "mean_collisions": 0.56, "mean_intent_rate": 0.981, "policy_loss": 0.001265, "value_loss": 179.228953, "entropy": 0.025316, "total_loss": 89.615489, "clip_frac": 0.007568, "sps": 407.8, "elapsed_s": 4507.9}
+{"update": 883, "steps_done": 1808384, "mean_step_reward": 2.3058, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 143.795, "mean_ep_length": 62.3, "mean_collisions": 0.67, "mean_intent_rate": 0.978, "policy_loss": 0.000395, "value_loss": 200.916045, "entropy": 0.024875, "total_loss": 100.458169, "clip_frac": 0.005737, "sps": 412.8, "elapsed_s": 4512.9}
+{"update": 884, "steps_done": 1810432, "mean_step_reward": 2.2985, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 141.546, "mean_ep_length": 61.6, "mean_collisions": 0.67, "mean_intent_rate": 0.98, "policy_loss": -0.000527, "value_loss": 188.27113, "entropy": 0.025522, "total_loss": 94.134781, "clip_frac": 0.004761, "sps": 410.9, "elapsed_s": 4517.9}
+{"update": 885, "steps_done": 1812480, "mean_step_reward": 2.3993, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 147.052, "mean_ep_length": 61.3, "mean_collisions": 0.15, "mean_intent_rate": 0.981, "policy_loss": -0.000305, "value_loss": 169.717437, "entropy": 0.023089, "total_loss": 84.858183, "clip_frac": 0.001709, "sps": 414.8, "elapsed_s": 4522.8}
+{"update": 886, "steps_done": 1814528, "mean_step_reward": 2.3594, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.341, "mean_ep_length": 63.6, "mean_collisions": 0.3, "mean_intent_rate": 0.977, "policy_loss": 0.000206, "value_loss": 147.18083, "entropy": 0.03019, "total_loss": 73.590319, "clip_frac": 0.000732, "sps": 410.3, "elapsed_s": 4527.8}
+{"update": 887, "steps_done": 1816576, "mean_step_reward": 2.3557, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 134.924, "mean_ep_length": 57.3, "mean_collisions": 0.31, "mean_intent_rate": 0.978, "policy_loss": -0.0001, "value_loss": 133.289473, "entropy": 0.02572, "total_loss": 66.644379, "clip_frac": 0.003296, "sps": 412.0, "elapsed_s": 4532.8}
+{"update": 888, "steps_done": 1818624, "mean_step_reward": 2.3967, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 141.241, "mean_ep_length": 58.9, "mean_collisions": 0.11, "mean_intent_rate": 0.979, "policy_loss": 0.000176, "value_loss": 115.508365, "entropy": 0.027088, "total_loss": 57.754087, "clip_frac": 0.001831, "sps": 402.1, "elapsed_s": 4537.9}
+{"update": 889, "steps_done": 1820672, "mean_step_reward": 2.2034, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 131.487, "mean_ep_length": 59.5, "mean_collisions": 1.11, "mean_intent_rate": 0.974, "policy_loss": -0.000428, "value_loss": 146.249186, "entropy": 0.031946, "total_loss": 73.123845, "clip_frac": 0.005371, "sps": 408.8, "elapsed_s": 4542.9}
+{"update": 890, "steps_done": 1822720, "mean_step_reward": 2.1871, "min_step_reward": -7.5874, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 137.324, "mean_ep_length": 62.9, "mean_collisions": 1.41, "mean_intent_rate": 0.976, "policy_loss": -0.002313, "value_loss": 217.784369, "entropy": 0.030422, "total_loss": 108.889566, "clip_frac": 0.009521, "sps": 413.0, "elapsed_s": 4547.9}
+{"update": 891, "steps_done": 1824768, "mean_step_reward": 2.055, "min_step_reward": -10.5361, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 132.791, "mean_ep_length": 64.9, "mean_collisions": 2.29, "mean_intent_rate": 0.975, "policy_loss": -0.0, "value_loss": 264.566578, "entropy": 0.032008, "total_loss": 132.282971, "clip_frac": 0.007446, "sps": 409.0, "elapsed_s": 4552.9}
+{"update": 892, "steps_done": 1826816, "mean_step_reward": 2.1, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 127.088, "mean_ep_length": 60.2, "mean_collisions": 1.63, "mean_intent_rate": 0.97, "policy_loss": 0.002288, "value_loss": 203.632836, "entropy": 0.043374, "total_loss": 101.818273, "clip_frac": 0.011719, "sps": 417.6, "elapsed_s": 4557.8}
+{"update": 893, "steps_done": 1828864, "mean_step_reward": 2.3576, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 131.71, "mean_ep_length": 55.9, "mean_collisions": 0.28, "mean_intent_rate": 0.974, "policy_loss": 0.000168, "value_loss": 135.140631, "entropy": 0.027327, "total_loss": 67.57021, "clip_frac": 0.007446, "sps": 410.9, "elapsed_s": 4562.8}
+{"update": 894, "steps_done": 1830912, "mean_step_reward": 2.2602, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 148.825, "mean_ep_length": 65.9, "mean_collisions": 0.97, "mean_intent_rate": 0.975, "policy_loss": -0.000341, "value_loss": 161.248045, "entropy": 0.027958, "total_loss": 80.623403, "clip_frac": 0.004395, "sps": 413.3, "elapsed_s": 4567.7}
+{"update": 895, "steps_done": 1832960, "mean_step_reward": 2.1456, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 142.796, "mean_ep_length": 66.5, "mean_collisions": 1.77, "mean_intent_rate": 0.98, "policy_loss": -2.3e-05, "value_loss": 203.109622, "entropy": 0.026799, "total_loss": 101.554519, "clip_frac": 0.007812, "sps": 418.7, "elapsed_s": 4572.6}
+{"update": 896, "steps_done": 1835008, "mean_step_reward": 2.1223, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 131.191, "mean_ep_length": 61.8, "mean_collisions": 1.64, "mean_intent_rate": 0.97, "policy_loss": -0.00031, "value_loss": 239.164698, "entropy": 0.030701, "total_loss": 119.581731, "clip_frac": 0.003052, "sps": 405.9, "elapsed_s": 4577.6}
+{"update": 897, "steps_done": 1837056, "mean_step_reward": 2.3611, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 144.707, "mean_ep_length": 61.2, "mean_collisions": 0.38, "mean_intent_rate": 0.982, "policy_loss": 0.000661, "value_loss": 139.598778, "entropy": 0.020098, "total_loss": 69.799848, "clip_frac": 0.002075, "sps": 408.0, "elapsed_s": 4582.7}
+{"update": 898, "steps_done": 1839104, "mean_step_reward": 2.3357, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.868, "mean_ep_length": 63.8, "mean_collisions": 0.47, "mean_intent_rate": 0.975, "policy_loss": -0.000563, "value_loss": 167.020252, "entropy": 0.023985, "total_loss": 83.509323, "clip_frac": 0.002686, "sps": 409.8, "elapsed_s": 4587.7}
+{"update": 899, "steps_done": 1841152, "mean_step_reward": 2.3571, "min_step_reward": -7.5859, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 141.815, "mean_ep_length": 60.2, "mean_collisions": 0.32, "mean_intent_rate": 0.977, "policy_loss": 0.000159, "value_loss": 139.566811, "entropy": 0.024434, "total_loss": 69.78332, "clip_frac": 0.005249, "sps": 413.9, "elapsed_s": 4592.6}
+{"update": 900, "steps_done": 1843200, "mean_step_reward": 2.3116, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 137.398, "mean_ep_length": 59.4, "mean_collisions": 0.44, "mean_intent_rate": 0.969, "policy_loss": 0.001042, "value_loss": 147.828201, "entropy": 0.03168, "total_loss": 73.914826, "clip_frac": 0.009155, "sps": 411.7, "elapsed_s": 4597.6}
+{"update": 901, "steps_done": 1845248, "mean_step_reward": 2.3225, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 141.01, "mean_ep_length": 60.7, "mean_collisions": 0.59, "mean_intent_rate": 0.981, "policy_loss": 0.000327, "value_loss": 170.832387, "entropy": 0.022915, "total_loss": 85.416291, "clip_frac": 0.002441, "sps": 412.8, "elapsed_s": 4602.6}
+{"update": 902, "steps_done": 1847296, "mean_step_reward": 2.4037, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 155.166, "mean_ep_length": 64.5, "mean_collisions": 0.16, "mean_intent_rate": 0.983, "policy_loss": -5.6e-05, "value_loss": 170.466201, "entropy": 0.020416, "total_loss": 85.232841, "clip_frac": 0.001465, "sps": 409.4, "elapsed_s": 4607.6}
+{"update": 903, "steps_done": 1849344, "mean_step_reward": 2.3371, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 140.986, "mean_ep_length": 60.3, "mean_collisions": 0.44, "mean_intent_rate": 0.98, "policy_loss": -0.000634, "value_loss": 134.402915, "entropy": 0.024511, "total_loss": 67.200579, "clip_frac": 0.004028, "sps": 416.5, "elapsed_s": 4612.5}
+{"update": 904, "steps_done": 1851392, "mean_step_reward": 2.4106, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 144.915, "mean_ep_length": 60.1, "mean_collisions": 0.06, "mean_intent_rate": 0.98, "policy_loss": -0.000294, "value_loss": 143.00751, "entropy": 0.022746, "total_loss": 71.503233, "clip_frac": 0.002197, "sps": 416.6, "elapsed_s": 4617.4}
+{"update": 905, "steps_done": 1853440, "mean_step_reward": 2.2553, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 139.511, "mean_ep_length": 61.9, "mean_collisions": 0.94, "mean_intent_rate": 0.976, "policy_loss": -0.000148, "value_loss": 196.373041, "entropy": 0.029269, "total_loss": 98.18608, "clip_frac": 0.00708, "sps": 406.0, "elapsed_s": 4622.4}
+{"update": 906, "steps_done": 1855488, "mean_step_reward": 2.3202, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 141.443, "mean_ep_length": 60.9, "mean_collisions": 0.56, "mean_intent_rate": 0.977, "policy_loss": 0.000107, "value_loss": 184.288925, "entropy": 0.027949, "total_loss": 92.14429, "clip_frac": 0.003296, "sps": 414.9, "elapsed_s": 4627.4}
+{"update": 907, "steps_done": 1857536, "mean_step_reward": 2.3955, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 161.159, "mean_ep_length": 67.3, "mean_collisions": 0.17, "mean_intent_rate": 0.981, "policy_loss": 0.000204, "value_loss": 147.917228, "entropy": 0.022985, "total_loss": 73.958587, "clip_frac": 0.00293, "sps": 415.6, "elapsed_s": 4632.3}
+{"update": 908, "steps_done": 1859584, "mean_step_reward": 2.3346, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 153.134, "mean_ep_length": 65.6, "mean_collisions": 0.52, "mean_intent_rate": 0.977, "policy_loss": -6.7e-05, "value_loss": 163.623065, "entropy": 0.020871, "total_loss": 81.811257, "clip_frac": 0.003418, "sps": 414.4, "elapsed_s": 4637.3}
+{"update": 909, "steps_done": 1861632, "mean_step_reward": 2.262, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 143.951, "mean_ep_length": 63.7, "mean_collisions": 0.94, "mean_intent_rate": 0.979, "policy_loss": -0.000444, "value_loss": 167.387566, "entropy": 0.027654, "total_loss": 83.693063, "clip_frac": 0.001465, "sps": 416.6, "elapsed_s": 4642.2}
+{"update": 910, "steps_done": 1863680, "mean_step_reward": 2.3318, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 151.313, "mean_ep_length": 64.8, "mean_collisions": 0.53, "mean_intent_rate": 0.979, "policy_loss": -0.000755, "value_loss": 140.281032, "entropy": 0.024096, "total_loss": 70.139519, "clip_frac": 0.005005, "sps": 415.6, "elapsed_s": 4647.1}
+{"update": 911, "steps_done": 1865728, "mean_step_reward": 2.3384, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.082, "mean_ep_length": 63.3, "mean_collisions": 0.53, "mean_intent_rate": 0.981, "policy_loss": 0.000592, "value_loss": 151.114608, "entropy": 0.020602, "total_loss": 75.557691, "clip_frac": 0.002563, "sps": 410.0, "elapsed_s": 4652.1}
+{"update": 912, "steps_done": 1867776, "mean_step_reward": 2.2784, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 134.972, "mean_ep_length": 59.2, "mean_collisions": 0.74, "mean_intent_rate": 0.977, "policy_loss": 0.0007, "value_loss": 191.69876, "entropy": 0.026076, "total_loss": 95.849818, "clip_frac": 0.005127, "sps": 417.7, "elapsed_s": 4657.0}
+{"update": 913, "steps_done": 1869824, "mean_step_reward": 2.3948, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 153.662, "mean_ep_length": 64.2, "mean_collisions": 0.19, "mean_intent_rate": 0.982, "policy_loss": -8.9e-05, "value_loss": 141.55303, "entropy": 0.018423, "total_loss": 70.776241, "clip_frac": 0.002319, "sps": 401.1, "elapsed_s": 4662.1}
+{"update": 914, "steps_done": 1871872, "mean_step_reward": 2.3182, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 139.124, "mean_ep_length": 60.0, "mean_collisions": 0.56, "mean_intent_rate": 0.979, "policy_loss": 0.000256, "value_loss": 148.251334, "entropy": 0.022436, "total_loss": 74.125699, "clip_frac": 0.002197, "sps": 411.1, "elapsed_s": 4667.1}
+{"update": 915, "steps_done": 1873920, "mean_step_reward": 2.3635, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 147.89, "mean_ep_length": 62.5, "mean_collisions": 0.36, "mean_intent_rate": 0.982, "policy_loss": -0.000188, "value_loss": 144.817403, "entropy": 0.018538, "total_loss": 72.408328, "clip_frac": 0.000488, "sps": 410.2, "elapsed_s": 4672.1}
+{"update": 916, "steps_done": 1875968, "mean_step_reward": 2.372, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 149.646, "mean_ep_length": 63.1, "mean_collisions": 0.31, "mean_intent_rate": 0.981, "policy_loss": -0.000136, "value_loss": 149.281569, "entropy": 0.022606, "total_loss": 74.640421, "clip_frac": 0.000732, "sps": 411.9, "elapsed_s": 4677.1}
+{"update": 917, "steps_done": 1878016, "mean_step_reward": 2.4214, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 160.041, "mean_ep_length": 66.1, "mean_collisions": 0.06, "mean_intent_rate": 0.984, "policy_loss": -5.1e-05, "value_loss": 165.784791, "entropy": 0.016932, "total_loss": 82.892175, "clip_frac": 0.000244, "sps": 408.5, "elapsed_s": 4682.1}
+{"update": 918, "steps_done": 1880064, "mean_step_reward": 2.3943, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 163.7, "mean_ep_length": 68.4, "mean_collisions": 0.23, "mean_intent_rate": 0.983, "policy_loss": 0.000794, "value_loss": 162.074156, "entropy": 0.018604, "total_loss": 81.037686, "clip_frac": 0.001831, "sps": 410.4, "elapsed_s": 4687.1}
+{"update": 919, "steps_done": 1882112, "mean_step_reward": 2.3846, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 154.603, "mean_ep_length": 64.4, "mean_collisions": 0.16, "mean_intent_rate": 0.983, "policy_loss": -0.000535, "value_loss": 181.514165, "entropy": 0.020429, "total_loss": 90.756343, "clip_frac": 0.006836, "sps": 412.0, "elapsed_s": 4692.0}
+{"update": 920, "steps_done": 1884160, "mean_step_reward": 2.3542, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 157.098, "mean_ep_length": 67.3, "mean_collisions": 0.6, "mean_intent_rate": 0.982, "policy_loss": -0.000136, "value_loss": 168.137992, "entropy": 0.019548, "total_loss": 84.068665, "clip_frac": 0.002808, "sps": 409.9, "elapsed_s": 4697.0}
+{"update": 921, "steps_done": 1886208, "mean_step_reward": 2.3128, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 145.802, "mean_ep_length": 63.0, "mean_collisions": 0.67, "mean_intent_rate": 0.98, "policy_loss": 0.000312, "value_loss": 167.149981, "entropy": 0.019738, "total_loss": 83.575104, "clip_frac": 0.005371, "sps": 412.2, "elapsed_s": 4702.0}
+{"update": 922, "steps_done": 1888256, "mean_step_reward": 2.2084, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 134.87, "mean_ep_length": 61.2, "mean_collisions": 1.24, "mean_intent_rate": 0.977, "policy_loss": -0.001863, "value_loss": 235.952585, "entropy": 0.029746, "total_loss": 117.974132, "clip_frac": 0.008545, "sps": 405.1, "elapsed_s": 4707.1}
+{"update": 923, "steps_done": 1890304, "mean_step_reward": 2.3021, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 149.997, "mean_ep_length": 65.3, "mean_collisions": 0.74, "mean_intent_rate": 0.977, "policy_loss": 0.000178, "value_loss": 158.641485, "entropy": 0.02325, "total_loss": 79.320688, "clip_frac": 0.003174, "sps": 411.6, "elapsed_s": 4712.0}
+{"update": 924, "steps_done": 1892352, "mean_step_reward": 2.4122, "min_step_reward": -7.5391, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 155.126, "mean_ep_length": 64.2, "mean_collisions": 0.06, "mean_intent_rate": 0.982, "policy_loss": -0.000279, "value_loss": 128.320208, "entropy": 0.019741, "total_loss": 64.159627, "clip_frac": 0.001831, "sps": 412.8, "elapsed_s": 4717.0}
+{"update": 925, "steps_done": 1894400, "mean_step_reward": 2.4289, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 146.235, "mean_ep_length": 60.2, "mean_collisions": 0.0, "mean_intent_rate": 0.982, "policy_loss": 2.1e-05, "value_loss": 124.351894, "entropy": 0.020305, "total_loss": 62.175765, "clip_frac": 0.000977, "sps": 407.5, "elapsed_s": 4722.0}
+{"update": 926, "steps_done": 1896448, "mean_step_reward": 2.3801, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 151.544, "mean_ep_length": 63.7, "mean_collisions": 0.16, "mean_intent_rate": 0.977, "policy_loss": -0.000272, "value_loss": 135.439113, "entropy": 0.024041, "total_loss": 67.719044, "clip_frac": 0.003418, "sps": 413.9, "elapsed_s": 4727.0}
+{"update": 927, "steps_done": 1898496, "mean_step_reward": 2.3703, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 135.489, "mean_ep_length": 57.1, "mean_collisions": 0.19, "mean_intent_rate": 0.974, "policy_loss": -0.000377, "value_loss": 130.079879, "entropy": 0.026863, "total_loss": 65.039294, "clip_frac": 0.003906, "sps": 413.7, "elapsed_s": 4731.9}
+{"update": 928, "steps_done": 1900544, "mean_step_reward": 2.3631, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 140.052, "mean_ep_length": 59.4, "mean_collisions": 0.32, "mean_intent_rate": 0.977, "policy_loss": -0.00011, "value_loss": 99.316707, "entropy": 0.020018, "total_loss": 49.658044, "clip_frac": 0.001587, "sps": 411.3, "elapsed_s": 4736.9}
+{"update": 929, "steps_done": 1902592, "mean_step_reward": 2.2853, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 158.256, "mean_ep_length": 69.2, "mean_collisions": 0.9, "mean_intent_rate": 0.981, "policy_loss": 0.000328, "value_loss": 147.776448, "entropy": 0.020124, "total_loss": 73.88835, "clip_frac": 0.003418, "sps": 404.8, "elapsed_s": 4742.0}
+{"update": 930, "steps_done": 1904640, "mean_step_reward": 2.2309, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 130.962, "mean_ep_length": 58.7, "mean_collisions": 1.03, "mean_intent_rate": 0.977, "policy_loss": 0.000624, "value_loss": 201.412822, "entropy": 0.020483, "total_loss": 100.70683, "clip_frac": 0.002441, "sps": 399.3, "elapsed_s": 4747.1}
+{"update": 931, "steps_done": 1906688, "mean_step_reward": 2.3569, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 162.447, "mean_ep_length": 69.0, "mean_collisions": 0.48, "mean_intent_rate": 0.983, "policy_loss": -9.5e-05, "value_loss": 125.266842, "entropy": 0.018402, "total_loss": 62.633142, "clip_frac": 0.002563, "sps": 413.5, "elapsed_s": 4752.0}
+{"update": 932, "steps_done": 1908736, "mean_step_reward": 2.3543, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 147.461, "mean_ep_length": 62.6, "mean_collisions": 0.39, "mean_intent_rate": 0.979, "policy_loss": -7.4e-05, "value_loss": 129.693226, "entropy": 0.01828, "total_loss": 64.846356, "clip_frac": 0.005249, "sps": 412.6, "elapsed_s": 4757.0}
+{"update": 933, "steps_done": 1910784, "mean_step_reward": 2.3212, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 137.145, "mean_ep_length": 58.6, "mean_collisions": 0.47, "mean_intent_rate": 0.98, "policy_loss": 0.000354, "value_loss": 142.825495, "entropy": 0.019928, "total_loss": 71.412902, "clip_frac": 0.005981, "sps": 413.1, "elapsed_s": 4762.0}
+{"update": 934, "steps_done": 1912832, "mean_step_reward": 2.3394, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 137.409, "mean_ep_length": 59.2, "mean_collisions": 0.49, "mean_intent_rate": 0.977, "policy_loss": 0.000348, "value_loss": 171.45539, "entropy": 0.022716, "total_loss": 85.727816, "clip_frac": 0.001709, "sps": 414.4, "elapsed_s": 4766.9}
+{"update": 935, "steps_done": 1914880, "mean_step_reward": 2.3964, "min_step_reward": -10.5505, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 144.481, "mean_ep_length": 60.2, "mean_collisions": 0.11, "mean_intent_rate": 0.979, "policy_loss": -0.00046, "value_loss": 123.721741, "entropy": 0.02038, "total_loss": 61.860207, "clip_frac": 0.002563, "sps": 408.2, "elapsed_s": 4771.9}
+{"update": 936, "steps_done": 1916928, "mean_step_reward": 2.1995, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 144.42, "mean_ep_length": 65.7, "mean_collisions": 1.48, "mean_intent_rate": 0.982, "policy_loss": -0.001175, "value_loss": 199.46053, "entropy": 0.021046, "total_loss": 99.72888, "clip_frac": 0.003906, "sps": 412.6, "elapsed_s": 4776.9}
+{"update": 937, "steps_done": 1918976, "mean_step_reward": 2.4253, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 168.889, "mean_ep_length": 69.7, "mean_collisions": 0.07, "mean_intent_rate": 0.984, "policy_loss": -0.00029, "value_loss": 153.353868, "entropy": 0.016257, "total_loss": 76.676481, "clip_frac": 0.004272, "sps": 412.7, "elapsed_s": 4781.9}
+{"update": 938, "steps_done": 1921024, "mean_step_reward": 2.3911, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 149.362, "mean_ep_length": 62.5, "mean_collisions": 0.21, "mean_intent_rate": 0.982, "policy_loss": -0.000259, "value_loss": 157.271849, "entropy": 0.019457, "total_loss": 78.635471, "clip_frac": 0.001831, "sps": 409.1, "elapsed_s": 4786.9}
+{"update": 939, "steps_done": 1923072, "mean_step_reward": 2.3887, "min_step_reward": -10.5612, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 146.426, "mean_ep_length": 61.3, "mean_collisions": 0.21, "mean_intent_rate": 0.983, "policy_loss": 6.3e-05, "value_loss": 128.466938, "entropy": 0.016946, "total_loss": 64.233362, "clip_frac": 0.002686, "sps": 405.0, "elapsed_s": 4791.9}
+{"update": 940, "steps_done": 1925120, "mean_step_reward": 2.2071, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 134.008, "mean_ep_length": 60.6, "mean_collisions": 1.18, "mean_intent_rate": 0.978, "policy_loss": -0.001155, "value_loss": 168.480263, "entropy": 0.024487, "total_loss": 84.238732, "clip_frac": 0.004761, "sps": 413.8, "elapsed_s": 4796.9}
+{"update": 941, "steps_done": 1927168, "mean_step_reward": 2.4024, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 144.113, "mean_ep_length": 60.0, "mean_collisions": 0.15, "mean_intent_rate": 0.982, "policy_loss": 0.000231, "value_loss": 140.160878, "entropy": 0.019342, "total_loss": 70.080476, "clip_frac": 0.003662, "sps": 416.8, "elapsed_s": 4801.8}
+{"update": 942, "steps_done": 1929216, "mean_step_reward": 2.422, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 143.346, "mean_ep_length": 59.2, "mean_collisions": 0.0, "mean_intent_rate": 0.982, "policy_loss": -0.000625, "value_loss": 126.871624, "entropy": 0.020097, "total_loss": 63.434986, "clip_frac": 0.002808, "sps": 415.1, "elapsed_s": 4806.7}
+{"update": 943, "steps_done": 1931264, "mean_step_reward": 2.2358, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 125.672, "mean_ep_length": 56.3, "mean_collisions": 0.97, "mean_intent_rate": 0.975, "policy_loss": 5.1e-05, "value_loss": 149.426165, "entropy": 0.023199, "total_loss": 74.712902, "clip_frac": 0.000732, "sps": 411.4, "elapsed_s": 4811.7}
+{"update": 944, "steps_done": 1933312, "mean_step_reward": 2.3851, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 160.305, "mean_ep_length": 67.2, "mean_collisions": 0.23, "mean_intent_rate": 0.981, "policy_loss": 0.000457, "value_loss": 103.570827, "entropy": 0.020536, "total_loss": 51.785666, "clip_frac": 0.002686, "sps": 409.5, "elapsed_s": 4816.7}
+{"update": 945, "steps_done": 1935360, "mean_step_reward": 2.3155, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 143.562, "mean_ep_length": 62.0, "mean_collisions": 0.58, "mean_intent_rate": 0.978, "policy_loss": 0.002881, "value_loss": 152.805198, "entropy": 0.023359, "total_loss": 76.405246, "clip_frac": 0.006348, "sps": 416.3, "elapsed_s": 4821.6}
+{"update": 946, "steps_done": 1937408, "mean_step_reward": 2.3399, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 152.822, "mean_ep_length": 65.4, "mean_collisions": 0.58, "mean_intent_rate": 0.981, "policy_loss": -0.000434, "value_loss": 129.341989, "entropy": 0.020557, "total_loss": 64.670356, "clip_frac": 0.002563, "sps": 413.4, "elapsed_s": 4826.6}
+{"update": 947, "steps_done": 1939456, "mean_step_reward": 2.3301, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 146.093, "mean_ep_length": 62.7, "mean_collisions": 0.55, "mean_intent_rate": 0.979, "policy_loss": 0.000191, "value_loss": 124.498177, "entropy": 0.022327, "total_loss": 62.249058, "clip_frac": 0.00293, "sps": 403.0, "elapsed_s": 4831.7}
+{"update": 948, "steps_done": 1941504, "mean_step_reward": 2.1818, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 148.374, "mean_ep_length": 68.0, "mean_collisions": 1.53, "mean_intent_rate": 0.978, "policy_loss": 0.001126, "value_loss": 173.738247, "entropy": 0.025755, "total_loss": 86.869992, "clip_frac": 0.002808, "sps": 413.8, "elapsed_s": 4836.6}
+{"update": 949, "steps_done": 1943552, "mean_step_reward": 2.1937, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 130.157, "mean_ep_length": 59.4, "mean_collisions": 1.32, "mean_intent_rate": 0.977, "policy_loss": -0.000729, "value_loss": 195.424159, "entropy": 0.023541, "total_loss": 97.711115, "clip_frac": 0.004028, "sps": 410.7, "elapsed_s": 4841.6}
+{"update": 950, "steps_done": 1945600, "mean_step_reward": 2.3397, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 145.445, "mean_ep_length": 62.2, "mean_collisions": 0.52, "mean_intent_rate": 0.98, "policy_loss": -0.001021, "value_loss": 179.728891, "entropy": 0.021912, "total_loss": 89.863205, "clip_frac": 0.005249, "sps": 412.7, "elapsed_s": 4846.6}
+{"update": 951, "steps_done": 1947648, "mean_step_reward": 2.3405, "min_step_reward": -7.5867, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 158.468, "mean_ep_length": 67.7, "mean_collisions": 0.6, "mean_intent_rate": 0.982, "policy_loss": -0.000331, "value_loss": 152.327202, "entropy": 0.019909, "total_loss": 76.163072, "clip_frac": 0.00293, "sps": 417.0, "elapsed_s": 4851.5}
+{"update": 952, "steps_done": 1949696, "mean_step_reward": 2.4162, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 152.337, "mean_ep_length": 63.0, "mean_collisions": 0.06, "mean_intent_rate": 0.982, "policy_loss": -0.000111, "value_loss": 121.466432, "entropy": 0.019695, "total_loss": 60.732908, "clip_frac": 0.003662, "sps": 412.5, "elapsed_s": 4856.5}
+{"update": 953, "steps_done": 1951744, "mean_step_reward": 2.4318, "min_step_reward": -0.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 150.836, "mean_ep_length": 62.0, "mean_collisions": 0.0, "mean_intent_rate": 0.982, "policy_loss": 6.1e-05, "value_loss": 122.97689, "entropy": 0.01954, "total_loss": 61.48831, "clip_frac": 0.00061, "sps": 407.1, "elapsed_s": 4861.5}
+{"update": 954, "steps_done": 1953792, "mean_step_reward": 2.3971, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 144.411, "mean_ep_length": 60.2, "mean_collisions": 0.18, "mean_intent_rate": 0.982, "policy_loss": 0.000134, "value_loss": 123.289167, "entropy": 0.018905, "total_loss": 61.644529, "clip_frac": 0.000977, "sps": 412.1, "elapsed_s": 4866.5}
+{"update": 955, "steps_done": 1955840, "mean_step_reward": 2.4108, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 156.219, "mean_ep_length": 64.8, "mean_collisions": 0.09, "mean_intent_rate": 0.983, "policy_loss": 0.002072, "value_loss": 91.966073, "entropy": 0.018719, "total_loss": 45.984921, "clip_frac": 0.001953, "sps": 412.9, "elapsed_s": 4871.4}
+{"update": 956, "steps_done": 1957888, "mean_step_reward": 2.3597, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 148.918, "mean_ep_length": 63.2, "mean_collisions": 0.41, "mean_intent_rate": 0.981, "policy_loss": -0.000121, "value_loss": 155.6457, "entropy": 0.018915, "total_loss": 77.82254, "clip_frac": 0.001587, "sps": 404.7, "elapsed_s": 4876.5}
+{"update": 957, "steps_done": 1959936, "mean_step_reward": 2.3606, "min_step_reward": -7.5878, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 152.325, "mean_ep_length": 64.5, "mean_collisions": 0.38, "mean_intent_rate": 0.98, "policy_loss": -0.000536, "value_loss": 166.501496, "entropy": 0.023201, "total_loss": 83.24998, "clip_frac": 0.002319, "sps": 411.4, "elapsed_s": 4881.5}
+{"update": 958, "steps_done": 1961984, "mean_step_reward": 2.0455, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 123.069, "mean_ep_length": 60.2, "mean_collisions": 2.24, "mean_intent_rate": 0.979, "policy_loss": -0.001124, "value_loss": 240.033017, "entropy": 0.027172, "total_loss": 120.015112, "clip_frac": 0.004272, "sps": 411.3, "elapsed_s": 4886.4}
+{"update": 959, "steps_done": 1964032, "mean_step_reward": 2.3555, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 153.746, "mean_ep_length": 65.3, "mean_collisions": 0.45, "mean_intent_rate": 0.982, "policy_loss": -0.000584, "value_loss": 156.628756, "entropy": 0.019273, "total_loss": 78.313602, "clip_frac": 0.00293, "sps": 418.1, "elapsed_s": 4891.3}
+{"update": 960, "steps_done": 1966080, "mean_step_reward": 2.2317, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 142.59, "mean_ep_length": 63.8, "mean_collisions": 1.12, "mean_intent_rate": 0.977, "policy_loss": -0.00052, "value_loss": 190.377276, "entropy": 0.025353, "total_loss": 95.187866, "clip_frac": 0.003052, "sps": 412.2, "elapsed_s": 4896.3}
+{"update": 961, "steps_done": 1968128, "mean_step_reward": 2.1968, "min_step_reward": -10.5867, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 143.265, "mean_ep_length": 65.2, "mean_collisions": 1.47, "mean_intent_rate": 0.982, "policy_loss": 5.5e-05, "value_loss": 216.930474, "entropy": 0.017672, "total_loss": 108.465115, "clip_frac": 0.001831, "sps": 409.5, "elapsed_s": 4901.3}
+{"update": 962, "steps_done": 1970176, "mean_step_reward": 2.319, "min_step_reward": -7.5883, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 140.273, "mean_ep_length": 60.5, "mean_collisions": 0.56, "mean_intent_rate": 0.979, "policy_loss": 0.000237, "value_loss": 185.535501, "entropy": 0.024032, "total_loss": 92.767747, "clip_frac": 0.005127, "sps": 413.5, "elapsed_s": 4906.3}
+{"update": 963, "steps_done": 1972224, "mean_step_reward": 2.177, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 36, "mean_ep_return": 123.638, "mean_ep_length": 56.8, "mean_collisions": 1.31, "mean_intent_rate": 0.977, "policy_loss": -0.000695, "value_loss": 157.527883, "entropy": 0.023619, "total_loss": 78.763011, "clip_frac": 0.005615, "sps": 416.2, "elapsed_s": 4911.2}
+{"update": 964, "steps_done": 1974272, "mean_step_reward": 2.4165, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 149.364, "mean_ep_length": 61.8, "mean_collisions": 0.06, "mean_intent_rate": 0.983, "policy_loss": 6.7e-05, "value_loss": 113.897891, "entropy": 0.016227, "total_loss": 56.948851, "clip_frac": 0.000977, "sps": 398.2, "elapsed_s": 4916.3}
+{"update": 965, "steps_done": 1976320, "mean_step_reward": 2.3729, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 30, "mean_ep_return": 159.998, "mean_ep_length": 67.5, "mean_collisions": 0.37, "mean_intent_rate": 0.984, "policy_loss": -5.4e-05, "value_loss": 127.242181, "entropy": 0.016634, "total_loss": 63.620871, "clip_frac": 0.003784, "sps": 413.7, "elapsed_s": 4921.3}
+{"update": 966, "steps_done": 1978368, "mean_step_reward": 2.2871, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 136.309, "mean_ep_length": 59.6, "mean_collisions": 0.74, "mean_intent_rate": 0.975, "policy_loss": -0.000409, "value_loss": 152.493193, "entropy": 0.027252, "total_loss": 76.245915, "clip_frac": 0.006348, "sps": 408.8, "elapsed_s": 4926.3}
+{"update": 967, "steps_done": 1980416, "mean_step_reward": 2.4051, "min_step_reward": -7.5614, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 149.184, "mean_ep_length": 62.0, "mean_collisions": 0.12, "mean_intent_rate": 0.982, "policy_loss": 0.00034, "value_loss": 158.952914, "entropy": 0.020053, "total_loss": 79.476597, "clip_frac": 0.003662, "sps": 414.3, "elapsed_s": 4931.2}
+{"update": 968, "steps_done": 1982464, "mean_step_reward": 2.2969, "min_step_reward": -10.5561, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 143.257, "mean_ep_length": 62.5, "mean_collisions": 0.78, "mean_intent_rate": 0.979, "policy_loss": -0.000621, "value_loss": 202.954854, "entropy": 0.023228, "total_loss": 101.476574, "clip_frac": 0.004028, "sps": 409.1, "elapsed_s": 4936.2}
+{"update": 969, "steps_done": 1984512, "mean_step_reward": 2.2665, "min_step_reward": -10.5081, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 147.281, "mean_ep_length": 65.1, "mean_collisions": 0.97, "mean_intent_rate": 0.978, "policy_loss": -0.000667, "value_loss": 189.828131, "entropy": 0.025444, "total_loss": 94.913145, "clip_frac": 0.006592, "sps": 412.0, "elapsed_s": 4941.2}
+{"update": 970, "steps_done": 1986560, "mean_step_reward": 2.4063, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 150.411, "mean_ep_length": 62.4, "mean_collisions": 0.09, "mean_intent_rate": 0.981, "policy_loss": 0.001377, "value_loss": 130.072497, "entropy": 0.019362, "total_loss": 65.037433, "clip_frac": 0.002075, "sps": 413.8, "elapsed_s": 4946.2}
+{"update": 971, "steps_done": 1988608, "mean_step_reward": 2.3247, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 161.279, "mean_ep_length": 69.4, "mean_collisions": 0.66, "mean_intent_rate": 0.98, "policy_loss": -0.000466, "value_loss": 164.362355, "entropy": 0.021416, "total_loss": 82.180497, "clip_frac": 0.001465, "sps": 414.2, "elapsed_s": 4951.1}
+{"update": 972, "steps_done": 1990656, "mean_step_reward": 2.4191, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 32, "mean_ep_return": 154.89, "mean_ep_length": 64.0, "mean_collisions": 0.06, "mean_intent_rate": 0.983, "policy_loss": -0.000131, "value_loss": 119.871729, "entropy": 0.017739, "total_loss": 59.935555, "clip_frac": 0.000122, "sps": 409.2, "elapsed_s": 4956.1}
+{"update": 973, "steps_done": 1992704, "mean_step_reward": 2.3874, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 149.976, "mean_ep_length": 62.8, "mean_collisions": 0.21, "mean_intent_rate": 0.981, "policy_loss": -0.000585, "value_loss": 131.532751, "entropy": 0.018123, "total_loss": 65.76561, "clip_frac": 0.003784, "sps": 407.2, "elapsed_s": 4961.1}
+{"update": 974, "steps_done": 1994752, "mean_step_reward": 2.3187, "min_step_reward": -10.5785, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 148.815, "mean_ep_length": 64.3, "mean_collisions": 0.65, "mean_intent_rate": 0.982, "policy_loss": 0.000557, "value_loss": 142.278336, "entropy": 0.019026, "total_loss": 71.139535, "clip_frac": 0.005005, "sps": 414.5, "elapsed_s": 4966.1}
+{"update": 975, "steps_done": 1996800, "mean_step_reward": 2.2447, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 34, "mean_ep_return": 139.259, "mean_ep_length": 61.9, "mean_collisions": 0.97, "mean_intent_rate": 0.98, "policy_loss": 0.001259, "value_loss": 146.38829, "entropy": 0.027892, "total_loss": 73.195126, "clip_frac": 0.006714, "sps": 411.3, "elapsed_s": 4971.1}
+{"update": 976, "steps_done": 1998848, "mean_step_reward": 2.4139, "min_step_reward": -10.5789, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 155.122, "mean_ep_length": 64.3, "mean_collisions": 0.06, "mean_intent_rate": 0.983, "policy_loss": 0.000897, "value_loss": 109.792812, "entropy": 0.017049, "total_loss": 54.897132, "clip_frac": 0.002197, "sps": 414.1, "elapsed_s": 4976.0}